POM Health

POM Health Systems was founded to fill clear gaps in the way the United States health system screens for cancer and to give people an effective option to manage their cancer risk. Our cancer screening model is run by some of the most hardworking, compassionate, and intelligent physicians, businesspeople, and policymakers in the world, but is still responsible for identifying just 14% of all cancer cases, more than 50% of which will be caught at stage III+. These statistics might be acceptable for a disease which affects 1% of the population, but not, in our opinion, one that will affect 39% of the population in their lifetimes and is fatal in all cases if untreated.

In developing POM Health, we took a first principles approach to addressing these and other statistics which describe the efficacy of our cancer screening system. We observe that the United States has both a detection and an early detection problem. We have provided a summary table of the key challenges we see, and go on to provide more context and rationale below.

• Low breadth of screening: Apply technologies that can pick up cancer signals for as many cancers as possible. We can detect at some stage 90%+ (180+/200+) of all cancers with the tests in our screening packages.
• Late age of screening tests: Screen any adult patient.
• Low frequency of screening: Screen patients annually. Allow for one-time screens but push people towards annual plans.
• Low signal strength for early detection: Apply technologies with overlapping coverage over as many cancers as possible. We have overlapping coverage for 80%+ (160+/200+) of all cancers with the tests in our screening package.
• Unclear next step if cancer signal detected: Pair every person who gets a screen with a top-tier, vetted oncologist from a major research institution who can react appropriately no matter the results of the screening tests.
• No insurance coverage for most: Relax the insurance constraint and instead partner with the makers of these tests to make them less expensive. We have reduced prices on our screening menu by ~75% and patients can use their multiply tax-advantaged HSA/FSA funds for our screening tests.

Rationale: Detection

Taking detection first, it is a fact that we have the tools required to detect, at some stage, most cancers. Oncologists apply them primarily in a high-suspicion or post- diagnosis context. The challenge we see with detection is how and when those tools are applied. There are five guideline approved and insurance covered cancer screening tests available to the general population, breast, colon, cervical, lung, and prostate. However, there are more than 200 known cancers, and 7/10 cancer deaths occur in cancers for which there isn’t a guideline-approved screening.

The ages at which screening begins for 4/5 of these tests are 40, 45, 45, and 50 (cervical is the exception here, as screening is applied in an OB/Gyn setting from 21 on) and screening procedures default to multi-year gaps between tests, yet early-onset cancers have doubled across all cancers since 1990, with a significant portion of these coming from colorectal cancers. While an oncologist might see the warning signs in people who do not meet screening criteria, oncologists don’t play an active part in screening the healthy population. Primary care physicians, who are on the front lines of cancer screening, generally won’t use tests which aren’t covered by insurance, meaning they are missing many potential cancer signals by design. Furthermore, if a patient does not meet guideline-recommended criteria, it is likely they will discount symptoms that could, if followed, lead to a cancer diagnosis.

So why then if we have the tools don’t we use them to screen for more cancers in more people? There are many answers, but it mostly has to do with the cost of doing so. The primary payers in our system are insurance companies, and their models tell them that it is cheaper to treat some people than screen most people and treat those who have cancer. Oncologists advocate for more screening and have advanced guidelines over many decades, but insurance coverage and broad access requires long-term medical and economic data. This underlines one of the fundamental challenges with cancer screening that we seek to address at POM. screening guidelines are a compromise that doesn’t always produce the best outcome for the patient.

Rationale: Early Detection

Now on to early detection. Early detection improves 5-year survival rates across all cancers from ~20% to ~90%, better than any individual therapeutic. Once a cancer spreads, it is much more difficult to treat and is significantly more likely to recur (~10% risk of recurrence over 10 years at Stage I vs. ~90% at Stage IV). POM firmly believes that early detection is the single best tool we have to address cancer. Single and multi-cancer early detection tests are well into their long march towards insurance coverage and wider access. Companies like Grail, whose Galleri test POM uses in its screening plans, sell hundreds of thousands of their tests each year to consumers.

But why, if these tests are so effective, do they not sell millions? Early detection is a technically challenging endeavor, equivalent to finding the proverbial needle in a haystack. It has taken years to generate sensitivity, specificity, positive predictive value, and negative predictive value data that many people require to purchase the tests and employ them usefully in a clinical setting. Grail’s Galleri, which looks for a shared cancer signal across 50+ cancers, has a false positive rate of 0.4%, the lowest on the market, meaning a negative signal very likely indicates that there is no cancer present. About 62% of the time when it calls a cancer signal, a cancer will be truly present, and 93.4% of the time their powerful ML-guided analysis will be able to predict the tissue of origin, which helps improve triage. Still, until these tests are perfect, we believe they should be employed alongside other tests to help qualify any cancer / no cancer signals, and that analysis and triage should be oncologist-driven, rather than by primary care physicians or consumers on their own.

Our Philosophy

When we talk about our work to people in the community, we either get high praise or warnings that by screening more, we will find more, not all of which will be cancer, and triaging those findings could cause patients more harm and stress than good and add costs to the system. Underpinning this warning is a central medical principle of, first, not doing any harm to a patient. Unlike some of our peers, we take on the legal responsibility of medical care for our customers and are bound by the medical ethics of do no harm. We are not a tech company first. We are a medical provider first, and that mandate is codified in the type of entity through which we administer our screenings (i.e., a medical professional corporation). We fundamentally disagree that this principle should be used to justify inaction, however, because the severity of not looking and missing a cancer is almost always higher than the sum total of the stress and potential for harm of the triage process. Furthermore, while every medical provider should be mindful of adding unnecessary costs to the system, we firmly believe that a system which spends more on screening will spend less on treatment, and that the data will bear this out over time.

So what is the philosophy underpinning our screening plans? Our philosophy is to apply the best tests available at any given time which most efficiently provide reasonably comprehensive coverage over most cancers, and then study the impact of each test to make sure its use is justifiable. We will update our tests based on evidence we’ve gathered and we will develop frameworks for how to react most appropriately to the cancer signals and combinations of signals we generate. We will support the work of the makers of these tests as they seek insurance coverage, advocate for more and better screening, and will drive towards broader access with compassion.

POM Health founders