Flatiron Health, Inc. v. Tempus, Inc.

• Court: District Court, S.D. New York
• Decided: March 20, 2020
• Docket: 1:19-cv-08999
• Status: Unknown

Opinion

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK ---------------------------------- X FLATIRON HEALTH, INC., : 3/20/20 : Plaintiff, : 19 Civ. 8999 (VM) : - against - : DECISION AND ORDER : KENNETH CARSON, M.D., : : Defendant. : ---------------------------------- X VICTOR MARRERO, United States District Judge.

Plaintiff Flatiron Health, Inc. (“Flatiron”) brought this action against its former employee, defendant Kenneth Carson, M.D. (“Carson”), to enforce a non-compete agreement.1 Flatiron claims that Carson anticipatorily repudiated the terms of the parties’ Covenants Agreement (the “Covenants Agreement”) by informing Flatiron that he accepted a job at Tempus Labs, Inc. (“Tempus”) and planned to immediately begin work there. Flatiron seeks a declaratory judgment, pursuant to 28 U.S.C. Section 2201, determining that the Covenants Agreement bars Carson from working for Tempus and soliciting Flatiron customers and employees for one year following his separation from

1 Flatiron has since voluntarily dismissed its claims against Tempus Labs, Inc. (Dkt. No. 32.) 1 employment by Flatiron, and that it also bars Carson from retaining, using, or disclosing any of Flatiron’s trade secrets and confidential information. In addition, Flatiron asks the Court for injunctive relief restraining Carson, until September 26, 2020, from violating the terms of the

Covenants Agreement, including by working for Tempus, soliciting Flatiron’s customers, or revealing Flatiron’s trade secrets or confidential information. The Court held a bench trial on January 27, 28, and 30, 2020 to adjudicate Flatiron’s claims. On February 19, 2020, the Court issued a judgment denying Flatiron’s requests for a declaratory judgment and an injunction and indicated that a formal Decision and Order stating the Court’s findings, reasoning, and conclusions would follow. (See Dkt. No. 94.) Shortly thereafter, Flatiron filed a motion, under Federal Rule of Civil Procedure 62(d) (“Rule 62(d)”),

asking the Court to enjoin Carson from working for Tempus or, alternatively, from violating certain limitations on the scope of his employment at Tempus, pending appeal of this Court’s decision to the Second Circuit. Carson has submitted a response to that request, and Flatiron has

2 replied. The Court resolves that motion in a separate Decision and Order. The Court now issues this decision setting forth the Court’s findings of fact and conclusions of law in support of its February 19, 2020 judgment as required by Federal

Rule of Civil Procedure 52(a). I. FINDINGS OF FACT2 A. CARSON’S BACKGROUND AND EXPERIENCE Carson is a physician specializing in hematology and oncology. He obtained his medical degree from the University of Southern California in 2000. After completing his residency at Duke University Medical Center in 2003, he

2 The factual recitation that follows derives from the Trial Transcript (“Tr.”) and exhibits, the January 23, 2020 Declaration of Kenneth Carson, M.D., Ph.D. (“Carson Decl.”), the January 23, 2020 Direct Testimony of Melisa Tucker (“Tucker Decl.”), the January 23, 2020 Direct Testimony of Zach Weinberg (“Weinberg Decl.”), the January 23, 2020 Direct Testimony of Geoffrey Calkins (“Calkins Decl.”), the January 23, 2020 Declaration of Ryan Fukushima (“Fukushima Decl.”), the January 13, 2020 Rule 30(b)(6) Deposition of Noel Auguston (“Auguston Dep.”), the January 20, 2020 Deposition of Erik Lefkofsky (“Lefkofsky Dep.”), the January 10, 2020 Deposition of Gary Palmer (“Palmer Dep.”), the January 19, 2020 Deposition of Neal Meropol (“Meropol Dep.”), the January 21, 2020 Deposition of Nat Turner (“Turner Dep.”), and the January 17, 2020 Deposition of Michael Axelson (“Axelson Dep.”), as well as the pleadings and motion papers on the public record. No further citations to these sources will be made herein except as specifically cited. While the Court has reviewed and considered all of the live and deposition testimony and accompanying exhibits admitted in evidence in connection with the trial in this matter, the Court addresses only those portions of the evidence relevant to its legal conclusions. 3 became an Associate in Medicine there. From 2004 to 2006, Carson worked as a Postdoctoral Fellow at the Research Institute for Healthcare Studies at Northwestern University. He completed fellowships in oncology and hematology at Northwestern University between 2005 and

2008. In 2008, Carson joined the faculty of the Washington University School of Medicine as a Clinical Instructor. Carson received a promotion to Assistant Professor of Medicine in the Division of Oncology in 2010 and served in that role until August 2019. Carson obtained a Ph.D. in Health Policy and Administration from the University of Illinois at Chicago in 2014. His dissertation evaluated racial disparities in the treatment of a type of lymphoma in veteran patients using the United States Department of Veterans Affairs (“VA”) healthcare system’s national database. The project

required Carson to oversee data collection -- by medical students and others -- from electronic patient medical records and to analyze that data to draw inferences about treatment disparities. Carson served as Chief of Oncology and Hematology at the VA Medical Center in St. Louis, Missouri from 2013 to 4 2016. He continued treating patients there one day a week through December 2019. Carson has published more than 70 research articles in peer-reviewed publications, and he published only three of those articles in connection with his role at Flatiron.

Carson applied the research and data analysis skills he developed in graduate school to produce many of these articles. For example, Carson has authored articles based on his participation in the Research on Adverse Drug Events and Reports (“RADAR”) project. The RADAR project relies on aggregated data from physician queries, clinical trials, published case reports, Federal Drug Administration databases, and manufacturers to investigate severe and previously unrecognized adverse drug and device reactions. Carson has also published articles that employ regression modeling to explore the association between obesity and survival of United States veterans with lymphoma and

myeloma. In addition, Carson has published articles regarding race-based disparities in the incidence and prognosis of multiple myeloma.

5 B. FLATIRON’S BUSINESS Flatiron is a healthcare technology and data services company. Flatiron has several business lines, which serve separate sets of customers. 1. Flatiron’s Clinical Data Aggregation Business

Flatiron’s largest line of business is its real-world evidence (“RWE”) service, which converts raw clinical data from patient records into a structured format so that the data can be used for research purposes.3 After structuring the data, Flatiron aggregates the data into data sets. Flatiron generates revenue by selling data sets to biopharmaceutical companies, as well as some regulatory agencies and researchers. Flatiron’s RWE business does not sell data to clinicians to improve their point of care decisions.4

3 Flatiron also collects prospective data, meaning that Flatiron intentionally captures data that would generally not be included in a patient’s record to address particular research questions or support certain studies. To collect such data, Flatiron partners with community oncologists and academic medical centers. 4 The written direct testimony of Melisa Tucker (“Tucker”), Flatiron’s Vice President of Product Management and Operations for RWE, implied that community oncologists are customers of Flatiron’s RWE business line. (See Tucker Decl. ¶ 14 (“Flatiron’s customers include community oncologists, academic medical centers, . . . biopharmaceutical companies, regulatory agencies, and other healthcare authorities. Flatiron often creates custom data sets, tailored to these customers’ specialized . . . needs and preferences.”).) That statement appears to 6 Flatiron has developed methodologies and software systems for gathering, curating, and analyzing data from electronic health records. For example, to curate data, Flatiron has formulated rules governing how Flatiron converts information conveyed by physician notes, or other

raw data in a patient record, into numeric values for variables in Flatiron’s data set. Carson concedes that the specific rules, methodologies, and algorithms that Flatiron uses to curate and aggregate data from electronic health records are proprietary. Flatiron develops and implements these methods and systems using cross-functional teams consisting of software engineers, oncologists, clinical data specialists, data entry personnel, and others. For example, Flatiron’s clinical data team writes policies and procedures to govern how Flatiron’s data entry personnel curate data from unstructured records, and Flatiron’s research oncology team

must generally sign off on those policies and procedures.

be inconsistent with other trial evidence. Flatiron’s RWE business sells data to certain academic researchers, some of whom are also practicing physicians who may apply their research findings in their own practices. (Tr. 172:18-173:3, 221:12-18.) That sort of spillover does not alter the Court’s determination that Flatiron’s RWE business does not directly serve clinicians as such. 7 Research oncologists also describe clinically relevant concepts and rules, which software engineers incorporate into Flatiron’s software codes. 2. Flatiron’s Community Oncology Software Business Flatiron sells an electronic medical records software

product to doctor’s offices. Doctors use the software to record patient data, which Flatiron collects. 3. Trial Matching Services Flatiron’s OncoTrials product helps community oncologists identify and screen patients for clinical trials. A separate Flatiron product, called Patient Recruitment, helps pharmaceutical companies and other trial sponsors identify patients eligible for their trials. 4. The CGDB Flatiron does not own or operate a clinical laboratory, but it has a relationship with Foundation Medicine (“Foundation”), which runs a genomic testing lab

business. Through its collaboration with Foundation, Flatiron built a database of linked clinical and genomic data, known as the clinico-genomic database (“CGDB”). Flatiron uses the CGDB to serve pharmaceutical and biotech companies and academic researchers. Flatiron does not use the CGDB data to serve individual physicians making 8 treatment decisions with regard to individual patients.5 The CGDB also does not contain complete information about the frequency or use of genomic tests. Rather, pharmaceutical companies generally use the CGDB to identify how patients with a certain genetic abnormality respond to standard

therapies. C. CARSON’S EMPLOYMENT AT FLATIRON Carson began working at Flatiron as a Senior Medical Director in the Research Oncology Group in November 2016.6 Carson testified that he accepted the position, in part, because he believed that his prior experience analyzing

5 In response to questions on redirect, Tucker asserted that Flatiron is partnering with Foundation “on things like how do we surface, better surface test reports, genomic test reports back to individual providers.” (Tr. 231:11-13.) At most, Tucker’s testimony would indicate that Flatiron and Foundation are currently raising questions about whether they have the capacity and desire to do something with genetic data that would serve providers, not that they were doing so while Carson was at Flatiron. Moreover, the direct testimony of Tucker and Zach Weinberg (“Weinberg”), Flatiron’s Co-Founder, President, and Chief Operating Officer, contained much discussion of Flatiron’s partnership with Foundation, yet neither mentioned any contemplated project focused on providing genetic test reports to providers. A July 10, 2019 email regarding amendments being made to Flatiron’s “collaboration terms” with Foundation “to ensure that [the] collaboration agreement contemplates new initiatives and products” lists categories of “key” additions but makes no mention of any provider-focused project. (See Pl.’s Ex. 107.) The Court observes that Tucker gave this testimony only after she, having sat at Plaintiff’s counsels’ table during the trial, listened to Carson testify about the limited nature of his new role at Tempus. (See Tr. 250:13-16.) For these reasons, the Court declines to credit Tucker’s testimony on this issue. 6 Flatiron had four other Senior Medical Directors in the Research Oncology Group. 9 large data sets in graduate school and his previous employment experience made him a good fit. He added that the position at Flatiron also appealed to him because it allowed him to continue treating patients one day a week and to work closely with his longtime friend and former co-

author Dr. Amy Abernethy (“Abernethy”). At the time, Abernethy was Flatiron’s Chief Medical Officer and Chief Scientific Officer. Carson continued serving as a Senior Medical Director in Flatiron’s Research Oncology Group until September 26, 2019. Around October of 2018, Flatiron gave Carson the additional title of Oncology Senior Methodologist. In early 2019, Flatiron promoted Carson to Senior Medical Director, Level II. At Flatiron, Carson was responsible for providing guidance and oversight across a number of products in Flatiron’s RWE business segment. Carson supported

Flatiron’s biopharmaceutical clients, and sometimes researchers at academic institutions, in conducting research using Flatiron’s off-the-shelf or customized databases. Specifically, after a customer signed a contract with Flatiron, Carson oversaw the preparation of data

10 targeted to the customer’s needs. He also helped customers identify key insights from the data. Flatiron’s biopharmaceutical clients sought evidence to support drug development, applications for regulatory approval, and product placement. In general, these clients

wanted outcomes research -- that is, they wanted to understand the variables related to a particular outcome, such as survival or tumor response, across a population of cancer patients. Carson conducted this type of outcomes research in pursuit of his Ph.D. Accordingly, in his role at Flatiron, Carson applied his pre-existing knowledge and experience concerning research methods, oncology, and hematology. While Carson helped biopharmaceutical companies and other Flatiron customers use Flatiron’s products, he was not involved in how data produced by Flatiron was ultimately disseminated to or utilized by physicians. That

is, Carson was not involved in identifying ways for individual physicians to use Flatiron’s data to improve outcomes for their patients. Carson had various leadership and management responsibilities at Flatiron. In 2017, Carson became a member of Flatiron’s five-person RWE business leadership 11 team. Carson was the lead research oncologist for various products and projects, including: Flatiron’s customized data sets at certain points in time; Flatiron’s off-the- shelf databases for five of twenty cancer types or diseases; Flatiron’s hybrid control arms product; the CGDB;

and Flatiron’s West Coast expansion, an initiative to enlarge Flatiron’s customer base on the West Coast. As the lead research oncologist for the CGDB for periods of time in 2017 and 2018, Carson was involved in day-to-day decisions regarding CGDB products. His work in connection with the CGDB served pharmaceutical companies, which generally sought information about how patients with certain genetic characteristics responded to a particular treatment. D. FLATIRON’S COVENANTS AGREEMENT Flatiron generally requires all of its employees to sign the same Covenants Agreement. Flatiron conditioned

Carson’s offer of employment on his acquiescence to this agreement. Carson signed Flatiron’s Covenants Agreement shortly before beginning work at Flatiron. Prior to signing the Covenants Agreement, Carson received advice from an attorney regarding the agreement. Carson read and reviewed the Covenants Agreement but did 12 not negotiate the terms of its non-compete provision with Flatiron. Rather, consistent with the instructions set forth in the Covenants Agreement, Carson listed the inventions he had been working on prior to his employment at Flatiron and his existing non-disclosure or

confidentiality obligations. The Covenants Agreement contains provisions prohibiting the disclosure of confidential information, including trade secrets, business plans, pricing information, and customer information. Specifically, the Covenants Agreement defines “Confidential Information” as: all of the trade secrets, know-how, ideas, business plans, pricing information, the identity of and any information concerning customers or suppliers, computer programs (whether in source code or object code), procedures, processes, strategies, methods, systems, designs, discoveries, inventions, production methods and sources, marketing and sales information, information received from others that [Flatiron] is obligated to treat as confidential or proprietary, and any other technical, operating, financial and other business information that has commercial value, relating to [Flatiron], its business, potential business, operations or finances, or the business of [Flatiron]’s affiliates or customers, of which I may have acquired or developed knowledge or of which I may in the future acquire or develop knowledge of during my work for [Flatiron], or from my colleagues while working for [Flatiron].

(Pl.’s Ex. 3 at 2.) The non-disclosure clause provides: I will use the Confidential Information only in the performance of my duties for [Flatiron]. I will not 13 disclose the Confidential Information, directly or indirectly, at any time during or after my employment by [Flatiron] except to persons authorized by [Flatiron] to receive this information. I will not use the Confidential Information, directly or indirectly, at any time during or after my employment by [Flatiron], for my personal benefit, for the benefit of any other person or entity, or in any manner adverse to the interests of [Flatiron]. I will take all action reasonably necessary to protect the Confidential Information from being disclosed to anyone other than persons authorized by [Flatiron].

(Id.) In addition, the Covenants Agreement contains a non- compete provision (the “Non-Compete”). The Non-Compete provides that “for one (1) year after termination of [his] employment” the employee will not: directly or indirectly, whether as owner, sole proprietor, partner, shareholder, director, member, consultant, agent, founder, co-venture partner or otherwise, (i) do anything to divert or attempt to divert from the Company any business of any kind, including, without limitation, solicit or interfere with any of the Company’s customers, clients, members, business partners or suppliers, (ii) solicit, induce, recruit or encourage any person engaged or employed by the Company to terminate his or her employment or engagement, or (iii) engage, invest or participate in (x) any business that is similar to those which [Flatiron] has created, has under development or are the subject of active planning from time to time during [his] employment by the Company, or (y) a Competing Business. For purposes of this Agreement, a “Competing Business” means the business of providing software products, data analysis, data, clinical trial research services, analytics and electronic medical record systems to hospitals, physicians, community practices, health care centers, and pharmaceutical companies in the oncology industry[.] 14 (Id. at 5.) Nat Turner, Flatiron’s Co-Founder and Chief Executive Officer, stated in his deposition that he interprets the Non-Compete as barring Carson from working for Tempus even as a janitor. (Turner Dep. 130:7-14 (“Q: . . . So I’m trying to explore what your belief is about that contractual bar, and you’ve previously said that there’s no position that Carson could work –- Dr. Carson could work at within Tempus, including janitor, during that one year, correct? A: That’s correct.”).) Zach Weinberg (“Weinberg”), Flatiron’s Co-Founder, President, and Chief Operating

Officer, expressed the same view in his deposition. (Weinberg Dep. 147:23-148:2.) E. CARSON’S DEPARTURE FROM FLATIRON In late 2018 or early 2019, Carson began to consider leaving Flatiron. He testified that various considerations led him to explore employment opportunities elsewhere: a desire to work in the same city where his family lived, his wife’s interest in moving to Chicago, a desire to focus on work that would improve treatment outcomes for individual patients, concerns about the quality of data Flatiron was disseminating externally, dissatisfaction with Flatiron’s

15 corporate culture, and questions about his ability to grow professionally at Flatiron. At the outset of Carson’s job search, he expected his family to remain in St. Louis, Missouri and, accordingly, pursued opportunities in that area. Washington University

in St. Louis offered him a position, and Carson was also presented with a prospect to invest in a local venture capital fund. Carson felt, however, that these opportunities would not allow him to use his full skill set or have a large-scale impact on the treatment of cancer patients.7 Carson first heard about a position at Tempus from a recruiter who reached out to him in April 2019. As described in more detail below, he interviewed for a position at Tempus in May and June 2019. See infra Section I(G). Meanwhile, Carson and his family decided to move to

Chicago. Carson’s wife was considering a position at Northwestern University, and moving to Chicago would allow their family to be near her aging parents. In light of this

7 Carson also stated that the potential profit from the investment opportunity was not comparable to the salary he could earn at a full- time job. 16 decision, Carson pursued opportunities at Northwestern University and Rush University, and he received an offer from Rush University. Ultimately, on August 13, 2019, Carson accepted a position at Tempus as Vice President of Clinical Solutions.

See infra Section I(G). On August 27, 2019, Carson gave Flatiron formal notice of his decision to depart Flatiron and join Tempus. Carson continued working for Flatiron for approximately one month after he gave notice. He testified that, to his knowledge, Flatiron did not restrict his access to any information during this time. Carson did, however, limit his exposure to potential trade secrets and confidential information during that time. Carson’s employment at Flatiron concluded on September 26, 2019. F. TEMPUS’S BUSINESS

Tempus is a healthcare technology company focused on providing physicians with data to inform their decisions regarding the treatment of individual patients. Tempus’s clinical laboratory is at the core of its organization. As discussed in more detail below, Tempus’s clinical lab testing business serves physicians, while separate business lines serve pharmaceutical companies. Because Tempus’s physician and pharmaceutical customers have different 17 needs, Tempus employs separate sets of employees to serve them. 1. Tempus’s Clinical Lab Testing Business Tempus’s clinical laboratory provides lab testing services, including genomic sequencing, to help doctors and

other health care providers understand the nature of a patient’s cancer. Individual doctors send samples of patients’ genetic material to Tempus. Tempus then analyzes the samples, identifying any genetic mutations present in the samples, and returns a lab report to the physicians. A Tempus lab report contains not only the patient’s test results but also information to help the physician understand and act upon the results. For example, the report may identify drugs that are indicated or contraindicated based on the patient’s mutations and any clinical trials for which the patient may be eligible.8 It also identifies whether the patient has a hereditary

mutation, thereby alerting the physician to consider screening the patient’s family members. In some instances,

8 To identify clinical trials for which a patient may be eligible, the report uses information that is publicly available on https://clinicaltrials.gov. (Accord Lefkofsky Dep. 15:18-16:7 (“[T]here’s nothing kind of magical about that”).) 18 Tempus’s report provides an insight based on its aggregated real-world data set. For example, the report may show how the patient’s tumor mutational burden9 compares to that of other patients in Tempus’s data set -- that is, whether it is in the 50th or 80th percentile. An online portal

accessible to physicians shows the various therapies that have been provided to patients in Tempus’s data set that are similar to the physician’s patient, along with the physician-reported outcomes of those patients.10 Carson contends and sought to prove at trial that Tempus’s clinical lab testing business does not compete with Flatiron. He argues that Flatiron does not own or operate a clinical laboratory and does not provide patient- specific lab reports to doctors or other health care

9 Tumor mutational burden, according to Ryan Fukushima (“Fukushima”), Tempus’s Chief Operating Officer, is a measurement of the quantity of mutations found in a tumor. 10 During the cross-examination of Fukushima, counsel for Flatiron asked Fukushima to point to any statement in his declaration that put the Court on notice that the Tempus lab report contained information derived from real-world data. For reasons discussed below, this was all much ado about nothing. See infra Section I(H). However, the Court notes that Fukushima’s declaration unambiguously conveyed that Tempus’s lab reports may present information derived from real-world data. (See Fukushima Decl. ¶ 24 (explaining that the lab report may present “information about other similar patients that have been tested by Tempus with certain characteristics”); see also id. at ¶ 25 (“In some circumstances, Tempus may also provide an insight that has been observed in Tempus’s aggregated data set showing how often a new therapy has been given in the real world and the real world outcomes associated with that therapy.”).) 19 providers. (See, e.g., Meropol Dep. 35:13-16 (“In terms of the narrow question of whether Flatiron provides a report on an individual molecular test to an individual physician, the answer is we do not”).) In fact, according to Carson, Flatiron’s RWE business does not sell data to physicians to

improve their point of care decisions. 2. Tempus’s Clinical Data Aggregation Business Tempus’s pharmaceutical data licensing business licenses subsets of Tempus’s aggregated real-world data to pharmaceutical companies. This segment of Tempus’s business competes with the partnership of Foundation and Flatiron. Tempus provides pharmaceutical companies with various types of real-world data, including aggregated genomic data generated in Tempus’s clinical laboratory, clinical records data from electronic medical record systems or electronic data warehouses at hospitals, and medical images from Tempus’s clinical laboratory. The pharmaceutical companies

then, using their own employees, analyze the data to inform their research and business decisions. 3. Tempus’s TIME Trial Program Tempus also serves its pharmaceutical clients through its TIME Trial Program, which aims to maximize patient participation in clinical trials. Tempus pre-screens 20 patients for clinical trials and uses a trial-matching algorithm to identify trials in which patients may enroll. Tempus then works to enroll local oncology sites in trials that might otherwise be limited to academic medical centers. G. TEMPUS’S RECRUITMENT AND HIRING OF CARSON

Tempus is a growing company and, as of early 2019, was searching for high-quality talent. Specifically, Tempus was looking for practicing oncologists, neurologists, and radiologists with experience managing research teams and designing and producing studies using clinical trial or real-world data. As Tempus’s Chief Operating Officer, Ryan Fukushima (“Fukushima”) oversees human resources and recruiting efforts at Tempus. In early 2019, he had a conversation about Tempus’s recruiting needs with Noel Auguston (“Auguston”) at Perspective, LLC (“Perspective”), a

recruiting firm that Tempus retained. Pursuant to the terms of Tempus’s engagement letter with Perspective, Perspective prepared a position specification for “VPs of Research” roles to memorialize the criteria that Perspective would apply to identify candidates. The specification described the ideal candidates as oncologists, neurologists, or 21 radiologists “from highly respected academic institution[s]” who were currently practicing or working in industry and had experience “leading and galvanizing medical teams designing and executing studies.” (Def.’s Ex. 45 at 62.)

The specification explained that Tempus’s VPs of Research for Oncology, Neurology, and Radiology would be expected to: Bring a clinical perspective to the organization, informing the product/service portfolio and strategy. Advise the organization on clinical evidence needs/gaps.

Drive outcomes research strategy and execution that is consistent with the current and future product and service offerings, incorporating datasets to support Tempus’s treating physicians.

Cultivate and maintain relationships with the appropriate national and international investigators to achieve broad visibility and support for the company’s portfolio. Interact with key opinion leaders at hospitals to assure incorporation of the latest advances in clinical and treatment guidelines into the clinical plan.

Maintain a foothold in the clinical setting, ideally by continuing to practice medicine on an agreed upon schedule.

(Id. at 61.) The specification also indicated that the VPs of Research would “facilitat[e] development plans and specializ[e] in assembling real-world datasets and 22 developing approaches to integrate technology into dataset creation.” (Id.) As Auguston explained, however, position specifications do not necessarily reflect actual available positions; in this case, the VP of Research specification

was crafted as a “bucket” to reflect multiple potential positions that Tempus would define with greater precision once Tempus had identified desirable candidates. (Auguston Dep. 34:21-37:13.) According to Auguston, such a recruiting strategy is sensible for growing companies, such as Tempus, that are hiring in the context of rapid organizational evolution and expansion. (See id. at 35:19-21.) Beginning in March 2019, Perspective commenced a broad search for candidates for Tempus’s “VP of Research, Oncology” position. Perspective routinely provided Tempus with “Progress Reports” on Perspective’s recruiting efforts.11 As the Progress Reports make clear, Perspective

did not specifically target Flatiron employees. Rather, Perspective considered oncologists with a variety of

11 See, e.g., Def.’s Exs. 49, 56, 58, 65, 69, 77, 85, 97, 104, 110, 131. 23 backgrounds, including numerous professors and many oncologists at academic medical centers. Carson was among the candidates Perspective identified for the VP of Research, Oncology position. Perspective determined that Carson fit the search criteria because of

his active clinical practice at an academic medical center and his experience conducting research using information from patient records, as demonstrated by his dissertation. In April 2019, Sara DeAmicis (“DeAmicis”), a recruiter at Perspective, contacted Carson on behalf of Tempus. Carson expressed interest in the position while flagging that he was subject to a non-compete. Meanwhile, on May 20, 2019, Tempus hired Dr. Michael Axelson (“Axelson”), another oncologist, to support

Tempus’s pharmaceutical business lines as Head of Clinical Research. Axelson previously worked as a Medical Officer in the Office of Hematology and Oncology Products at the Food and Drug Administration before joining Bristol Myers Squibb as Head of Oncology. While Axelson met Tempus’s immediate need for an oncologist to support Tempus’s pharmaceutical business line, Tempus still sought an oncologist who could support Tempus’s clinical solutions business by identifying

24 ways to increase physicians’ utilization and understanding of Tempus’s services. The day after Tempus hired Axelson, Fukushima had a preliminary phone conversation with Carson. Perspective made Fukushima aware of Carson’s Non-Compete prior to that

conversation, and Carson also brought up the Non-Compete during the call. Fukushima indicated that it would be worthwhile for Carson to engage in further discussions with Tempus because Tempus could potentially identify a role for Carson that would not overlap with Carson’s role at Flatiron. At Fukushima’s recommendation, Carson subsequently interviewed with several Tempus representatives while in Chicago for the 2019 American Society of Clinical Oncology annual meeting. Carson met with Erik Lefkofsky (“Lefkofsky”), Tempus’s Founder and Chief Executive Officer, and Gary Palmer (“Palmer”), Tempus’s Chief Medical

Officer, on May 31, 2019. He met with Fukushima and Kevin White (“White”), Tempus’s Chief Scientific Officer, on June 4, 2019. Rather than focusing on Carson’s fit for a particular position, Lefkofsky, Palmer, Fukushima, and White had relatively high-level conversations with Carson. Carson testified that he specifically recalls Lefkofsky 25 suggesting that positions besides the VP of Research, Oncology role were potentially available. At his June 4 interview with Fukushima, Carson explained that he was interested in leaving Flatiron, in part, because he wanted to transition away from working

with pharmaceutical clients and focus more on improving the treatment of individual patients. At the same time, Carson cautioned that his Non-Compete could pose challenges. In his conversations with Tempus personnel, Carson specifically expressed his concerns that his Non-Compete would prevent him from taking the VP of Research, Oncology position, as described by Perspective, because it was similar to his current role at Flatiron. Accordingly, Fukushima and Carson’s conversation on June 4 explored how Tempus’s clinical testing business could potentially leverage Carson’s research skills and his experience as a practicing oncologist. Carson did not have further

discussions with Palmer, White, or Lefkofsky after his interviews with them in Chicago. Based on the interviews, Lefkofsky, White, and Palmer expressed to Fukushima that they believed Carson would be a good fit at Tempus. Lefkofsky and Fukushima agreed that Tempus could use an oncologist to work in Tempus’s clinical 26 lab testing business line and that Carson had the qualifications and skills to be successful in that position. On June 26, Fukushima obtained a copy of Carson’s Non-Compete from Perspective. After reviewing Carson’s Non- Compete with Tempus’s in-house counsel Erik Phelps

(“Phelps”) and conducting an assessment of Tempus’s need for an oncologist to work in the clinical laboratory business, Tempus decided to create a role for Carson that would not overlap with his role at Flatiron. On June 28, 2019, Carson spoke with Fukushima and Phelps by phone to discuss whether Tempus could offer him a role sufficiently distinct from his role at Flatiron such that Carson could avoid litigation with Flatiron over his Non-Compete. Carson and Fukushima discussed how Tempus could utilize Carson in a role in Tempus’s clinical testing business that would focus on improving physician’s utilization of the information in Tempus’s lab reports.

Carson was excited about this position and viewed it as an excellent opportunity to use both his experience treating patients and his research skills. Carson also viewed the role as aligned with his conviction that improving public health requires investing in dissemination and implementation work –- that is, work focused on ensuring 27 that the strategies researchers identify to improve health and prevent disease are effectively communicated to and implemented by health care providers. (Id. ¶¶ 53-54; Tr. 106:8-24, 108:3-15.) In Carson’s own words: The challenge of moving health research innovations from discovery to practice is complex and multifaceted. One of the most critical issues impeding improvements in public health today is the enormous gap between what we know can optimize health and health care (research and discovery) and what actually gets implemented by clinical physicians (dissemination and implementation). The science of dissemination and implementation (also known as “D&I”) seeks to narrow this gap by studying and understanding how best to cause evidence-based strategies that improve health and prevent disease to be effectively delivered in clinical and public health practice.

Carson Decl. ¶ 53. Following the June 28 phone conversation, Lefkofsky and Fukushima worked to formally define Carson’s role. They decided to title Carson’s role as that of Vice President of Clinical Solutions to make clear that Carson would work exclusively in Tempus’s clinical laboratory business. Tempus sent Carson an offer for the position of Vice President of Clinical Solutions on July 30, 2019. The offer provided the following description of Carson’s role: Your primary responsibilities will be focused on advancing our clinical solutions, which we offer predominantly through the genomic profiling we do for physicians in our CLIA/CAP lab, together with assisting Tempus’ efforts to analyze data that Tempus 28 has collected (and will continue to collect) to help identify key insights from that data that may improve clinical decisions and overall therapeutic care. Additionally, we believe that your role as a practicing physician will be instrumental to helping our teams in identifying ways to bring the insights derived from Tempus’ data back to other treating physicians in ways that will help improve patient outcomes, which is central to Tempus’ mission.

(Def.’s Ex. 106 at 2.) The offer letter further specified that Tempus “has no interest whatsoever in any information or intellectual property [Carson] may have regarding [his] previous employer.” (Id.) It directed that, “[i]n the event any employee at Tempus requests information from [Carson] that would require [him] to violate the terms of any agreement [he] entered into with [his] former employer, [Carson] should not provide any such information, inform the requesting employee that [he] is not allowed to disclose any such information, and also inform Tempus’s General Counsel.” (Id.) In addition, the offer was conditioned on Carson signing a certification confirming that he understood he was “prohibited from using [or] disclosing . . . Flatiron confidential information in the scope of [his] Tempus duties and responsibilities” and that doing so would be “grounds for discipline, up to and including termination.” (Id. at 2, 6-7.) 29 After Tempus sent Carson the July 30, 2019 offer letter, Carson and Tempus engaged in further discussions regarding Carson’s compensation, and how they would proceed in the event Flatiron sued to enforce the Non-Compete. Tempus agreed to pay legal fees associated with a suit by

Flatiron to enforce the Non-Compete, and Tempus also agreed to pay Carson up to six months’ salary in the event he was enjoined from working for Tempus. Tempus revised Carson’s offer letter to reflect these discussions. Carson accepted the position by signing the offer letter and accompanying materials on August 13, 2019. After accepting his offer from Tempus and giving notice to Flatiron, Carson and Fukushima corresponded about Tempus’s broad, long-term vision and whether Carson would eventually be able to take on a role in shaping that vision. Carson’s communications indicated that he looks forward to working at a company where, in the long term, he

will be involved in strategic planning decisions. H. CARSON’S POSITION AT TEMPUS According to Fukushima, Lefkofsky, and Carson, as Vice President of Clinical Solutions at Tempus, Carson will be responsible for improving physicians’ understanding and use of Tempus lab reports. Because the next generation 30 sequencing technology that Tempus uses is a relatively new technology, doctors are still learning how to use patients’ genomic reports to make treatment decisions. Consequently, Carson’s role will entail assessing how care providers are currently using the Tempus lab reports and identifying ways

to help physicians use the reports more effectively to make informed treatment decisions for their patients. For example, Carson may assess whether there are particular types of cancer for which physicians are not ordering genetic tests even though genetic tests could inform decisions about how to treat that type of cancer. He may also evaluate whether physicians are using information in Tempus’s lab reports about germline mutations, defined as mutations which can be passed down to children, to determine when genetic counseling and/or screening for a patient’s family is appropriate. In performing this role, Carson will oversee focus groups comprised of health care

providers who use Tempus lab reports. These focus groups will help Tempus gather ideas for improving care providers’ understanding and use of Tempus’s lab reports to inform their treatment decisions. Based on this articulation of Carson’s position at Tempus, the Court is persuaded that Carson’s role as Vice 31 President of Clinical Solutions at Tempus would not overlap with his former employment at Flatiron. At Tempus, Carson will be analyzing a type of data different from that which he evaluated at Flatiron, in a different manner, and in service of a different set of customers. At Flatiron,

Carson helped pharmaceutical customers build and analyze data sets to understand the role of different variables at the population level, with the ultimate goal of assisting the companies’ drug development efforts. Carson had no interactions with physicians to improve therapeutic care while at Flatiron. In contrast, at Tempus, Carson’s role would involve helping physicians understand how information presented in Tempus’s lab reports –- information generated without Carson’s involvement -- should guide the physicians’ treatment decisions. As Carson stated: “[L]ooking at populations to derive inferences [as he did at Flatiron] is very different from looking at individual

patient reports, clinical reports and looking at how to apply that evidence,” as Carson will do at Tempus. (Tr. 21:17-19). Put simply, Carson’s role at Flatiron involved generating new evidence to assist pharmaceutical companies, while Carson’s role at Tempus is to assist doctors in applying existing evidence to treat individual patients. 32 Moreover, Carson will not participate in generating the information that goes into Tempus’s lab reports. As Carson testified, “rather than being involved in the data curation process, I will instead receive already-completed genomic reports and assist doctors in interpreting this

finished report to better understand the best treatment options for individual cancer patients.” (Carson Decl. ¶ 85.) Indeed, while Carson’s Non-Compete remains in effect, he will not be extracting and analyzing data from electronic health records in any manner that could overlap with his role at Flatiron. On this point, according to Fukushima, “[Carson] will not be curating clinical data from electronic health records or performing data analysis for pharmaceutical clients.” (Fukushima Decl. ¶ 58.) Any data Carson collects and analyzes will be data regarding physicians’ use of Tempus’s tests. Carson may, for example, analyze data on the types of cancer for which

physicians order genetic tests or how often physicians prescribe patients the drugs that the lab report identifies as potentially appropriate for them. The Court finds that

33 conducting such analyses will not risk the use or disclosure by Carson of any Flatiron trade secrets.12 Because Carson will not be involved in curating,

structuring, or analyzing the data that is ultimately presented in Tempus’s lab reports, whether the evidence presented in the report is based on real-world data is irrelevant to whether Carson’s role at Tempus overlaps with his role at Flatiron.13 Regardless, the Court observes that the vast majority of information presented in Tempus’s lab reports comprises information obtained from clinical trials or published studies, not information drawn from real-world settings. As noted above, Tempus lab reports may, on occasion, present insights derived from real-world data.14

12 For example, without risking the disclosure of any Flatiron trade secrets, Carson could request from Tempus data on whether, after a Tempus report identifies an indicated drug for a particular patient, doctors prescribe the drug. The statistical methodologies Carson would be using in this project are of the sort he unquestionably mastered in graduate school, if not earlier. 13 For the same reason, whether Tempus has a single database that serves both pharmaceutical companies and healthcare providers is irrelevant. 14 When Fukushima testified that Carson was “going to be using aggregated data collected from real world settings,” Fukushima was referring to how Carson would be helping physicians understand Tempus’s lab reports, which occasionally include insights based on real-world data -- for example, that a patient’s tumor mutational burden is in the 85th percentile compared to that of other patients in Tempus’s real- world data sets. (Tr. 420:3-5; see Tr. 445:12-447:25.) The Court is persuaded that Fukushima did not mean that Carson would be working with or having contact with the Tempus employees who aggregate and structure data sets for pharmaceutical companies. Flatiron’s Rule 62(d) 34 Carson will not need information about how the data underlying Tempus’s lab reports were curated and analyzed to help physicians understand the reports. Indeed, any insights from real-world data that are contained in the reports will be presented alongside all available “material

information (e.g., supporting PubMed citation, size of the population underlying the Insight) and/or statistical analyses (e.g., p-values, confidence intervals) intended to give . . . adequate context to evaluate the potential relevance of the Insight to the patient.” (Pl.’s Ex. 203.) Carson will remain an “end user” of that data, working to help physicians understand the actionability of the insights generated by others. (Tr. 113:21.15)

Memorandum (“Flatiron Mem.”) misinterprets other portions of Fukushima’s testimony as indicating that Carson will be analyzing real- world data in a manner that would overlap with his role at Flatiron. (See Mem. at 8-9.) Fukushima has explained that Carson’s dissemination and implementation work, which is focused on “clinical actionability,” will “help Tempus prioritize solutions that will leverage Tempus’s data to drive better clinical decisions at the point of care.” (Fukushima Decl. ¶ 63.) Similarly, he has stated that Carson’s dissemination and implementation work will “make sure Tempus is . . . providing [doctors with] actionable insights from Tempus’s large library of data.” (Id. ¶ 58.) Neither of these statements in any way suggests that Carson himself will be analyzing data from patient records to generate insights that physicians can apply. 15 While the Court’s discussion focuses on Tempus’s clinical reports, the foregoing findings also apply with respect to the additional information available in the portal through which some physicians access the reports. The portal provides a list of aggregated historical data on the drugs prescribed to patients with specific types of cancer 35 The Court credits Carson’s and Fukushima’s testimony that Carson’s role at Tempus will be limited in the manner described above for the duration of Carson’s Non-Compete. The alignment of Carson’s qualifications and interest in dissemination and implementation work with Tempus’s

business needs and profit motives for conducting dissemination and implementation work lends credibility to their representations regarding the scope of Carson’s role. The Court finds unpersuasive Flatiron’s various attempts to establish that Carson and Fukushima did not accurately represent the bounds of Carson’s role at Tempus. For example, Carson’s expression of interest with regard to the VP of Research, Oncology position does not indicate that Carson intends to perform the functions described in the VP of Research, Oncology position specification while his Non-Compete is in effect. While Carson expressed interest in hearing more about Tempus and that role in

particular, he consistently articulated to both Perspective

and the physician-reported response to those drugs. (Lefkofsky Dep. 18:19-24:13.) Lefkofsky explained that “if [Carson] was going to work on cohort, the kind of things he would be doing is looking at the presentation of data we have and trying to make it more useful.” (Id. at 47:5-8.) That type of presentation design work would not overlap with Carson’s role at Flatiron. 36 and Tempus his concern that he could not accept the VP of Research, Oncology position in light of his obligations to Flatiron. Moreover, the Court credits Carson’s testimony that he believed it would be not only legally problematic but also inappropriate to accept a position at Tempus that

overlapped with his position at Flatiron. As Carson expressly testified: “[C]ertainly a position like this would have so much overlap with what I was doing at Flatiron, that not only would I be sued, but it would be inappropriate.” (Tr. 47:12-15; see also Pl.’s Ex. 8 at 4 (Carson: “We invested so much [at Flatiron] and I want to see it succeed as part of a greater evidence ecosystem”).) That Perspective and Tempus continued referring to Carson’s role in their communications with one another as “VP of Research, Oncology” after he was hired for the VP of Clinical Solutions role, see, e.g., Def.’s Ex. 135, is without probative significance. Trial testimony indicated

that the position specifications that Perspective uses often do not reflect actual roles but function to attract highly-qualified candidates around whom Tempus can tailor roles in a manner consistent with the candidates’ and Tempus’s objectives. (See Auguston Dep. 34:21-37:13; Lefkofsky Dep. 66:10-67:7.) Although it is “[v]ery common 37 for a role to change around a candidate,” the progress reports that Perspective sent to Tempus continued to reflect the titles of the initial position specifications. (Auguston Dep. 36:17-18, 71:22-24.) Because Perspective always referred to hires using the titles from the initial

specifications, it made sense for Tempus to use the same terms when communicating with Perspective. The communications demonstrate nothing more than that Perspective and Tempus used a shared, albeit imprecise, language. Nor does Carson’s offer letter suggest that Carson’s actual role at Tempus will be broader than Carson, Fukushima, and Lefkofsky have represented in this litigation, such that it would overlap with Carson’s role at Flatiron. The offer letter identifies one of Carson’s responsibilities as “assisting Tempus’ efforts to analyze data that Tempus has collected (and will continue to

collect) to help identify key insights from that data that may improve clinical decisions and overall therapeutic care.” (Def.’s Ex. 106 at 2.) Carson explained that this provision refers to identifying insights from data “on utilization of the Tempus test” –- for example, “who is using the test, who isn’t using the test, [and] other 38 pieces of information” so that Tempus can improve the dissemination and utilization of this test to “help change therapeutic care.” (Tr. 22:24-23:17.) Lefkofsky similarly interpreted the provision as tasking Carson with “helping make the test more useful and

helping doctors understand the test and . . . having them get the most out of it.” (Lefkofsky Dep. 39:13-20.) If viewed in isolation, that sentence of Carson’s offer letter could be interpreted more broadly. However, read in context with the other restrictions Tempus has imposed on Carson –- namely, that Carson is prohibited from using or disclosing Flatiron confidential information in his work at Tempus -- this provision cannot reasonably be read to suggest that Tempus has hired Carson to analyze real-world data.16 For this reason, that Carson’s offer letter does not contain an explicit list of prohibited activities is irrelevant. In light of the broad restrictions in the company

16 Similarly, while Carson will have some discretion to define his position description and authority to decide how he spends his time at Tempus, nothing in the record indicates that Tempus would allow Carson, at least while the Non-Compete remained in effect, to task himself with functions similar to those he performed at Flatiron. To the contrary, doing so would violate Tempus’s explicit prohibition that Carson not use or disclose Flatiron confidential information, thereby subjecting himself to discipline or potential liability. 39 certification on which Carson’s offer was conditioned, any list of prohibited activities would be superfluous. That Palmer stated that he expected Carson to work on the pharmaceutical side of Tempus’s business is unsurprising and does not alter the Court’s analysis for

two primary reasons. First, Palmer will not be supervising Carson’s day-to-day responsibilities. Tempus has a matrix organizational structure in which employees receive day-to- day supervision from one person while reporting directly to someone else. As Lefkofsky testified: “[I]t’s a matrix, so . . . the doctors might all be under the doctors, but it doesn’t mean they’re actually supervising what they do or really hav[ing] any day-to-day involvement with what they do.” (Lefkofsky Dep. 32:11-20; see also Axelson Dep. 22:12- 23:2 (explaining the distinction between a “direct manager” and “day-to-day boss” at Tempus).) Thus, in Tempus’s matrix organizational structure, physicians report to other

physicians. Under this managerial arrangement, though Carson will formally report directly to Palmer, who is a physician and Tempus’s Chief Medical Officer, Michael Yasiejko (“Yasiejko”), as the Business Head of Clinical Testing, will oversee Carson’s day-to-day responsibilities.

40 Second, aside from conducting a high-level interview of Carson on May 31, 2019, Palmer had minimal involvement in recruiting Carson and developing Carson’s role. Fukushima and Lefkofsky largely handled this process themselves. Given that Carson will be working more closely

with Yasiejko than Palmer on a day-to-day basis, that Fukushima and Lefkofsky did not involve Palmer more in hiring Carson and defining his position is unsurprising. The Court notes that Tempus demonstrated that it took seriously and endeavored to facilitate Carson’s compliance with his Covenants Agreement obligations to Flatiron. Such showing, and affirmative steps consistent with it, was manifest in the manner in which Tempus, in consultation with its General Counsel, carefully formulated a role at Tempus for Carson that would not overlap with the functions and services he performed at Flatiron. This is therefore not a case in which an employee, acting in concert with and

with the support of a prospective employer, sets out, knowingly and purposefully, to flout a non-compete agreement. Nor is it suspect that after his Non-Compete expires Carson’s role may expand to encompass broader responsibilities, potentially even including conduct that is now prohibited by the Non-Compete. In fact, 41 acknowledging this point, Weinberg testified that the day after Carson’s one-year Non-Compete expires, Carson is free to work in any position at Tempus. (Tr. 310:1-4.) Nor does other evidence in the record contradict Carson’s testimony regarding the limits confining his role

at Tempus while his Non-Compete is in effect. It is not surprising that a person possessing Carson’s qualifications and experience would want to work for an employer that would seek and value his professional contribution on matters regarding the company’s vision and agenda, in Carson’s case thus enabling him to maximize his impact on oncology care delivery. Reflecting this observation, Carson asked Fukushima: “Will I have the opportunity (a seat at the table) to help shape [Tempus’s] vision?” (Pl.’s Ex. 17 at 2.) (See also Pl.’s Ex. 20 (“I am ready to . . . have a seat at the table where decisions are made.”); Pl.’s Ex. 8 at 5 (“I’m going to dive into Tempus to find out what I can

do there to make a difference and help shape the agenda there . . . .”).) Nonetheless, the Court credits Carson’s representation that in the short-term, meaning “at least twelve months,” he will be advancing Tempus’s mission by focusing on dissemination and implementation projects and would be barred from playing the larger policy and 42 decision-making role he envisions. Indeed, Carson’s testimony and written communications express recognition of the distinction between his short-term and long-term roles at Tempus. He stated, for example, that: “The opportunity to lead vision within one of the domains of the broader

vision is fine. Other pieces can fall into place later once I have deeper experience and have demonstrated capabilities internally.” (Pl.’s Ex. 17 at 1.) The Court is not persuaded by Flatiron’s argument that Carson’s short-term role will require him to discuss Tempus’s data curation and analysis processes with other Tempus employees. Whether Carson would even require such information to perform his short-term services is doubtful. Moreover, Flatiron incorrectly assumes that Carson would need to converse with other Tempus employees to obtain such information. Carson could, of course, request written descriptions of all of Tempus’s data curation rules and a

dictionary of all the variables in Tempus’s data sets. Carson would not tip off any Tempus employees to Flatiron’s trade secrets by requesting other Tempus employees to provide him with broad categories of written information. On this point, Carson has demonstrated that he can draw the line between knowledge he possesses by virtue of his 43 education and experience prior to joining Flatiron, and potentially confidential knowledge he possesses by virtue of his employment at Flatiron. The Court finds evidence in the trial record indicating that Carson has demonstrated ability to observe confidentiality obligations in various

settings, including while working with competing pharmaceutical companies at Flatiron, while serving as a doctor, and while making public presentations during his employment at Flatiron. (See, e.g., Tr. 143:9-144:21; see also Meropol Dep. 67:6-72:11.) In sum, the Court finds that Carson’s proposed role at Tempus differs substantially from his prior work at Flatiron. Further, the Court is persuaded that, in performing his duties at Tempus, Carson will be drawing largely upon the research skills and medical knowledge he developed in pursuit of his Ph.D. and as a practicing physician. To this extent, he will be working in a part of

Tempus’s business that does not compete with Flatiron, that serves different customers than does Flatiron’s RWE business, and that provides a product that is not sufficiently similar to any Flatiron product. For these reasons, Carson’s role at Tempus will not require him to

44 draw upon any Flatiron confidential or proprietary knowledge. I. TEMPUS’S MECHANISMS FOR ENFORCING THE NON-COMPETE As discussed above, Carson’s offer from Tempus was conditioned on Carson signing a certification confirming

that he understands he is “prohibited from using [or] disclosing . . . Flatiron confidential information in the scope of [his] Tempus duties and responsibilities” and that doing so would be “grounds for discipline, up to and including termination.” (Def.’s Ex. 106 at 2, 6-7.) Fukushima confirmed that, in addition to promoting compliance with Carson’s obligations to Flatiron through the threat of discipline, Tempus has processes and systems in place to prevent a violation of those obligations. Specifically, Fukushima explained that access to Tempus’s real-world data is password protected and restricted on a need-to-know basis, and that Tempus would not grant Carson

access to that data. J. PURPORTED TRADE SECRETS AND CONFIDENTIAL INFORMATION Prior to his departure from Flatiron, Carson returned all Flatiron documents and devices, deleted the Flatiron email account from his phone, and removed himself from company Slack groups. (Carson Decl. ¶ 98.) Carson signed an 45 acknowledgment confirming that he destroyed all Flatiron documents and materials in his possession. (Def.’s Ex. 150.) Flatiron does not dispute this testimony but asserts that Carson retains knowledge of various categories of Flatiron’s purported trade secrets and confidential

information. 1. Methods and Technical Information a. Data Curation and Aggregation Carson testified that, in his role at Flatiron, he had access to confidential and proprietary information regarding Flatiron’s aggregated data sets, including the variables available in those data sets. (Tr. 31:11-32:2.) But that does not mean that Carson would be expected to recall all of the variables contained in even a single data set. (See Tr. 124:22-125:8.) Because he was involved in preparing data targeted to specific customers’ needs, however, Carson would likely be able to recall the

capacities and limitations of Flatiron’s off-the-shelf and customized data sets, including the CGDB. That is, he can likely recall some of the variables that are and are not available in those data sets. Carson also testified that he had access to confidential and proprietary information regarding 46 Flatiron’s data curation rules and methods for data aggregation. (See Tr. 33:11-21.) For example, Carson provided information about relevant clinical and medical concepts that needed to be incorporated into Flatiron’s data curation process. Other team members, including

statisticians, clinical data specialists, software engineers, nurse practitioners, and others, used the information provided by Carson to develop specific rules or algorithms. Carson then reviewed and signed off on the final rules and algorithms. The Flatiron data curation rules are far too complex and voluminous for Carson to have committed them entirely to memory. However, Carson likely recalls some confidential information about how Flatiron resolved certain challenging data curation issues that were escalated to him. For example, he may remember some information about how Flatiron curates the disease progression, response,

mortality, and line of therapy data points. Flatiron continuously reviews and updates its data curation rules. The need for revision arises when, for example, new therapies come on the market or new uses are identified for existing therapies. Melisa Tucker (“Tucker”), Flatiron’s Vice President of Product Management 47 and Operations for RWE, testified that after a period of six months, Flatiron’s rules “may be somewhat outdated” although not in need of complete revision. (Tr. 226:15- 228:5.17) She estimated that “[l]ess than 10 percent” of rules would be “completely different” in six months,

compared to between “ten and twenty percent” after one or two years. (Id. at 227:24-228:5.) Flatiron asserts that given Carson’s exposure to a project involving machine learning, he could be expected to recall some confidential information about Flatiron’s successes and failures with using machine learning to predict certain data points. (See id. ¶¶ 51-52.) But the only document Flatiron offers as evidence of Carson’s work on this project dates the project to August 2018. Flatiron has not demonstrated that this information is not stale. In fact, Weinberg acknowledged that information that is over a year old is of diminished value to competitors. (See

Weinberg Decl. ¶ 33.)

17 Weinberg’s written direct testimony stated that Carson was familiar with “Flatiron’s proprietary methodology for driving research from concept to data interpretation, so as to turn data analysis into a valuable product.” (Weinberg Decl. ¶ 49.) The Court is not persuaded that Flatiron possesses any such overarching proprietary methodology for conducting research. 48 Even so, such information is not relevant to Carson’s role at Tempus because he will not be participating in data curation and analysis projects similar to those which Flatiron conducts. With regard to the CGDB, in particular, no Flatiron witness has identified knowledge Carson gained

from helping pharmaceutical companies utilize the CGDB’s de-identified data that would bear upon his work at Tempus helping individual physicians understand and utilize Tempus’s patient-specific lab reports, which present data that have already been curated and analyzed. Moreover, reiterating a point the Court stressed above, even if Carson recalled some Flatiron confidential information, it does not necessarily lead to a conclusion that he would reveal it, internally or externally, while working at Tempus during the remaining months of the Non- Compete. Carson testified credibly that he would not, and the Court credits the evidence Carson submitted describing

the steps he and Tempus took to establish safeguards and limitations to ensure that he does not violate the Non- Compete while it remains in effect. b. Potentially Patentable Technologies Carson attended several Flatiron patent review committee meetings in 2016 and early 2017 at which 49 participants discussed confidential information regarding technologies that could potentially be patentable. However, none of Flatiron’s witnesses explained how any specific information shared at those meetings in 2016 or 2017 would, if disclosed by Carson in 2020, harm Flatiron. On this

score, the Court again repeats Weinberg’s acknowledgment that Flatiron information that is over a year old is of diminished value to competitors. c. Clinical Trial Matching Trial evidence indicated that Carson did not participate in Flatiron’s clinical trial matching business and knew very little about it. Carson recalled only one occasion, in 2017, when he learned about a clinical trial matching project and expressed belief that Flatiron ultimately did not pursue the project. The Court found Carson exceptionally forthcoming and credible when asked, during cross-examination, about his access to and knowledge

of confidential and proprietary information. Accordingly, the Court credits Carson’s testimony that he does not recall much, if anything, about Flatiron’s clinical trial

50 matching business.18 Geoffrey Calkins (“Calkins”), Flatiron’s Senior Vice President of Product Management and Operations, contends that Carson attended a customer meeting in 2018 at which Flatiron described its “vision and product roadmap” for clinical trial matching. (Calkins

Decl. ¶ 34.) Even if Carson were to remember the information presented at that meeting, Flatiron has not sufficiently demonstrated that this information has not become stale in the last seventeen months. And if Carson remembers such information, again recalling the Court’s observation described above, and for the reasons stated, it does not necessarily mean that Carson would reveal it during the remaining months of the Non-Compete. d. External Control Arms External control arms constitute a technique for using real-world patients to augment or replace the control arms

18 Geoffrey Calkins (“Calkins”), Flatiron’s Senior Vice President of Product Management and Operations, testified that Carson had “knowledge of Flatiron’s clinical trial matching capabilities and strategies,” but offered no facts to corroborate this conclusory assertion about knowledge. (See Calkins Decl. ¶ 34.) That Carson “received or had access to” documents regarding the clinical trial matching business or attended one meeting at which someone spoke about that business (see id.) does not sufficiently demonstrate the extent of Carson’s actual knowledge of that information, especially when considering the vast amount of information that would have been received by or accessible to Carson at Flatiron. For these reasons, the Court does not credit Calkins’s testimony on this point. 51 traditionally used in clinical trials. Flatiron uses external control arms in its work with pharmaceutical companies. Although this technique is not unique to Flatiron, Carson testified that he was exposed to trade secrets and confidential information regarding Flatiron’s

external control arms product. (See Tr. 27:7-15.) Regardless, the Court credits trial evidence indicating that Carson will not be creating external control arms or working with pharmaceutical companies at Tempus while his Non-Compete is in effect. e. Prospective Evidence Generation The Court finds that Carson’s exposure to prospective evidence generation projects at Flatiron was limited, and that Flatiron has not demonstrated that the information to which he had access is not stale. Carson provided feedback to a Flatiron team as it was designing a particular prospective study of imaging in or around February and

March 2019. Specifically, Carson reviewed the analytic plan for the study with the team’s statistician and Ariel Bourla (“Bourla”) and provided certain necessary sign-offs. Bourla is a Flatiron Medical Director with a medical degree and Ph.D., and she, not Carson, was the lead scientist on the project. As Tucker acknowledged: “[Carson] wasn’t the day- 52 to-day research oncologist . . . .” (Tr. 190:24.) In the course of Carson’s limited work on that project, he received confidential information, but the Court credits his testimony that he does not recall that information. Although Carson was aware of Flatiron’s plans to develop

prospective CGDB projects, his information on that subject is now over one year old and there is no evidence that the material is not stale at this time. Tucker identified no specific information known to Carson regarding prospective evidence that postdates March 2019. In any event, the Court is persuaded that Carson’s role at Tempus will not entail prospective evidence generation while his Non-Compete is in effect. 2. Business and Financial Information a. The RWE Business Leadership Team As a member of Flatiron’s RWE Business Leadership Team, Carson received non-public information on a quarterly

basis regarding sales and revenue associated with Flatiron’s RWE business line. His role on this team also gave him insight into some of Flatiron’s confidential strategic plans, including the RWE business line’s strategic plan for the next three years. At times, the team also discussed Flatiron’s strategies for responding to 53 regulatory requests and audits. However, Carson stated that he does not recall any specific pricing or revenue information for any of Flatiron’s customers. (Carson Decl. ¶ 28.) The Court found Carson’s testimony relating to this point credible and credits the evidence.

b. Pricing Statements of work for certain products in Flatiron’s RWE business line required Carson’s sign-off at certain points in time. Statements of work are proposals or contracts between Flatiron and its customers providing for a certain type of work to be done at a particular price. For example, as of July 2017, Carson’s signoff was required for statements of work on certain data sets, including customized and off-the-shelf data sets and CGDB. (Tr. 29:10-30:7.) Beginning in early 2018, however, reviewing statements of work became a responsibility shared by a number of research oncologists, and Carson no longer

reviewed all statements of work. Carson conceded that the exact prices Flatiron charges to customers are confidential. Carson, however, will not be using such pricing information in his role at Tempus because he will not be involved in marketing or selling any Tempus products that compete with Flatiron products or participating in any 54 strategic business decisions while the Non-Compete is in effect. 3. Customer Information Carson testified that he obtained confidential information from Flatiron’s biopharmaceutical clients

regarding the specific research questions they were seeking to answer and projects they were interested in doing in the future using Flatiron’s products. (Tr. 30:8-31:10, 143:5- 12.) Because Carson will not be working with pharmaceutical companies or participating in any strategic business decisions at Tempus for the duration of his Non-Compete, this information would not be relevant to his role at Tempus. II. CONCLUSIONS OF LAW A. ANTICIPATORY REPUDIATION

Flatiron claims that Carson anticipatorily repudiated his employment agreement by accepting a position at Tempus. Anticipatory repudiation of a contract occurs when a party to that contract makes a “positive and unequivocal” expression of intent not to perform. Princes Point LLC v.

55 Muss Dev. LLC, 87 N.E.3d 121, 124 (N.Y. 2017).19 The expression may be either a statement by the obligor to the obligee indicating that he will breach the contract or a voluntary act by the obligor that renders him “unable or apparently unable to perform” his obligations under the

contract. Id. at 124. For reasons discussed below, the Court concludes that the Non-Compete is unenforceable as Flatiron seeks to apply it to Carson in the circumstances presented by this case and, accordingly, that Carson has not anticipatorily repudiated the Non-Compete. 1. Enforceability of the Agreement Because restrictive covenants in employment agreements may restrain competition, “impinge on individual agency,”

and restrict “an employee’s ability to make a living,” New York courts subject such covenants to heightened judicial scrutiny. Oliver Wyman, Inc. v. Eielson, 282 F. Supp. 3d 684, 693 (S.D.N.Y. 2017) (collecting cases); BDO Seidman v. Hirshberg, 712 N.E.2d 1220, 1223 (N.Y. 1999). Non-compete provisions are enforceable under New York law only if they

19 The parties agree that this dispute is governed by New York law. 56 are (1) reasonable in duration and geographic scope, (2) “necessary to protect the employer’s legitimate interests,” (3) “not harmful to the general public,” and (4) “not unreasonably burdensome to the employee.” BDO Seidman, 712 N.E.2d at 1223 (citations omitted). “A violation of any

prong renders the [non-compete] invalid.” Id. a. Whether the Agreement is Greater than Necessary to Protect Legitimate Interests

A non-compete must be “no greater than required for the protection of [the employer’s] legitimate interests.” Int’l Bus. Machs. Corp. v. Visentin, No. 11-cv-399, 2011 WL 672025, at *21 (S.D.N.Y. Feb. 16, 2011), aff’d, 437 F. App’x 53 (2d Cir. 2011). Cognizable employer interests include the “(1) protection of trade secrets, (2) protection of confidential customer information, (3) protection of the employer’s client base, and (4) protection against irreparable harm where the employee’s services are unique or extraordinary.” Oliver Wyman, 282 F. Supp. 3d at 694 (quotations and citations omitted). Here, the Non-Compete provides that, for one year following an employee’s termination, the employee will not: directly or indirectly, whether as owner, sole proprietor, partner, shareholder, director, member, consultant, agent, founder, co-venture partner or otherwise, (i) do anything to divert or attempt to 57 divert from [Flatiron] any business of any kind, including, without limitation, solicit or interfere with any of [Flatiron]’s customers, clients, members, business partners or suppliers, (ii) solicit, induce, recruit or encourage any person engaged or employed by [Flatiron] to terminate his or her employment or engagement, or (iii) engage, invest or participate in (x) any business that is similar to those which [Flatiron] has created, has under development or are the subject of active planning from time to time during [his] employment by [Flatiron], or (y) a Competing Business. For purposes of this Agreement, a “Competing Business” means the business of providing software products, data analysis, data, clinical trial research services, analytics and electronic medical record systems to hospitals, physicians, community practices, health care centers, and pharmaceutical companies in the oncology industry[.] (Pl.’s Ex. 3 at 5.) The “similar to” clause of the Non-Compete prohibits employees from participating in “any business” that is somehow “similar to those [businesses] which Flatiron has created” or that Flatiron was actively planning while the employee was at Flatiron. (Id.) The clause’s reference to “those [businesses] which Flatiron has created” suggests that the term “business” in this clause refers to business lines or divisions, rather than corporate entities. Resolving ambiguity surrounding the term “business” by adopting the narrower available definition is appropriate given that Flatiron developed the language of the Non- Compete. See 151 W. Assocs. v. Printsiples Fabric Corp., 58 460 N.E.2d 1344, 1345 (N.Y. 1984) (“[A]mbiguities in a contractual instrument will be resolved . . . against the party who prepared or presented it . . . .”). Here, the clause contains no criteria to provide notice of which similarities are and are not relevant. At its extreme, this

clause could prevent Carson from working at any entities, including universities, that analyze health-related data. Accordingly, it operates as a broad and amorphous restraint.20 The Non-Compete also prohibits employees from working for a “Competing Business.” It defines a Competing Business as one that provides specified services to specified customers in the oncology industry. Specifically, a Competing Business is one that provides:

20 As detailed above, the Court finds that (1) Tempus’s clinical laboratory testing business does not compete with any Flatiron business, (2) the product Tempus’s clinical laboratory business provides is not similar to any Flatiron product, (3) Flatiron does not provide patient-specific lab reports or any other patient-specific reports to clinicians, (4) clinicians are not direct customers of Flatiron’s RWE business, and (5) the manner in which Tempus’s lab reports and portal utilize real-world data does not overlap in any meaningful way with the manner in which Flatiron’s RWE business uses real-world data. See supra Sections I(B) and (H). Nonetheless, the language of the Non-Compete does not indicate whether, in light of these distinctions, Carson’s position in Tempus’s clinical laboratory business falls outside the scope of the Non-Compete’s “similar to” clause. 59 software products, data analysis, data, clinical trial research services, analytics and electronic medical record systems to hospitals, physicians, community practices, health care centers, and pharmaceutical companies in the oncology industry.

Carson asks the Court to construe the term “and” within that definition conjunctively, such that only businesses that provide all the listed services to all the listed customers would qualify as Competing Businesses. Such a definition would exclude Tempus, which does not provide electronic medical record systems, and likely many other businesses with which Flatiron purportedly competes. The Court cannot construe the clause in this manner because it would produce an absurd result. See Mastrovincenzo v. City of New York, 435 F.3d 78, 104 (2d Cir. 2006) (“[W]hen interpreting contracts under New York state law . . . absurd results should be avoided.” (quotations and citations omitted)). The Court must interpret the term “and” within that definition disjunctively. Accord United States v. Fisk, 70 U.S. 445 (1865) (“[C]ourts are often compelled to construe ‘or’ as meaning ‘and,’ and again ‘and’ as meaning ‘or.’”). However, under this interpretation, any company that provides any one of the listed services to any one of the listed customers is a Competing Business. For example, the clause would prohibit 60 Flatiron employees from working for companies that provide payroll or human resources “software products” to hospitals. The terms of the Non-Compete are thus facially overbroad, and the “similar to” clause, in particular, is

impermissibly vague. Courts routinely deem unenforceable non-competes that broadly restrain employees from working for companies that sell goods or services similar to those of the employer. See, e.g., GFI Brokers, LLC v. Santana, No. 06-cv-3988, 2008 WL 3166972, at *10 (S.D.N.Y. Aug. 6, 2008) (deeming “impermissibly broad” a non-compete that purported to prevent an employee from working “for a competitor in a non-competitive capacity”); Silipos, Inc. v. Bickel, No. 06-cv-2205, 2006 WL 2265055, at *6 (S.D.N.Y. Aug. 8, 2006) (holding that a non-compete was overbroad insofar as it prevented an employee from working for any firm engaged in “the Business of the Company, expansively

defined” such that the employee would “be barred from the entire industry”); see also, e.g., Crye Precision LLC v. Duro Textiles, LLC, No. 15-cv-1681, 2016 WL 1629343 (S.D.N.Y. Apr. 22, 2016) (holding that non-compete was “impermissibly vague and overbroad” where it prohibited the licensee from making “similar” products but offered “no 61 criteria to provide notice of what [the licensor] considers to be similar”), aff’d, 689 F. App’x 104 (2d Cir. 2017). In Columbia Ribbon & Carbon Manufacturing Co., Inc. v. A-1-A Corp., 369 N.E.2d 4, 6 (N.Y. 1977), the New York Court of Appeals considered a non-compete that prevented a

company’s employees from working for any firm that sold goods similar to those sold by the company. The Court of Appeals concluded that the non-compete did “no more than baldly restrain competition” because it contained no “limitations keyed to uniqueness, trade secrets, confidentiality, or even competitive unfairness.” Id. Flatiron’s Non-Compete similarly lacks limitations keyed to trade secrets, confidentiality, and competitive unfairness and thus impermissibly “prohibits competition in areas where [Flatiron] simply has no legitimate business interest.” Visentin, 2011 WL 672025, at *21. For example, as Flatiron would apply it, the Non-Compete would prohibit

Carson from working for Tempus in a business line –- here, a clinical laboratory business – that does not compete with any Flatiron business. See id. (holding that agreement was overbroad where it prohibited an employee from “working for a competitor in a business in which IBM does not even participate”). Carson established that his dissemination 62 and implementation project at Tempus is “unrelated to what he did for [Flatiron],” and Flatiron has not shown that Carson “possess[es] information he could misappropriate in [that] area[].” Id.; see GFI Brokers, 2008 WL 3166972, at *10 (holding that a company’s non-compete was unenforceable

where its prohibitions were not limited “to situations where [the employee] would have an opportunity to exploit the ‘information and relationships’ gained from his work at [the company]”). To equip himself to improve the outcomes of cancer patients on a large scale through the use of data analytics, Carson pursued both a medical degree and Ph.D. in Health Policy and Administration. Obtaining those degrees required the investment of substantial time, effort, and money. As the New York Court of Appeals has recognized, “it is not reasonable for a man to be excluded from a profession for which he has been trained when he

does not compete with his former employer by practicing it.” Karpinski v. Ingrasci, 268 N.E.2d 751, 754 (N.Y. 1971). Flatiron also has not shown that the Non-Compete is necessary to protect the company against irreparable harm from losing the “unique or extraordinary” services of an 63 employee. “[T]o demonstrate that a former employee performed unique or extraordinary services, the employer must show that the employee was irreplaceable and that the employee’s departure caused some special harm to the employer.” Pure Power Boot Camp, Inc. v. Warrior Fitness

Boot Camp, LLC, 813 F. Supp. 2d 489, 510 (S.D.N.Y. 2011) (citations omitted); Purchasing Assocs. Inc. v. Weitz, 196 N.E.2d 600, 605 (N.Y. 1960) (similar). While Carson’s qualifications are impressive and evidence in the record indicates that he provided valuable services to Flatiron, the Court finds that Flatiron has not shown that it suffered “special harm” when Carson left, or that other experienced candidates with an M.D. and Ph.D. could not effectively replace him. Based on the above considerations, the Court concludes that the Non-Compete is broader than necessary to protect Flatiron’s legitimate business interests. Accordingly, it

is unenforceable. BDO Seidman, 712 N.E.2d at 1223 (“A violation of any prong renders the covenant [not to compete] invalid.”). b. Duration and Geographic Scope The “durational reasonableness of a non-compete agreement is judged by the length of time for which the 64 employer’s confidential information will be competitively valuable.” Estee Lauder Cos. v. Batra, 430 F. Supp. 2d 158, 180 (S.D.N.Y. 2006). In light of the frequency with which Flatiron amends its data curation rules and the time after which its other confidential information becomes stale, a

non-compete no longer than one year in duration is reasonable. Given the facial overbreadth of the Non-Compete, the Court declines to reach the question of whether the world- wide geographical scope of the provisions is reasonable. c. Whether the Agreement Unreasonably Burdens Carson Even though, in view of its findings above with regard to the breadth prong of the non-compete inquiry, the Court need go no further in this analysis, it finds that the Non- Compete would unreasonably burden Carson because it would prohibit him from working at any company that provides any type of data to any health care provider in the oncology industry. The application of data analytics to improve

patient outcomes in oncology is essentially Carson’s “chosen field of medicine.” Long Island Minimally Invasive Surgery, P.C. v. St. John’s Episcopal Hosp., 164 A.D.3d 575, 577 (N.Y. App. Div. 2d Dep’t 2018), leave denied, 117 N.E.3d 819 (N.Y. 2019). The Non-Compete, as Flatiron 65 purports to apply it, is unreasonably burdensome insofar as it prohibits Carson from performing work within his chosen field. See id. Flatiron’s arguments to the contrary are unconvincing. That Tempus agreed to pay Carson six months’ salary does

not offset the burden imposed on Carson by the one-year Non-Compete. In addition, because oncologists are expected to provide patients with long-term care, it is likely that employers in the oncology industry would be uninterested in hiring an oncologist for a one-year term. Accordingly, to obtain and accept a full-time position as an oncologist, whether as a clinical professor or otherwise, Carson would have to forego the opportunity to work at Tempus. Nor are any investment opportunities or expert witness engagements likely to offset the income and professional advancement Carson would have to forego from Tempus. d. Whether the Agreement is Injurious to the Public Similarly, the Court finds that it would be harmful to

the public to enforce the Non-Compete as Flatiron claims here. Preventing Carson from working at the job where he would be most productive and make the largest impact on oncology care inflicts not only a private cost to Carson but also a social cost. See Harlan M. Blake, Employee 66 Agreements Not to Compete, 73 HARV. L. REV. 625, 682-83 (1960) (“[T]he social cost of preventing an employee from going to a job at which he would be more productive is theoretically equal, given an efficient market, with the economic loss to the individual.”). On that public policy

ground, New York courts disfavor broad non-competes. See BDO Seidman, 712 N.E.2d at 1223. e. Partial Enforceability The New York Court of Appeals has “expressly recognized and applied the judicial power to sever and grant partial enforcement for an overbroad employee restrictive covenant.” Brown & Brown, Inc. v. Johnson, 25 N.E.3d 357, 362 (N.Y. 2015) (quoting BDO Seidman, 712 N.E.2d at 1226). At the same time, the Court of Appeals

recognizes that “[a] legitimate consideration against the exercise of th[e] power [to partially enforce overbroad non-competes] is the fear that employers will use their superior bargaining position to impose unreasonable anti- competitive restrictions, uninhibited by the risk that a court will void the entire agreement, leaving the employee free of any restraint.” BDO Seidman, 712 N.E.2d at 1226. As Professor Harlan M. Blake put it:

67 For every covenant that finds its way to court, there are thousands which exercise an in terrorem effect on employees who respect their contractual obligations and on competitors who fear legal complications if they employ a covenantor . . . Thus, the mobility of untold numbers of employees is restricted by the intimidation of restrictions whose severity no court would sanction. If severance is generally applied, employers can fashion truly ominous covenants with confidence that they will be pared down and enforced when the facts of a particular case are not unreasonable. This smacks of having one’s employee’s cake, and eating it too.

Blake, supra, at 682-83. In light of this serious public policy concern, the Court of Appeals directed courts to determine whether partial enforcement is appropriate by “conduct[ing] a case- specific analysis, focus[ed] on the conduct of the employer in imposing the terms of the agreement.” BDO Seidman, 712 N.E.2d at 1226. Under this approach “partial enforcement may be justified” if: the employer demonstrates an absence of overreaching, coercive use of dominant bargaining power, or other anti-competitive misconduct, but has in good faith sought to protect a legitimate business interest, consistent with reasonable standards of fair dealing . . . .

Id. Applying this test, the Court of Appeals deemed partial enforcement appropriate where “the covenant was not imposed as a condition of [the] defendant’s initial employment, or even his continued employment, but in connection with 68 promotion to a position of responsibility and trust . . . .” Id. Noting that the agreement at issue in BDO Seidman was not facially overbroad, the court reasoned that nothing indicated that the employer “imposed the covenant in bad faith, knowing full well that it was overbroad.” Id.

Applying these principles, as an initial matter, the Court observes that the “similar to” and “Competing Business” prongs of the Non-Compete cannot simply be severed from the non-solicitation prong. That prong provides that Carson is prohibited from: directly or indirectly, . . . do[ing] anything to divert or attempt to divert from the Company any business of any kind, including, without limitation, solicit or interfere with any of the Company’s customers, clients, members, business partners or suppliers[.]

This non-solicitation clause is also overbroad because, for example, it makes no distinction between those customers with whom Carson developed a relationship by virtue of his work for Flatiron and those with whom he did not. As the Court of Appeals explained in BDO Seidman, an employer does not have a legitimate interest in the preservation of its “entire client base where, as here, there is no evidence that the employee obtained a competitive advantage by using confidential information.” Id. at 1224. Rather, an 69 employer’s legitimate interest is limited to the protection of only those “relationships the employee acquired in the course of employment.” Id. Flatiron has not demonstrated that Carson acquired relationships with all of its customers during the course of his employment. Accordingly,

the non-solicitation clause is “greater than is needed to protect [Flatiron’s] legitimate interests” insofar as it prohibits Carson from soliciting Flatiron customers with whom he did not develop a relationship through “assignments to perform direct, substantive . . . services” on behalf of Flatiron. Id. at 1225. Ultimately, in light of the considerations set forth in BDO Seidman, the Court concludes that Flatiron has not demonstrated its entitlement to partial enforcement of the Non-Compete. Flatiron imposed the Non-Compete as a condition of Carson’s initial employment. In addition, the reasons for which the Non-Compete is overbroad are obvious,

calling into question Flatiron’s good faith in proposing the Non-Compete. No drafter could reasonably overlook that the language of the “similar to” prong is impermissibly vague and overbroad absent a list of clear and specific criteria by which similarity may be assessed. Likewise, the definition of a “Competing Business” plainly encompasses 70 entities with which Flatiron does not compete. That Flatiron generally “required all employees, regardless of position” to agree to the same anti-competitive covenants further suggests that Flatiron did not, in good faith, ask Carson, or any other particular employee, to bear only

those restrictions Flatiron deemed “necessary to protect its legitimate business interests.” Brown & Brown, 158 A.D.3d at 1150. That Flatiron’s Co-Founder and Chief Executive Officer believed the Non-Compete would bar Carson from working as a janitor at Tempus (see Turner Dep. 130:6-14) confirms the Court’s conclusion that the Non- Compete’s overbreadth is so obvious that Flatiron could not, in good faith, require almost all employees to agree to its terms. New York courts have declined to partially enforce overbroad non-competes under similar circumstances. See Brown & Brown, 158 A.D.3d at 1149-50 (declining to

partially enforce non-compete that was a condition of initial employment, imposed on all employees, and clearly overbroad under New York law); Long Island Minimally Invasive Surgery, 164 A.D.3d at 578 (declining to partially

71 enforce non-compete that was a condition of initial employment and “clearly overbroad”).21 While Carson has argued that the Non-Compete in the Covenants Agreement is unenforceable, he has not asked the Court to invalidate the entire Covenants Agreement or

refuse to enforce its non-disclosure provision. Accordingly, the Court will not decline to enforce the non- disclosure provision. 2. Conclusion The Court’s conclusion that the Non-Compete is unenforceable necessarily leads to a further conclusion that Carson has not anticipatorily breached the Non- Compete. The Court determines, however, that the non- disclosure provision remains enforceable.

B. PERMANENT INJUNCTIVE RELIEF To obtain a permanent injunction, a plaintiff must show that:

21 Even if partial enforcement of the Non-Compete were appropriate, the Court would not find that Carson anticipatorily breached the Covenants Agreement. Under a narrowed interpretation, the Non-Compete would prohibit only “associations by [Carson] that threaten [Flatiron’s] legitimate interests against unfair competition from [Carson’s] new employer.” GFI Brokers, 2008 WL 3166972, at *10. That Carson’s association with Tempus does not threaten Flatiron’s legitimate interests against unfair competition follows directly from the Court’s findings of fact regarding the scope of Carson’s role at Tempus. 72 (1) it has suffered an irreparable injury; (2) that remedies available at law, such as monetary damages, are inadequate to compensate for that injury; (3) that, considering the balance of hardships between the plaintiff and defendant, a remedy in equity is warranted; and (4) that the public interest would not be disserved by a permanent injunction.

eBay v. MercExchange, L.L.C., 547 U.S. 388, 391 (2006). 1. Irreparable Injury To demonstrate irreparable harm, Flatiron must establish either that Carson “actually misappropriated [Flatiron’s] trade secrets” or that Carson’s “new position [at Tempus] will inevitably require disclosure of Flatiron’s trade secrets or confidential information.” Visentin, 2011 WL 672025, at *16. Flatiron does not contend that Carson has actually misappropriated its trade secrets or confidential information. Accordingly, it must prove that Carson will inevitably disclose Flatiron’s trade secrets or confidential information. See id. To determine whether Flatiron has established that Carson will inevitably disclose trade secrets, the Court must first consider whether Carson “possesses trade secrets” and then evaluate considerations “bearing on whether disclosure is inevitable.” W. Publ’g Corp. v. Coiteux, No. 16-cv-6825, 2017 WL 4339486, at *3 (S.D.N.Y. Aug. 28, 2017). 73 a. Claimed Trade Secrets Under New York law, a trade secret is “any formula, pattern, device, or compilation of information which is used in one’s business, and which gives him an opportunity to obtain an advantage over competitors who do not know or

use it.” Visentin, 2011 WL 672025, at *8 (quotations and citations omitted). Courts consider the following factors when determining whether certain information constitutes a trade secret: 1) the extent to which the information is known outside of the business; (2) the extent to which it is known by employees and others involved in the business; (3) the extent of measures taken by the business to guard the secrecy of the information; (4) the value of the information to the business and its competitors; (5) the amount of effort or money expended by the business in developing the information; (6) the ease or difficulty with which the information could be properly acquired or duplicated by others.

Id.

Flatiron contends that Carson possesses knowledge of three categories of purported trade secret information: (1) methods and technical information, (2) business strategy, partnership, business development opportunity, and financial information, and (3) client confidential information. 74 i. Methods and Technical Information Flatiron has shown that Carson has knowledge of certain information regarding Flatiron’s methods and technical information. Specifically, the Court is persuaded that Carson knows about certain capacities and limitations

of Flatiron’s off-the-shelf and customized data sets, including the CGDB. See supra Section I(J)(1)(a). That is, Carson could recall some variables that are and are not available in those data sets. See id. Flatiron has also persuasively demonstrated that Carson possesses confidential information about how Flatiron resolved certain challenging data curation issues, such as disease progression, response, mortality, and line of therapy. See id. In addition, the Court finds that Carson has knowledge of information regarding Flatiron’s external control arms product. See id. Flatiron has established that the preceding types of

information qualify for trade secret protection. In fact, Carson himself acknowledged that such information is confidential and proprietary. (See Tr. 31:11-32:2 (confirming that Carson had access to confidential and proprietary information concerning Flatiron’s data sets, including the variables in those data sets); 33:11-12 75 (confirming that Carson had access to confidential and proprietary information regarding Flatiron’s data curation rules and data aggregation methods).) Indeed, as Calkins stated, Flatiron employs a variety of measures to prevent disclosure of this information. (See Calkins Decl. ¶¶ 13-

28.) For example, a committee reviews all of Flatiron’s proposed publications to assess whether they would disclose confidential or proprietary information. (Flatiron has invested substantial effort in developing rules and methods for data curation and in building its databases. As these databases and rules add value to products that can be sold to pharmaceutical companies, information about Flatiron’s databases and rules would be of value to competitors. Based on the rate at which Flatiron updates its data curation rules, the Court is not persuaded that all of this information is stale. Flatiron has not established Carson’s knowledge of

trade secret information regarding potentially patentable technologies, prospective evidence generation, and clinical trial matching. With regard to potentially patentable technologies, Flatiron has not carried its burden of demonstrating that any information to which Carson had access in 2016 and early 2017 would still possess 76 competitive value. See supra Section I(J)(1)(b); Am. Airlines, Inc. v. Imhof, 620 F. Supp. 2d 574, 583 (S.D.N.Y. 2009) (reasoning that stale information is not protectable). Likewise, Flatiron has not identified specific information known to Carson about prospective

evidence generation that postdates March 2019, and the Court is not persuaded that Flatiron has sufficiently shown that information pre-dating March 2019 is still of competitive value. See supra Section I(J)(1)(e); Visentin, 2011 WL 672025, at *10 (requiring specificity with regard to information known to the defendant). In addition, the Court is not persuaded that Flatiron has established that Carson recalls any information regarding Flatiron’s Clinical Trial Matching services, or that any information to which he had access in 2017 or 2018 remains of competitive value. See supra Section I(J)(1)(c). ii. Business Information

With regard to business information, Flatiron has demonstrated that Carson has knowledge of information about Flatiron’s strategic plans for its RWE business line and how Flatiron prices the RWE business’s products and services. See supra Section I(J)(2). Strategic business information, as distinct from knowledge of the intricacies 77 of a business and its financials, is protectable as a trade secret. Compare Estee Lauder, 430 F. Supp. 2d at 175-176 (holding knowledge of “brand strategies” and future product plans was trade secret information), with Silipos, 2006 WL 2265055, at *5 (explaining that “knowledge of the

intricacies of a business operation” does not qualify for trade secret protection). Similarly, although “pricing data is generally not entitled to trade secret protection,” see Silipos, 2006 WL 2265055, at *5, strategic information about a company’s pricing structure is protectable as a trade secret. See also Estee Lauder, 430 F. Supp. 2d at 175. The Court assumes, without deciding, that the pricing information of which Carson had knowledge is protectable as a trade secret. iii. Customer Information Flatiron has demonstrated that, while at Flatiron, Carson obtained knowledge of specific research questions

that Flatiron’s pharmaceutical clients were seeking to answer or hoping to explore. See supra Section I(J)(3). The Court assumes without deciding that this information is entitled to trade secret protection.

78 b. Inevitable Disclosure To determine whether an employee working in a new position will inevitably disclose his former employer’s trade secrets, New York courts consider: (1) the extent to which the new employer is a direct competitor of the former employer; (2) whether the employee’s new position is nearly identical to his old one, such that he could not reasonably be expected to fulfill his new job responsibilities without utilizing the trade secrets of his former employer; (3) the extent to which the trade secrets at issue would be valuable to the new employer; and (4) the nature of the industry and its trade secrets. Int’l Bus. Machs. Corp. v. Papermaster, No. 08-cv-9078, 2008 WL 4974508, at *7 (S.D.N.Y. Nov. 21, 2008) (quotations and citations omitted). Flatiron need not prove that inevitable disclosure is certain. Rather, the “likely inevitability of even inadvertent disclosure is sufficient to establish a real risk of irreparable harm . . . .” Id. at *10 (compiling cases). The Court concludes that Flatiron has not carried its burden of showing that Carson cannot “reasonably be expected to fulfill his new job responsibilities without utilizing the trade secrets of his former employer.” Visentin, 2011 WL 672025, at *16. First, Carson will be working in Tempus’s clinical laboratory business, which does not compete with any of Flatiron’s business services. 79 See supra Section I(F)(1). As described above, Tempus’s clinical laboratory conducts lab tests, including genomic sequencing, and provides patient-specific lab reports that present a patient’s test results along with information to help the physician understand and act upon the results. The

Court finds that Flatiron does not own or operate a similar clinical laboratory and does not provide patient-specific lab reports, or any other patient-specific reports, to inform doctors’ individual treatment decisions. In fact, Flatiron’s RWE business does not directly sell data to clinicians. Second, with the assistance of legal counsel, Carson and Tempus proactively narrowed Carson’s prospective role at Tempus so as to avoid any meaningful overlap with Carson’s prior role at Flatiron for the duration of Carson’s Non-Compete and thus ensure Carson’s compliance with his Covenants Agreement with Flatiron. Such an

arrangement is not novel or inherently suspect. See Visentin, 2011 WL 672025, at *16 (holding that employee would not inevitably disclose trade secrets where “[r]ecognizing that there was a potential risk with regard to [the employee’s] prior [role at a competitor], [the employee] and [his new employer] agreed to limit the scope 80 of [his] responsibilities for the first twelve months of his employment”). The Court has found that Carson’s role at Tempus will be limited in the manner described by Carson, Fukushima, and Lefkofsky. See supra Section I(H). In this role, Carson

will not participate in: curating data from electronic health records, analyzing such data to draw population- level inferences or support pharmaceutical companies, working with pharmaceutical companies, or making strategic business decisions. See id. His access to Tempus’s databases will also be restricted. See supra Section I(I). And he would be barred from discussing with any Tempus employee or clients any knowledge about Flatiron’s confidential information or trade secrets. He would also be prohibited from applying such knowledge in his role at Tempus. So limited, Carson’s role does not present a risk that he will inevitably disclose Flatiron’s trade secrets.

Flatiron’s various arguments why Carson’s role is nonetheless “nearly identical” to his role at Flatiron are not supported by the evidence and so are not persuasive. First, Flatiron contends that Carson will inevitably draw upon Flatiron’s proprietary technical and methodological information as he generates the insights from real-world 81 data that are presented in Tempus’s lab reports. This argument, however, ignores evidence in the record demonstrating that Carson will not participate in curating, structuring, or analyzing the data that goes into Tempus’s lab reports. See supra Section I(H). The record makes clear

that Carson will be merely an “end user” of the data in the Tempus lab reports. (Tr. 112:22-23.) Rather than producing the data that goes into the lab report, Carson will be analyzing how physicians utilize the existing data in the lab reports. See supra Section I(H). In conducting such dissemination and implementation research, Carson will be drawing on the basic statistical methods that he mastered as a Ph.D. student and not necessarily applying knowledge he gained while working at Flatiron. Flatiron’s technical and methodological trade secrets will not be of any value to Carson (or Tempus) as he conducts such dissemination and implementation research.

Second, Flatiron has not convincingly demonstrated a likelihood that Carson will be placed in situations where he will unintentionally disclose Flatiron’s trade secrets. First, the record does not lend credence to Flatiron’s speculation, intensely advanced by Weinberg, that Carson will be attending meetings with all of Palmer’s other 82 direct reports, including Axelson, at which the attendees will discuss Tempus’s pharmaceutical business. (See, e.g., Weinberg Decl. ¶ 83.) In fact, under Tempus’s matrix organizational structure, Axelson and Carson would raise substantive issues about their ongoing projects with their

respective “day-to-day” managers, not with Palmer. As Axelson explained, he meets with Palmer monthly to review vacation schedules, upcoming conferences, travel expenses, and a few updates about Tempus. (Axelson Dep. 22:12-23:2.) When Axelson needs to discuss substantive projects, he goes to his “day-to-day” supervisor, Ryan Fukushima, rather than to Palmer. (Id.) Likewise, Carson will be discussing substantive medical projects not with Palmer but with his “day-to-day” boss, Yasiejko, the Business Head of Clinical Testing. See supra Section I(H). Accordingly, the record does not support Flatiron’s notion that Tempus’s direct reporting lines will place Carson in situations where the

disclosure of Flatiron’s trade secrets is likely. Moreover, if any circumstance arises in which Tempus managers met to discuss matters that would raise potential Non-Compete compliance issues, obvious easy remedies are available to avoid such problems: Carson can absent himself or be recused from such discussions. The Court credits the good 83 faith efforts Tempus and Carson made to put in place restrictions on Carson to prevent such occurrences, whether knowing or inadvertent. See supra Section I(I). Third, Flatiron suggests that because most Tempus employees collaborate across business lines, Carson will

also collaborate across business lines. See, e.g., Weinberg Decl. ¶ 79. This argument is a non-sequitur. In fact, Tempus carefully crafted Carson’s role to ensure that he will be “isolated in a business line that [i]s totally distinct and separate from the one Flatiron [i]s in.” (Lefkofsky Dep. 73:18-74:3.) Fourth, Flatiron asserts that disclosure of Flatiron’s trade secrets is likely because, to help physicians understand Tempus’s lab reports, Carson will need to have conversations with other Tempus employees about how any real-world data presented in the reports was curated, aggregated, and analyzed. (See, e.g., Weinberg Decl. ¶ 82.)

The Court is not persuaded, and finds that the Tempus lab report itself provides all of the information needed to interpret and contextualize any insights from real-world data that it contains. See supra Section I(H). As discussed above, Tempus’s report provides information on population size, p-values, confidence intervals, and associated PubMed 84 citations. (See Pl.’s Ex. 203.) Thus, Carson will not need to look much beyond the four corners of the report.22 Finally, in arguing that it has suffered irreparable injury, Flatiron points to language in the Covenants Agreement providing that a breach or threat to breach the

agreement would cause Flatiron to “suffer immediate and irreparable harm and that monetary damages would be an inadequate remedy.” If this language alone were decisive, then a determination that Carson has breached or attempted to breach the Non-Compete would compel the conclusion that Flatiron had suffered irreparable harm. However, as elaborated above, the Court holds that Carson has not breached or threatened to breach the Covenants Agreement. Thus, even if the Court were to give this clause the “great weight” that Flatiron claims it merits, it makes no difference in this case. Cf. Alpha Cap. Aktiengesellschaft v. Adv. Viral Res. Corp., No. 02-cv-10237, 2003 WL 328302,

at *6 (S.D.N.Y. Feb. 11, 2003).

22 Flatiron’s argument also incorrectly assumes that Carson can obtain information only from other Tempus employees through conversation. As discussed above, Carson could, for example, request written descriptions of all of Tempus’s data curation rules and a dictionary of all the variables in Tempus’s data sets without revealing any of Flatiron’s confidential or trade secret information. See supra Section I(H). 85 2. Conclusion Based on the above discussion and in consideration of all evidence in the record, the Court concludes that Flatiron has not demonstrated a risk that Carson will disclose Flatiron’s trade secrets.

Because Flatiron has not demonstrated that it has suffered irreparable harm from Carson’s acceptance of employment by Tempus, Flatiron is not entitled to an injunction. See eBay, 547 U.S. 388, 391. III. ORDER Accordingly, it is hereby

ORDERED that the request of plaintiff Flatiron Health Inc. (“Flatiron”) for a declaratory judgment, pursuant to 28 U.S.C. Section 2201, that defendant Kenneth Carson (“Carson”) is contractually barred from working for Tempus Labs, Inc. in any capacity and from soliciting Flatiron customers is DENIED; and it is further ORDERED that Flatiron’s request for a declaratory judgment that Carson is contractually bound not to retain, use, or disclose to third parties any of Flatiron’s trade secrets and confidential information is GRANTED; and it is further

86 ORDERED that Flatiron’s request for a permanent injunction is DENIED.

SO ORDERED.

Dated: New York, New York 19 March 2020

_________________________ VICTOR MARRERO U.S.D.J.