ARK’s Expected Value For Exact Sciences In 2027: $140 Per Share
According to ARK’s research, Exact Sciences (EXAS, “Exact”) has been a pioneering force in the growing field of oncology testing and will likely remain so during the next five years. Leveraging technology and experience from its Cologuard and Oncotype DX testing franchises, Exact Sciences is expanding its services across the entire cancer care continuum—from prevention to late-stage disease. After severe disruptions in cancer care during the COVID-19 pandemic, the molecular oncology industry now seems poised to continue delivering sustained secular growth. Exact’s competitive advantages include its commercial scale, brand recognition, deep regulatory experience, payor relationships, industry and research partnerships, clinical evidence, and technology platform. As illustrated in our open-source model, in our base case, shares of Exact Sciences could compound at an average annual rate of 25%, reaching $140 by 2027. We divide the remainder of this paper into three parts:
- Part I summarizes the assumptions, key inputs, and outputs of ARK’s Exact Sciences model.
- Part II details the molecular oncology landscape, including the impact of the COVID-19 pandemic and Exact’s expansion across the cancer burden continuum.
- Part III delves into Exact’s portfolio, including its ongoing work in screening and prevention, prognostics and risk stratification, as well as advanced cancer testing.
PART I: ARK’s EXACT SCIENCES MODEL
Model Assumptions and Key Inputs
In our model, cancer prevention and screening services drive the bulk of Exact Sciences’ valuation. This line item is composed primarily of Cologuard—Exact’s flagship colorectal cancer (CRC) screening test, as illustrated below. Thanks to improvements in test performance, sample-type compatibility, enabling infrastructure, and national health policy, more than 80% of eligible Americans are likely to be screened for CRC in the future. In our view, Exact will continue to lead this category of testing through 2027. For this revenue segment, we also have modeled Exact’s expansion into hereditary cancer testing (HCT), multi-cancer earlier detection (MCED), and non-CRC single-cancer screening. We expect the MCED market to materialize near the end of the 2020s and Exact to command an enduring leadership position. HCT and non-CRC single-cancer screening also offer the potential for meaningful revenue upside for Exact’s Prevention and Screening business. We summarize our key model inputs and primary revenue drivers below.
Key Prevention and Screening Inputs
|Input Name||Explanation||Base Case Value|
|Long-Term CRC Screening Adherence Rate (%)||The percentage of eligible individuals who adhere to guideline-recommended CRC screening—a long-term asymptote, not the value in 2027.||80%|
|Sales Model Mix (Cologuard)||The percentage of Cologuard volume driven by outbound sales rather than digital integration, rescreenng, and network effects.||60%|
|Takeover Time (Cologuard)||The number of years for Cologuard to move through its middle-adoption phase.||15|
|Long-Term CRC Market Share (%)||Exact’s share of annual CRC screening volume—a long-term asymptote, not the value in 2027. Presently, Exact’s share of this market is ~7.5% and we expect that to reach ~16.5% by the end of 2027.||40%|
|Pre-Reimbursement MCED Market Fraction (%)||The fraction of the MCED total addressable market (TAM) unlockable prior to national guideline inclusion and reimbursement.||10%|
|Onset Year of Rapid-Growth (MCED, Post-Reimbursement)||The year that Exact’s reimbursed MCED test accelerates from its early adoption phase.||2029|
Exemplified by Oncotype DX, Exact’s second main revenue driver is molecular prognostic testing. This suite of tests helps oncologists differentiate between more and less aggressive forms of cancer—mitigating the risks of undertreatment or overtreatment, respectively. For example, some studies have shown Oncotype DX reduces unnecessary chemotherapy for 60% of early-stage breast cancer patients. Oncotype DX should be a durable source of Exact’s revenue growth as it expands into new cancer types and subtypes, new geographies, and potential new sample types like blood.
Capitalizing on its formidable, early-stage oncology salesforce and deep relationships with providers, Exact is likely to gain early traction in minimal residual disease (MRD) testing despite its relatively late market entry. We believe the commercial and strategic synergy between cancer screening and early-stage disease management will be vital to Exact’s growth. We have summarized several of these inputs for Prognostics and MRD in the tables and charts below. Although Exact offers tests for advanced cancer therapy selection like Oncomap, they do not drive our outlook materially.
Key Prognostics and MRD Model Inputs
|Input Name||Explanation||Base Case Value|
|Long-Term Risk-Stratification Market Share (%)||Exact’s share of US prognostic testing volume (breast, colorectal, etc.). The input value is a long-term asymptote, not the value in 2027. Presently, Exact’s share of this market is ~35% and we expect that to reach ~53% by 2027.||70%|
|Tumor-Informed MRD Test Frequency||The average number of MRD surveillance tests that a patient receives per year.||1.5|
|Onset Year of Fast Rapid Growth (Informed MRD)||The year that Exacts’ tumor-informed MRD test accelerates out from its early adoption phase.||2027|
|Naïve MRD Market Share (%)||Exact’s share of tumor-naïve MRD testing volume. The input value is a long-term asymptote, not the value in 2027.||10%|
Despite the speed of technological change in the molecular oncology testing industry, we view Exact Sciences as a sound investment opportunity. Led by a management team with demonstrated foresight, Exact has been nimble in a growing market. In addition to our revenue and share price forecasts, the table below shows other key performance indicators (KPIs).
Key Model Inputs
|Output Name||2016||2021||2027 Base Case||2027 Bull Case||2027 Bear Case|
|Cologuard Revenue ($K)||$99,380||$1,062,000||$2,986,000||$3,206,000||$2,351,000|
|Oncotype DX Revenue ($K)||N/A||$562,000||$1,088,000||$1,142,000||$1,051,000|
|Total Revenue ($K)||$99,380||$1,767,000||$5,276,000||$6,203,000||$4,150,000|
|Exit Growth Rate (%)||N/A||N/A||15%||17%||13%|
|Core EV/EBITDA Multiple*||N/A||N/A||20||24||18|
|Market Cap ($M)||$1,500||$13,900||$32,674||$46,674||$17,248|
|Share Price/Target ($)||N/A||$45||$139||$208||$74|
We use a sum-of-parts valuation for Exact Sciences that includes its core business—Cologuard and Oncotype DX—and its burgeoning pipeline of molecular oncology tests. To capture the profitability of Exact’s core business, we apply a mature EV/EBITDA ratio to this part of the business and use an EV/Sales multiple (6x) for Exact’s more nascent tests. In practice, Exact is likely to reinvest cash from its core business into its suite of pipeline tests, remaining cashflow neutral for several years. We believe this is prudent given Exact’s leading position in several large, untapped markets. In Part II of this article, we will discuss the molecular oncology landscape before, during, and after the COVID-19 pandemic, then dive into Exact Sciences specifically.
PART II: THE MOLECULAR ONCOLOGY LANDSCAPE
Cancer has created a significant unmet need in the healthcare industry. New cancer diagnoses have risen steadily over the past two decades. Cancer doesn’t care about interest rate fluctuations or changes in consumer preferences. As a result, the demand for more powerful diagnostics and therapeutics is growing secularly, as shown below.
Supercharged by affordable next-generation sequencing (NGS), molecular methods to screen, diagnose, risk-stratify, and monitor cancer have translated into clinical practice over the past decade. The tests measure biomolecules like DNA, RNA, and proteins to understand the unique disease of each patient, which is the basis of precision medicine—a dream inspired by the Human Genome Project in the nineties that has turned into reality.
Supplemented by histopathology, tumor tissue-based molecular testing has become the standard of care for some cancer subtypes. Exact Sciences’ Oncotype DX assay, for example, guides treatment for roughly 70% of American patients with HR-positive, HER2-negative breast cancer. That said, tissue-based tests have drawbacks. They require high-quality tumor tissue samples, the extraction of which is invasive and often requires hospital admission, sometimes preventing the procedure for patients who lack sufficient tumor tissue or are immunocompromised. Moreover, NGS workflows destroy tissue, making these types of tests one and done. Thankfully, the rapid pace of innovation in life sciences is beginning to overcome these obstacles.
While not a replacement for tissue-based testing, “liquid biopsies” have burst onto the clinical scene. In fact, blood-based, liquid biopsies are becoming a minimally invasive complement to tissue testing. Thanks to mobile blood-drawing services, they even can be done at home and repeated to monitor a tumor’s response to treatment over time.
Liquid biopsy is not a new concept. For decades, pathologists have relied on methods like flow cytometry to monitor liquid cancers like multiple myeloma. Compared to liquid cancers, solid tumors shed faint signals into the bloodstream that can require deep and expensive sequencing. Thanks to novel sample preparation methods, machine learning (ML) and, importantly, lower DNA sequencing costs, liquid biopsy for solid tumors has become more cost-effective, as illustrated below. Because solid tumors account for more than 90% of annual cancer diagnoses, these advances have grown the liquid biopsy market considerably.
Despite exceptional publication growth and broader clinical reimbursement, comprehensive molecular profiling of advanced cancers still accounts for less than 15% of the market. Earlier cancer detection and cancer monitoring, on which liquid biopsy will have profound impacts, are less than 1% of the market. ARK believes the unmet need and addressable markets are large enough to support several major players over the next few decades.
The Impact of the COVID-19 Pandemic on Oncology
The COVID-19 pandemic disrupted cancer and other healthcare practices severely, causing broad-based interruptions in care. Early in the pandemic, clinicians at the Cleveland Medical Center reported a significant (~5%) increase in oncology appointment cancellations. Researchers have estimated that patients missed nearly 30 million cancer screening visits during COVID-19. A more recent systematic review of 62 studies found that the pandemic interrupted 26% of cancer treatments, as supply chain bottlenecks deprived 79% of medical centers of anti-cancer medication and imaging equipment, and as serious staffing issues impacted 60% of hospitals. Hospitalization rates for cancer patients dropped 30% throughout the pandemic while clinical trials declined 65% at the height of the pandemic.
Though global cancer care has not rebounded completely, pandemic-related disruptions are diminishing, with clinical trials leading the way. According to a recent report by IQVIA, between 2016 and 2021 the number of oncology clinical trial starts increased 56%, with nearly half the growth in 2021 alone.
Apart from clinical trials, the recovery has lagged. Worldwide, oncologists are reporting caseloads ~25% below pre-pandemic levels, though they are much improved relative to the >60% reductions during the height of the pandemic. Although colonoscopies and mammograms have recovered to pre-pandemic levels, the overall cancer screening rate is down ~10%. That said, oncology writ large could be in a stronger position moving forward than otherwise would have been the case.
With tissue biopsies almost impossible during the pandemic, many oncologists defaulted to liquid biopsies. An Italian cancer center noted that liquid biopsy helped “reduce the danger of viral spread by avoiding invasive surgical procedures” during treatments of non-small cell lung cancer (NSCLC). In fact, the increased utilization created “concordance” datasets demonstrating that liquid biopsies can identify cancer mutations at nearly the same rate as tissue biopsies. In 2021, the American Society of Clinical Oncology (ASCO) meeting showcased more than 80 liquid biopsy studies completed during the pandemic. Interestingly, the evidence seems clear that liquid and tissue share >90% of mutations, each sample type carrying a unique mutational signature. In other words, comprehensive diagnostic strategies are likely to involve both tissue and liquid biopsies whenever possible.
Clinical evidence governs medical guidelines that, in turn, influence reimbursement decisions. During the pandemic, the US Food & Drug Administration (FDA) approved the first two NGS-based liquid biopsies as companion diagnostics for the management of late-stage lung cancer: Guardant360 CDx and FoundationOne Liquid CDx. Under its innovative MolDx framework, Medicare contractor Palmetto GBA has granted favorable local coverage determinations for several liquid biopsy tests, portending broader reimbursement. These policy decisions have made it easier for both patients and physicians to adopt liquid biopsy.
Despite the recovery from the COVID-19 pandemic and progress in genome-informed cancer care, the market does not seem to be optimistic about the prospects for molecular testing companies, not to mention growth companies. Indeed, based on the high cost of capital, many growth-oriented companies have cut back on aggressive spending plans. Several cancer testing companies, however, have maintained or raised annual revenue targets as growth rates have reaccelerated from COVID-19 lows. That said, the rapid emergence of COVID-19 variants like Delta and Omicron has created atypical year-over-year revenue growth comparisons, as shown below, complicating the analysis of testing companies’ growth trajectories. These data show that the pandemic interrupted steady growth in the uptake of molecular diagnostic tests.
Molecular oncology is a vast and underpenetrated market. After decades of declining costs, molecular testing is becoming more powerful, accessible, and cost-effective than traditional methods in some cases. Because they surface actionable biomarkers, molecular diagnostics enable the delivery of targeted therapies to cancer patients. For many reasons we’ll address, the market may not submit to a “winner-takes-most” dynamic, but we do expect a decade-long period of sustained industry consolidation. In our view, Exact Sciences will be one of the consolidators.
Mapping Molecular Oncology
Historically, many diagnostics companies have focused on commercializing single flagship tests. Spending years on the R&D necessary to design and validate a diagnostic assay, a company combines its clinical trial results with real-world evidence (RWE), strengthening the case for the test’s clinical value. As local coverage expands to national Medicare coverage and commercial payors follow suit, a test’s average selling price (ASP) tends to rise, adding to the operating leverage associated with lower cost-of-goods (COGS) and economies of scale. Yet, while successful in the testing space today, molecular oncology pioneers are broadening their aspirations.
Leading molecular testing companies are migrating away from the single franchise model toward full-service cancer care models. While technology can and does differentiate companies, we believe durable competitive advantage will require market entrenchment, mindshare dominance, service, and support. With solutions across the care continuum, diagnostics vendors can serve patients and physicians throughout the cancer treatment cycle, as shown below. This model creates higher attach rates between tests, lowers customer acquisition costs, and improves operating leverage with lower sales and marketing costs. Based on similar workflows, capital equipment, and technologies, many oncology tests have the potential for significant R&D leverage. Those companies with broad product portfolios and simple, single front-end ordering systems are likely to become entrenched structurally in the healthcare system.
Turbocharging this strategy with its acquisition of Genomic Health in 2019, Exact Sciences has built, bought, and integrated technologies to expand its potential in the oncology testing market. Combining its own Cologuard screening test with Genomic Health’s Oncotype DX—one of the largest prognostic testing franchises in the world—Exact has a substantial presence at the front end of the diagnostic funnel.
Exact’s Expansion Across the Cancer Burden Continuum
Approved by the US Food and Drug Administration (FDA) in late 2014, Cologuard—Exact Sciences’ flagship colorectal cancer (CRC) screening test—had screened more than two million people by late 2018. Since then, despite the COVID-19 outbreak, Exact has more than quintupled its cumulative CRC screening volume, testing ten million people on behalf of more than 300,000 unique healthcare providers in the US alone.
Across our strategies, ARK minimizes exposure to companies dependent on single assets, so called binary stocks. Acquiring fellow molecular oncology pioneer Genomic Health (GHDX) in 2019, Exact diversified its business importantly. Historically, Genomic Health had focused on cancer prognostics, otherwise known as risk-stratification. Oncologists use prognostic tests to determine the aggression of a tumor, aiming to reduce the risk of over- and under-treatment. As mentioned, Genomic Health’s—now Exact’s—Oncotype DX assay is a standard of care to guide treatment decisions in early-stage breast cancer. According to our research, since its inception, Oncotype DX has supported over 900 clinical trials.
Although some investors viewed Exact’s acquisition of Genomic Health as a departure from its focus on CRC screening, we interpreted the deal more positively. As we have written, cancer screening does not exist in a vacuum. The performance and health economics of a screening program depend on the entire cancer care pathway, including prognostic testing. A prognostic test can help determine if a tumor is indolent, or slow growing, lowering the probability of over-treatment with chemotherapy and/or radiation.
In our view, Exact’s expansion into prognostics was prescient. Since 2019, Exact has amalgamated technologies to enable hereditary disease testing, therapy selection, treatment response monitoring, and blood-based multi-cancer screening. As Exact broadens its testing portfolio, the company should benefit from its brand recognition, global sales infrastructure, digital channel, and access to patient samples for machine learning. Perhaps most important is Exact’s capacity to integrate disparate nodes of cancer care into a unified, longitudinal testing service, which should continue entrenching the company in healthcare systems around the globe.
PART III: EXACT’S ONCOLOGY PORTFOLIO
Cancer Screening and Prevention
Thanks to the control of infectious diseases, remarkable improvements in public health occurred during the first half of the twentieth century. During the past 50 years, progress in preventive medicine has been astounding, as shown below. The mortality rate associated with several cardiovascular diseases, for example, has declined nearly 60%, with disease prevention accounting for the majority (~75%) of the improvement. During the next 50 years, cancer mortality is likely to decline dramatically thanks to disease prevention and early detection.
Primary prevention predicts and prevents disease, while secondary prevention detects disease early in its progression. Cancer is a progressive disease that is much easier to treat if diagnosed early. Exact Sciences has commercial tests for both the primary and secondary prevention of cancer, as shown in the chart below.
A = Some people, especially those outside cancer screening guidelines, may opt for HCT first. For Exact, the number likely is low given the relative immaturity of its HCT offering.
B = Given national reimbursement coverage and medical guidelines, most patients begin with a secondary preventive screening test like Cologuard.
C = This pathway is bidirectional. Patients testing positive with HCT are likely to follow up with a screening test. Meanwhile, a percentage of screened patients is likely to undergo HCT as well, which will help practitioners personalize future screening and flag hereditary cancer risk in families.
D = Individuals with a negative test result repeat screening based on the medically recommended interval, typically ~3 years.
E = Individuals testing positive may move downstream to Exact’s prognostic and disease monitoring services.
Exact’s Work in Primary Prevention: Before Cancer Happens
With its acquisition of PreventionGenetics (“Prevention”), Exact Sciences gained access to broad expertise in germline sequencing and hereditary cancer testing (HCT). Clinicians can use HCT results to focus secondary screening strategies on those with higher hereditary risk. These test results can inform the type of screening, the age at which it should begin, its frequency, and other proactive risk-reduction strategies. While it has broad-based clinical utility, hereditary cancer testing also is an effective preventive tool in the primary care setting.
Prevention generated roughly $36 million in revenue, 2% of Exact’s total revenue, at a profit during 2021. Although it currently offers non-HCT services, Exact is likely to narrow Prevention’s menu toward oncology testing. Though it lags that of more established competitors like Invitae (NVTA) and Myriad Genetics (MYGN), Prevention’s HCT market share is likely to increase as Exact cross-sells HCT throughout its industry-leading commercial channels in primary care and oncology and integrates it into a unique EPIC electronic health record (EHR) implementation. In other words, we believe Exact Sciences will increase its competitive advantage as a full-service solution for healthcare providers by mitigating the friction associated with HCT ordering and clinical reporting.
Given its nascency and the lack of broad-based reimbursement for hereditary tests, we assume that the average selling price (ASP) of Exact’s HCT portfolio is a modest $500 and that it will increase 3% per year. If its attach rate were to hit just 5% of its core testing franchises, Cologuard and Oncotype DX, Exact’s HCT revenues could increase ~40% at an average annual rate during the next five years, from ~$36 million in 2021 to ~$205 million by 2027.
Exact’s Work in Secondary Prevention: As Soon as Cancer Happens
Single-Cancer Earlier Detection (SCED)
Cologuard, Oncoguard, and Exact’s emerging multi-cancer earlier detection (MCED) tests are forms of secondary cancer prevention. Launched in 2014, the stool-based version of Cologuard has been ordered by most primary care doctors in the United States. The franchise has strong brand recognition and physician goodwill that the company can leverage as it expands its cancer screening business. Thanks to technological improvements, product expansions, and several industry- and company-specific tailwinds, we believe Exact will grow its Cologuard business >20% per year during the next five years.
The screening population and adherence to screening guidelines are key to forecasting the size of the CRC screening market. In 2021, the US Preventive Services Task Force (USPSTF) expanded the recommended ages for CRC screening to 45–75 (from 50–75), expanding the screening population by roughly 20% to approximately 115 million Americans.
A recent systematic review of nearly 100,000 Medicare beneficiaries showed that the CRC-screening adherence rate increased from 50% in 2011 to nearly 70% in 2019. From the universe of all CRC screens in the study, fecal occult blood testing’s (FOBT’s) share decreased from 17.7% to 6.6% while Cologuard’s testing share increased from 0% to 14.2%. Although colonoscopy is the dominant form of CRC screening today, Cologuard is its complement—not a competitor—offering patients a less-invasive, low-cost test that has been important to the increased adherence to CRC screening guidelines. For that reason, Cologuard has taken share from other non-invasive screens during the past decade.
Moreover, the US Centers for Medicare and Medicaid Services (CMS) recently proposed changes to its 2023 Physician Fee Schedule that will reimburse any colonoscopy conducted in the US after a stool-based screening test, eliminating a major barrier to widespread Cologuard adoption. Previously, the confirmatory colonoscopy wouldn’t be covered.
In our model, the CRC screening adherence rate increases from 70% but plateaus at ~85% in the distant future—an improvement that increased preventive service adoption in the primary care setting could secure. Already, the percent of US outpatient preventive care visits increased from 30% in 2008 to 44% in 2018, as shown below.
Despite this meaningful progress, approximately 44 million average-risk adults in the US have yet to be screened for colorectal cancer. Both the private and the public sectors are working to improve adherence with a proliferation of technology-enabled, healthcare platforms like those at Signify Health [SGFY, now CVS Health (CVS)] that offer new strategies to improve CRC screening adherence, particularly in underserved demographics. We believe innovative primary care offerings from companies like One Medical (ONEM, now Amazon (AMZN)), Accolade (ACCD), Teladoc (TDOC), and many others are likely to extend the reach of preventive primary care in both Medicare and commercially insured populations. The Biden administration’s recent reinvigoration of the Cancer Moonshot initiative, along with the sustained health policy efforts of patient advocacy groups like the National Colorectal Cancer Roundtable (NCCRT), should continue to boost cancer screening. New competitive entrants like Guardant Health (GH) and Freenome (Private) also should help raise awareness and expand the cancer screening market.
In the context of industrywide screening tailwinds, Exact continues to improve and broaden the Cologuard franchise. In 2021 at the American Society of Clinical Oncology (ASCO) meeting, Exact Sciences shared preliminary data from a case-control study of its second generation, stool-based Cologuard test (“Cologuard 2.0”). Compared to the current version, these data showed improvements in both sensitivity, which increased from 92% to 95%, and specificity, which increased from 87% to 92%. The test’s ability to detect advanced adenoma (AA), a form of precancerous lesion, increased drastically from 42% to 57%. Detecting and then removing AAs can prevent cancer, making this metric crucial to the success of a CRC screening test. Test performance often degrades modestly in a prospective clinical trial. If performance holds relatively steady in the forthcoming pivotal trial, then Cologuard 2.0 should help Exact continue taking share from other less-invasive screening tests on the market. Exact expects pivotal trial results in late 2022 or early 2023.
Recently, Guardant Health published results from its pivotal ECLIPSE trial evaluating the performance of the company’s blood-based CRC screening test, Shield. The assay demonstrated 83% sensitivity at 90% specificity as well as a 13% detection rate for advanced adenoma. Based on these initial results, we think blood-based CRC screening is unlikely to disintermediate Cologuard 2.0, a stool-based test, in the intermediate term but should grow the overall market. According to our research, Guardant and Freenome have the most well-developed and robust blood-based CRC screening tests, though neither company currently has a commercial product yet.
Leveraging samples collected from its 24,000-participant BLUE-C clinical trial as well as a decade of experience with CRC biomarker discovery, Exact Sciences plans to validate the performance of Cologuard Blood, its own blood-based CRC screening test. CMS has set the minimum threshold for reimbursement at 74% sensitivity and 90% specificity for a blood-based CRC test. We anticipate that Cologuard Blood will perform better than the CMS minimum, and perhaps in-line with other CRC tests, and that small differences in test performance will matter less than brand awareness, ordering logistics, and commercial channels. For perspective, 90% of Americans over 45 see at least one Cologuard ad per week. Unlike its competitors, Exact Sciences will be able to process two sample types, blood and stool, all under Cologuard’s banner. We do not differentiate between Cologuard stool and blood revenue in our model.
Patient rescreening could become another adoption tailwind specific to Exact Sciences. Given the FDA-recommended screening interval of three years, Exact estimates that more than one million patients are due for another Cologuard test, an opportunity that should snowball as Cologuard scales, creating a significant recurring revenue stream. To bolster Cologuard adoption, Exact has invested in IT infrastructure and its customer experience since 2014. As rescreening becomes more seamless from the users’ perspective, sales and marketing should decline as a percent of revenue, increasing Exact’s operating margins.
Cologuard Blood will be Exact Sciences’ second blood-based cancer screening test. Once commercialized, it will join Oncoguard Liver, Exact’s blood-based test for the early detection of hepatocellular carcinoma (HCC, “Liver Cancer”). Liver cancer is the fastest-growing cause of cancer-related death in the US, making it a prime candidate for earlier detection. Oncoguard Liver is available today, though only in a patient-pay format. We anticipate that payors will offer reimbursement for Oncoguard Liver by the end of 2023 and that adoption should accelerate afterwards.
Exact could launch other blood-based, single-cancer screening tests by the end of our five-year forecast period. Having partnered with scientists and clinicians at Mayo Clinic since 2009 to identify biomarkers for 15 of the deadliest cancers, and having acquired university spin-out Thrive Earlier Detection (“Thrive”) in 2020, Exact initiated another biomarker discovery collaboration with the Johns Hopkins University School of Medicine. Note that Exact’s non-CRC, single-cancer screening tests fall under the Oncoguard brand name. Given the company’s focus on profitable growth, Oncoguard tests are not a material revenue driver in our base case, but, in our bull case, Exact commercializes at least two Oncoguard tests during the next five years.
Multi-Cancer Earlier Detection (MCED)
As described in ARK’s Big Ideas 2021, multi-cancer earlier detection (MCED) tests may have more potential than any single medical innovation in history to reduce cancer mortality deaths. In our view, after its acquisition of Thrive, Exact is the testing company best positioned to drive MCED into mainstream clinical care and activate one of the largest applied markets in medical genetics.
Unlike single-cancer screening, multi-cancer screening scours the blood for signs of many cancer types simultaneously and is likely to proliferate alongside single cancer testing, not cannibalize it. As discussed in our podcast with company co-founder and esteemed researcher Dr. Bert Vogelstein, Thrive’s IP (CancerSEEK) embodies decades of research at Johns Hopkins and has been evaluated in the largest prospective, interventional MCED trial in history (n=10,000). The Detect-A study showed that Thrive’s standalone legacy test more than doubled the number of cancers detected through screening, 65% of them in early stages.
Since the acquisition, Exact has been integrating Thrive’s DNA mutation, protein, and aneuploidy signatures into its research on methylation. As a reminder, methylation is a type of chemical modification to the outside of DNA that can be highly altered in cancer. In October 2022, Exact released a tranche of case-control data, highlighting the early performance of its unfinished MCED test. Compared to Detect-A, the revised test doubled overall sensitivity to 61%. Because of the different stages of development, study designs, stages and types of cancer, inclusion of diseased controls, and other variables, a comparison of MCED tests would be premature and unproductive at this point. Exact plans to release more case-control data before finalizing the test design ahead of its prospective, pivotal trial launch in 2023. In our view, widespread MCED adoption is unlikely to take place within the 2025-2027 timeframe. Not until the USPSTF publishes its recommendations and CMS announces its coverage decisions do we expect substantial adoption, which could take place in 2028-2029.
In our view, the obstacles to MCED will be clinical implementation and diagnostic follow-up—not technology or cost. Our main concerns are disease localization and overtreatment. After examining these two issues, we explain why Exact Sciences seems to have developed a sound approach to MCED.
Exact’s Work in Disease Localization
Unlike tests that screen for the presence of a specific cancer, a multi-cancer screening test returns positive results if the panel detects any cancer. While some tests like Grail’s Galleri leverage methylation signals to generate a tissue-of-origin (TOO) recommendation, that approach could complicate follow-up—more on this point below. By design, Exact’s MCED test offers no TOO readout but recommends standardizing screen-positive patients on a full-body confirmatory PET-CT scan.
Galleri’s TOO prediction rate is purported to be 89% but our analysis suggests a rate substantially lower. Galleri’s positive predictive value (PPV)—the fraction of positive results that are true positives—is roughly 40%, which means that 60% of the positives are false, but Grail has excluded them from its TOO accuracy calculation. All false positives generate an incorrect TOO prediction because the cancer does not exist. Including false positives in the denominator and applying Grail’s 89% TOO prediction accuracy only to true positives, Galleri seemingly would point a clinician to the correct cancer site less than 50% of the time.
Even highly specific tests in the average-risk, asymptomatic setting will generate many false positives, not because of faulty testing but because of the underlying dynamics of the population. Presently, we favor MCED approaches that take advantage of existing medical infrastructure and familiar care pathways. By coupling PET-CT with its MCED test, for example, Exact is standardizing downstream follow-up around a well-accepted imaging modality that boasts high sensitivity.
Given the glacial pace of evidence generation and organizational review, first-mover advantages in MCED seem unlikely to determine ultimate success. Exact’s structural, competitive advantages include a single, EHR-integrated ordering platform, industry-leading primary care commercial channels, experience and expertise in navigating regulatory and payor pathways, and brand recognition, among others.
Exact’s Efforts to Prevent Overtreatment
Overtreating cancer involves costly, often painful medical procedures on patients with non-aggressive (indolent) tumors, as shown below. In our view, leveraging prognostic testing to differentiate between aggressive and non-aggressive disease is one of the best ways to combat overtreatment, or undertreatment for that matter. As noted elsewhere in this paper, cancer screening does not exist in a vacuum. The steps taken after a positive screening result can matter more than the performance of the screening test itself. Exact Sciences’ commercial prognostic tests like Oncotype DX are among the most clinically validated and utilized in the US for several major cancer types. That said, while approved for breast and colon cancer, Oncotype DX does not apply, yet, to the swath of cancers included in Exact’s MCED panel.
Exact’s Work in Prognostics and Risk Stratification
In this section, we explore Exact’s commercial prognostic testing franchise—Oncotype DX—as well as its pipeline and integration into the company’s portfolio and technology platform, as depicted below. When patients present with sudden symptoms or receive positive screening results, clinicians first work to confirm the diagnosis. Next, oncologists will seek more precise ways to describe the cancer. How aggressive is the disease? Will the patient derive benefit from chemotherapy and/or radiotherapy? Once treated, how likely is the cancer to return? Combined with histopathology, molecular prognostic testing helps answer some of these questions.
A = Following an early-stage cancer diagnosis, most patients entering Exact’s prognostic funnel likely will begin with Oncotype DX. Many may also receive hereditary testing in the short-term and/or MRD testing in the long-term.
B = Some patients who test positive on a screening test will continue cancer treatment with Exact as their diagnostic provider.
C = During and after treatment, patients will receive MRD tests to monitor their disease.
D = Patients in long-term remission may exit surveillance monitoring.
E = Patients with more aggressive cancers take a more comprehensive therapy selection test.
Exact’s Oncotype DX: Predicting Cancer Recurrence and Response
Like Cologuard, Exact Sciences’ Oncotype DX franchise has enjoyed tremendous mindshare, clinical support, and adoption. It is a sturdy, steadily growing, profitable platform upon which Exact can launch pipeline tests under development.
Unlike therapy selection tests that steer patients toward specific drugs, oncologists often use Oncotype DX to help patients avoid costly, painful interventions like chemotherapy. Because of its long-term survival data across specific cancer subtypes, clinicians are more comfortable with its test results recommending de-escalation of treatment. Genomic Health’s TAILORx study (n=10,273), for example, the largest adjuvant breast cancer trial to date, helped establish Oncotype DX as a standard of care for node-negative, early-stage breast cancer. Today, nearly 100% of Americans have access to Oncotype DX as a covered benefit, as shown below.
Despite Oncotype DX’s success, Exact Sciences has not rested on its laurels. The recent RxPonder study (n=5,000) quantified the benefit of chemotherapy for node-positive, early-stage breast cancer. Based on these data, ASCO gave Oncotype DX the strongest overall recommendation for guiding both node-positive and node-negative early-stage breast cancer treatment. Moreover, Exact has broadened Oncotype DX to include ductal carcinoma in situ (DCIS)—another breast cancer subtype accounting for ~20% of new diagnoses. Beyond breast cancer, Oncotype DX is approved for several stages and subtypes of colorectal cancer. In our view, Exact will continue generating evidence for more clinical use cases and/or cancer subtypes, generating sustained growth for the global Oncotype DX franchise. As an example, Exact showed in its recent POLAR study that Oncotype DX also can predict radiotherapy benefit for early-stage breast cancer patients.
In addition to use-case expansion, Exact Sciences can open up new geographies to sustain Oncotype DX’s growth. Like ASCO, the European Society for Medical Oncology (ESMO) recommends Oncotype DX for guiding risk-stratification of early-stage breast cancer. Despite pandemic-related headwinds, Exact’s international Oncotype DX revenue grew from $11 million in 2019 to $110 million in 2021. Given the lower penetration rate and lack of competition overseas, international Oncotype DX revenue should grow more than 20% per year during the next five years.
Oncotype DX has provided Exact Sciences with other strategic and competitive advantages. While they have different use cases, Oncotype DX and Exact’s forthcoming MRD and HCT tests will serve the same patient and clinician populations. Thanks to Oncocyte DX and unlike its competitors, Exact has developed relationships with 98% of US oncologists, increasing the likelihood of early market traction for new tests and lower customer acquisition costs (CAC). According to our research, Cologuard and Oncotype DX have positioned Exact Sciences to generate roughly 4.25x more annual gross profit than its nearest oncology testing peer. Because of its scale, Oncotype DX brings Exact more than 100,000 early-stage cancer samples per year, an enviable source of training data for the machine learning algorithms supporting its pending MRD tests. Oncologists should be able to use Oncotype DX to target and tailor MRD testing for patients with early-stage disease—a two-pronged service that only Exact currently is positioned to offer. Indeed, Exact already has launched a study—ADAPT—to evaluate this portfolio synergy.
As noted earlier, hereditary cancer testing has been productive in the management of early-stage cancer. Invitae’s recent study (n=2,023) showed that 30.5% of patients across many solid cancer types harbored a pathogenic, hereditary variant. Most of these variants were clinically actionable and can inform decisions about precision therapy and/or clinical trial eligibility. The recent Phase III OlympiA trial (n=1,836), for example, demonstrated that early-stage breast cancer patients with pathogenic, hereditary BRCA1/2 variants, who also were treated with Lynparza (olaparib), saw statistically significant improvements in both invasive disease-free survival and distant disease-free survival, as shown below. Based on these findings, the FDA recently approved olaparib for the adjuvant treatment of certain early-stage breast cancers. As researchers present more supportive clinical evidence, Exact should be able to cross-sell its HCT services to the same oncologists who order Oncotype DX.
Exact’s Minimal Residual Disease (MRD) Testing: Detecting Cancer Recurrence and Response
MRD tests typically involve whole exome or genome sequencing of a patient’s tumor tissue. The lab selects cancer-specific (somatic) mutations that collectively serve as a fingerprint for the tumor. With this information, the lab crafts a “tumor-informed” liquid biopsy so that oncologists can monitor the tumor over time using a non-invasive blood draw. “Tumor-naïve” MRD testing does not require an upfront tissue sample but may be less sensitive for certain cancer types. Oncologists can use MRD testing to assess a patient’s response to therapy, detect recurrence and relapse ahead of medical imaging, or identify a tumor’s resistance to therapy.
MRD testing seems poised to become one of the fastest growing molecular diagnostic markets, thanks to vast amounts of emerging clinical data, the number of potential use cases, and rapidly declining input costs. Our research suggests that the market should support several major winners despite the likelihood of skirmishes in early days. Nascent, fragmented, and large, MRD will not be a winner-takes-all market. Based on the increasing number of cancer survivors and the need for regular testing as lifespans extend, according to our estimates, the US MRD testing market will scale to ~$20 billion annually over the next twenty years. Today, the market penetration is less than 1%.
Exact Sciences is not the first to market in solid tumor MRD testing. Natera’s (NTRA) Signatera test was first to the clinical market in 2019; Guardant Health launched its test, Guardant Reveal, in 2021; and Invitae opened its service, Personalized Cancer Monitoring (PCM), in 2022. Personalis (PSNL) plans to launch its own MRD test, NeXT Personal, sometime during the next few months. While Exact has yet to mention a specific launch date, our expectation is that its first MRD test will launch by the end of 2023 as a laboratory developed test (LDT).
We believe no company has a durable, technologically enabled monopoly over MRD testing. Certain approaches like Personalis’ (PSNL) NeXT and Haystack Oncology’s (Private) DUO, however, do offer part-per-million sensitivity, giving them an advantage with certain cancer types. MRD tests involve naturally occurring tumor biology, leaving little room to exclude competition with intellectual property (IP). Supply-side innovation, such as hyper-accurate and/or low-cost short-read sequencers from Element (Private), Pacific Biosciences (PACB), and Ultima (Private), as well as target enrichment solutions from Twist Bioscience (TWST) could help level the MRD playing field. In the intermediate term, MRD tests that check the boxes on sensitivity, specificity, and molecular quantitation probably can leverage off existing reimbursement codes.
Many of Exact’s platform technologies could play useful roles in the MRD process: among others, Biomatrica for sample stability; TAPS for single-workflow, low-cost methylation detection; Safe-Seq for sequence error suppression; and TARDIS for sensitive quantitation. Oncotype DX offers tens of thousands of early-stage cancer samples on which Exact can train its MRD machine learning classifiers.
Palmetto GBA, an influential Medicare Administrative Contractor (MAC), created the MolDx program to accelerate the approval and adoption of molecular diagnostic tests. Under the MolDx framework, once it covers a new class of test like MRD, Palmetto will reimburse all tests that perform in line with standards, fostering competition that benefits Medicare beneficiaries and eliminating duplicative clinical trials. Other MACs and commercial payers often follow Palmetto’s recommendations with time.
That said, we believe that its scarce assets have put Exact in a strong competitive position in the MRD market: brand recognition, established commercial channels, access to training data, and the multiplicative benefits of a vast commercial oncology portfolio. Unlike assay technology, those assets cannot be bought or built easily.
Exact’s Work in Advanced Cancer Testing
Advanced cancer testing, one of the more mature molecular diagnostic markets that also is known as therapy selection, has a 15% penetration rate. This class of molecular testing surfaces biomarkers that enable precision therapies, many approved only for advanced cancers with a significant unmet need. Most therapy selection tests are based on solid tumor tissue, though much of the same data can be gathered from liquid biopsy as well. As a result, therapy selection tests are more mature than most other tests. Advanced cancer testing should enjoy significant and sustainable growth as the number of precision therapies and biomarkers increases. Barriers to adoption include distribution, education, and clinical decision support, more so than cost or technology. Like other molecular tests, the advanced cancer testing market bifurcates around routine clinical use and biopharma trial use. Biopharma customers typically want comprehensive tests that include investigational content, whereas clinicians prefer quick, narrow panels. Exact has solutions for both groups—Oncomap for clinicians and Oncomap Extra for biopharma investigators, as shown below.
While not a main driver of our long-term forecast, the Oncomap franchise has synergies with the rest of Exact’s portfolio. Ultimately, some patients using Exact’s MRD service will relapse or advance, necessitating therapy selection testing.
Exact Sciences is a clear leader in the nascent molecular oncology testing landscape, with Cologuard and Oncotype DX offering durable sources of gross profit that can fund pipeline assets, debt repayment, and the maintenance of infrastructure. Thanks to the biological, epidemiological, and practical synergies among its testing categories, Exact is becoming more than the sum of its parts. As it grows and broadens its commercialized portfolio, Exact should gain more operating leverage, driving both revenue and profitability.
On its most recent earnings report, Exact delivered record quarterly revenue of $523 million—or 20% year-over-year growth. Management lowered its guidance for operating expenses by $113 million and pulled forward the timeline for adjusted EBITDA breakeven to the third quarter of 2023, citing broad-based efficiency gains throughout its business. During the past five years, Exact Sciences has generated ~$6 of revenue for each $1 spent on R&D. Although a slowdown in topline growth clipped that ratio to ~2.4X during COVID-19, it has recovered to 5.8x, an impressive rebound because Exact spends ~70% of R&D on pipeline tests that do not contribute to revenue. Given the workflow and IP overlap across its on-market and pipeline tests, we believe Exact can gain R&D leverage as the firm’s testing menu expands. Exact’s ability to leverage its selling, general, and administrative (SG&A) expenses, which are higher than R&D in absolute terms and as a percent-of-revenue, is key to its profitability. In late 2021, Exact doubled its primary care physician (PCP) salesforce by hiring ~400 sales representatives from Pfizer that were part of a previous co-marketing agreement between the two companies. Exact’s headcount also increased substantially because it acquired Thrive. In 2023, hiring should slow down meaningfully. Like its strategy after acquiring Genomic Health, Exact has frontloaded general and administrative (G&A) expenses on information technology (IT) infrastructure for automated billing and multi-product ordering, spending that should taper off in 2023 and 2024. Exact has guided to a 4% reduction in G&A spending for 2022, though we assume G&A grows 7% on average per year in our model.
Exact’s largest operating expense (OpEx) is sales and marketing (S&M) into which it has invested aggressively to advertise CRC screening in the 45-49 age group. The sustained lack of physician access is a headwind on sales productivity, which should reverse as the pandemic abates. Because of an overlap in sales call-points, cross-selling an enhanced portfolio should lead to higher revenue per sales rep. We believe CRC re-screening also will be a boon for S&M leverage. Exact has guided to a 6% reduction in S&M spending for 2022, though we assume S&M grows 8% on average per year in our model.
ARK also has exposure to other testing companies that compete with Exact, including Guardant Health (GH), Freenome (Private), Invitae (NVTA), Personalis (PSNL), and Veracyte (VCYT). Each has its unique strengths and weaknesses, and the success of one molecular oncology company need not come purely at another’s expense. As outlined in this article, cancer care is a vast, underserved, durable, and secularly growing market that is likely to consolidate. Exact is one of few companies that have the sales and IT infrastructure to support another multi-year period of growth. Its experienced and trained sales force is an important competitive advantage.
Patients and physicians will be prime beneficiaries of improvements in molecular oncology and targeted cancer care. As life science tools companies launch new platforms that enable lower costs and create new testing applications, the scope of the diagnostics industry will broaden significantly.
As molecular testing firms generate more evidence, provider utilization should increase. As more targeted therapies come online, the need for diagnostics should rise. As digital health and infrastructure companies widen healthcare’s scope, transparency, and accessibility, the addressable market for diagnostics should burgeon. At the dawn of a century that could significantly curtail cancer mortality rates, we believe Exact Sciences will play a leading role.
As with our open-source Roku, Tesla, and Zoom models, we are publishing our open-source Exact Sciences model to strengthen the quality of our research. By sharing our assumptions and modeling methodologies, we hope to solicit constructive feedback and criticism. Like open-source software, we believe that open-source research will be more robust and accurate than research conducted and published non-transparently behind closed doors.
The forecasted performance and price estimates herein are subject to revision by ARK and provided solely as a guide to current expectations. There can be no expectation that the specific security will achieve such performance or that there will be a return of capital. Past performance is not indicative of future results.
FORECASTED PERFORMANCE RESULTS ARE HYPOTHETICAL AND HIGHLY SPECULATIVE, AND PRESENT MANY RISKS AND LIMITATIONS. The recipient should not consider these estimated prices alone in making an investment decision. While ARK believes that there is a sound basis for the forecasts presented, no representations are made as to their accuracy, and there can be no assurance that such forecasts or returns will be achieved by the specific security.
The recipient is urged to use extreme caution when considering the forecasted performance, as it is inherently subjective and reflects ARK’s inherent bias toward higher expected returns. Any higher returns should be viewed as a measure of the relative risk of such investments, with higher forecasted performance generally reflecting greater risk. There is no guarantee that any results will align with the forecasted performance, and they might not be predictive. Some or all results may be substantially lower than projected results and, as with any investment, it is possible that you could lose money.
FORECASTED performance results (single security model simulation forecasts) have many inherent limitations. A recipient account might or might not hold this single security, and the account performance will be affected in proportion to its holding size and the amount of price fluctuation over time. No representation is being made that any client account will or is likely to achieve profits or losses tied to a security in the security model forecasts. In fact, there could be significant differences between these forecasted performance results and the actual results realized.
Forecasted performance has not been achieved by the security, and like all modeled, projected or hypothetical performance, it is important to note that there are multiple versions of a model, and ARK has a conflict of interest in that we have an incentive to show you the best performing results. These forecasts rely on models, which calculate hypothetical performance. Several of the limitations of hypothetical performance models include: 1) reliance on a variety of data obtained from sources that are believed to be reliable, but might be incorrect, inaccurate or incomplete and ARK does not guarantee the accuracy or completeness of any information obtained from any third party, 2) potential inclusion of inherent model creation biases, data discrepancies and/or calculation errors that could cause actual results to differ materially from those projected, 3) NO reflection of the impact that material economic and market factors might have had on investment decisions that would have been in actual portfolios being managed at the time and do not involve market risk, and 4) NO guarantee of future investment results. The forecasted results rely on assumptions, forecasts, estimates, modeling, algorithms and other data input by ARK, some of which relies on third-parties, that could be or prove over time to be incorrect, inaccurate or incomplete.
The forecasted returns are based on a variety of criteria and assumptions, which might vary substantially, and involve significant elements of subjective judgment and analysis that reflect our own expectations and biases, which might prove invalid or change without notice. It is possible that other foreseeable events that were not taken into account could occur. The forecasted performance results contained herein represent the application of the simulation models as currently in effect on the date first written above, and there can be no assurance that the models will remain the same in the future or that an application of the current models in the future will produce similar results because the relevant market and economic conditions that prevailed during the performance period will not necessarily occur. The results will not be updated as the models change, or any information upon which they rely changes. There are numerous other factors related to the markets in general or to the public equity security specifically that cannot be fully accounted for in the preparation of forecasted performance results, all of which can adversely affect actual results. For these reasons, forecasted performance results will differ, and could differ significantly from actual results. FORECASTED PERFORMANCE RESULTS ARE SUBJECT TO REVISION AND PRESENTED FOR ILLUSTRATIVE PURPOSES ONLY.
While ARK’s current assessment of the subject company may be positive, please note that it might be necessary for ARK to liquidate or reduce position sizes prior to the company attaining any forecasted valuation pricing due to a variety of conditions including, but not limited to, client specific guidelines, changing market conditions, investor activity, fundamental changes in the company’s business model and competitive landscape, headline risk, and government/regulatory activity. Additionally, ARK does not have investment banking, consulting, or any type of fee-paying relationship with the subject company.