Dr. Achim Regenauer discusses how far we are along the road to having multi-cancer early detection tests – essentially powerful liquid biopsies – in the clinical setting. It appears that this is much closer than previously assumed, and that these tests will complement, rather than replace, current detection options.
For the past 18 months, it may seem as if medical research has been preoccupied with the challenges of the pandemic, with little resources for or interest in other areas. This, however, is most definitely not the case. Below, you will discover how potentially groundbreaking advances are underway in oncology – these relate to liquid biopsy1, whereby, in theory, cancers are detected via a simple blood draw (test) long before any first symptoms or positive cancer diagnostics from medical investigations.
Liquid biopsy is the result of major advances that have taken place over the last decade in our ability to examine tumor-derived material in blood and other bodily fluids, including urine and saliva. This has been enabled through the development of sensitive assays capable of detecting rare, cancer-specific analytes derived from normal cells, such as circulating tumor cells (CTCs) and cell-free tumor DNA (ctDNA).
The World Economic Forum has identified liquid biopsy as one of the world’s top-10 emerging technologies2.
The cherry on the liquid biopsy cake would be the development of a simple, fast, accurate and inexpensive multi-cancer early detection test (MCED). Until recently, achieving this goal seemed far off because cancers are such a highly heterogeneous group on many levels (e.g., histology and genes).
Now, for the first time, an impressive, independent validation study has been published3. This study gives reason to assume that a large-scale MCED clinical application is conceivable in the near future. If further corroboration follows – and note, further corroboration is an important point that should not be overlooked – a cancer screening and diagnostics game-changer, with significant implications for Life & Health insurance, can no longer be excluded.
Currently, only five useful cancer (traditional) screening tests are available – each for one cancer type: breast, colorectal, cervical, high-risk lung and prostate. Combined, in the US these cancers (excluding prostate) only account for up to 29% of annual cancer incidence and 24% of cancer-related deaths among individuals aged 50-79 years. In addition, adherence to traditional screening programs for these cancers, such as those set up in the US and England, is often below national targets4,5.
Many other cancers, including esophageal, liver and pancreatic cancers, currently have no screening option. Of these, several (e.g., ovarian and pancreatic cancer) also present late, resulting in late-stage detection and late treatment starts, and therefore poor patient outlook. Highlighting the need to improve this situation, one can assume that approximately 70% of all cancer incidences are from cancers without available screening options.
Recent publications describe MCED tests that can detect signals from a wide range of cancers by means of next-generation genomic sequencing (NGS) technology: CancerSEEK from Thrive, Galleri from GRAIL, GENECAST’s ADPSTM, PanSeer from SINGLERA Genomics and PREDICT from Burning Rock.
These tests are being validated in ongoing clinical trials, with encouraging preliminary results6.
However, what’s regularly overlooked in the numerous premature, sensational, and hyped media reports, is that validation is far from an easy or fast process, and that both analytical and clinical validation are required before a routine MCED application can be promoted.
For the above-mentioned tests, clinical validation of the overall clinical practice has yet to be evaluated. Many questions are yet to be answered, questions such as how to interpret the test results, which part (risk group) of the population should MCED be applied to, which age range are the tests best targeted to, and above all, if a test result is positive, what then? That last question is of great significance. If, for example, no imaging technology, such as MRI, is yet able to visualize the tiny cancer spot, as it is below the detection threshold, how should clinicians react to a positive MCED test result? Attempt to remove it by surgery, chemotherapy or radiation, or just wait and see? What should be done if so-called indolent dormant cancers are detected, i.e., cancers that progress slowly and do not pose an imminent threat to a patient’s health? At the other extreme, how should testing for cancers that have no effective treatment be handled? This is just an arbitrary selection of open questions which clearly shows that there is still some way to go.
The Galleri test has been developed to detect more than 50 types of cancer through a single blood draw7, 45 of which have no current clinical screening option. Using NGS technology, the test identifies distinct DNA methylation patterns associated with specific cancers and that simultaneously provide information about the organ of origin.
In June 2021, a leading international cancer journal, Annals of Oncology, published a first validation sub-study of the Galleri test3. This sub-study investigated the performance of the test by means of specificity and sensitivity (see more on this below) in 2,823 people already diagnosed with cancer and in 1,254 people without cancer. The mean age of all participants was 60.6 years (55.4% female).
One of the most important requirements of a cancer screening test is high specificity for malignancy, which should ideally be close to 100%. As the prevalence of a particular cancer type can be assumed to be relatively low in the general population, high specificity is necessary to minimize possible harm due to false-positive findings. This is currently not the case for existing protein-based cancer biomarkers, such as PSA used in the screening of prostate cancer and CA 125 used in the screening of ovarian cancer.
In addition to high specificity, high sensitivity is necessary to enable the detection of small/low volume tumors8.
According to the above-mentioned publication, the specificity of the Galleri test was 99.5%, i.e., close to the ideal of 100%. From this 99.5%, it can be interpreted that a positive test result is correct (i.e., cancer exists) in 99.5% of the time, or to put it differently, only 1 in 200 positive results are false-positive results.
In contrast, the overall sensitivity parameter (the likelihood of detecting a cancer that really exists) of the Galleri test was rather low: just 51.5%. However, the sensitivity results were heavily dependent on the cancer stage (stage 1: 16.8%; stage 2: 40.4%; stage 3: 77.0%; stage 4: 90.1%) and type. Furthermore, the sensitivity was much higher in the prespecified group of 12 cancer classes that accounts for two thirds of annual cancer deaths in the US (anal, bladder, bowel, esophageal, stomach, head and neck, liver and bile duct, lung, ovarian, pancreatic cancers, lymphoma and multiple myeloma). For this prespecified group of cancer classes, the overall sensitivity was 76.3% for all stages and 67.6% for stages 1-3.
The study determined, therefore, that the Galleri test has the capability to accurately detect cancer, often before any symptoms arise and with a very low false-positive rate. The test could also predict with a high degree of accuracy where (in which organ) the cancer is located in the body.
The results show that across all cancers and stages, the Galleri test detects approximately 5 in 10 cancers. Although this is low, it must be stressed that for most cancers there is no other screening option available. So, if used alongside existing traditional screening tests, the overall detection of hidden/unknown cancers is substantially increased. The Galleri test does sometimes miss cancers, so it should not replace the currently available screening tests. It is, however, rarely wrong – so a positive test result would need further investigation.
This sub-study is just one of a comprehensive set of studies evaluating the Galleri test, collectively known as the “Circulating Cell-free Genome Atlas Study” (CCGA). The CCGA is a prospective, case-controlled study with a 5-year longitudinal follow-up, recruiting individuals aged over 20 years, looking at more than 50 distinctive cancer types in blood and tumor tissue samples from 15,254 individuals from 142 locations in North America, including individuals with new cancer and blood samples from people without a cancer diagnosis.
It should be stressed that as a case-controlled study, the CCGA is not reflective of performance in a screening population, let alone an insured population. Further large, prospective trials will follow with the objective to better examine the test’s feasibility for screening populations: the STRIVE, PATHFINDER and REFLECTION studies in the US, and the SUMMIT study in the UK.
It seems that the development of MCED tests is progressing faster than previously assumed. This is also reflected in the fact that two major national healthcare systems are already including possible applications of these tests in their future planning. In the US, the Galleri test has not yet been approved by the US Food and Drug Administration, but it is already commercially available (from the Providence Health System for an out-of-pocket cost of around $950)9; test results are available within two weeks. GRAIL is, however, striving for an FDA approval for the Galleri test in 202310.
At the end of 2020, GRAIL announced a commercial partnership with the UK’s National Health Service (NHS) to trial the Galleri test. This pilot trial will involve approximately 165,000 NHS patients, to be expanded to around 1 million people by 2025. First results are expected by 202311.
With major healthcare systems already taking MCED tests into account and the abovementioned finding that the Galleri test is likely to be a complementary test, rather than a substitute for currently recommended screening options, we begin to see the potential approach and extent of MCED testing.
MCED tests leverage newly emerging genomic and information technologies to usher in a transformative paradigm shift in cancer screening and revolutionize oncology within the next decade12. The Galleri test seems to be one of the closest to clinical application. In the mid-term, the likely implications for healthcare systems and Life and Health insurance are huge. It is certainly time for the insurance industry to monitor these advances in view of the potential impact on products, pricing, claims and underwriting approaches. It remains too early to assess the implications with sufficient certainty, but it is nonetheless useful to begin to sketch these out:
As described above, MCED tests have the potential to place oncology and healthcare systems on a completely new footing, and with that, insurers. The numerous challenges associated with liquid biopsy are often described as threats in insurance literature. There are, however, also numerous opportunities. Cancer-specific services for insureds, from annual check-ups, identification of cancer risk factors and advice on “cancer health”, to the mediation of highly specialized oncological treatments, together with financing for cost-intensive, new cancer therapeutics.
1 PartnerRe 2021 Cancer Advances series; https://www.partnerre.com/opinions_research/cancer-diagnostics-earlier-detection-impact-on-incidence-mortality/
2 World Economic Forum: Top 10 emerging technologies of 2017 https://www.weforum.org/agenda/2017/06/these-are-the-top-10-emerging-technologies-of-2017/
3 E. A. Klein et al.; Annals of Oncology; Clinical validation of a targeted methylation-based multi-cancer early detection test using an independent validation set https://www.annalsofoncology.org/article/S0923-7534(21)02046-9/fulltext
4 Hackshaw, A., Cohen, S.S., Reichert, H. et al. Estimating the population health impact of a multi-cancer early detection genomic blood test to complement existing screening in the US and UK. British Journal of Cancer (2021). https://www.nature.com/articles/s41416-021-01498-4
5 American Cancer Society; Cancer Facts & Figures 2020 https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/annual-cancer-facts-and-figures/2020/cancer-facts-and-figures-2020.pdf
6 Tomasz M. Beer. Novel Blood-Based Early Cancer Detection: Diagnostics in Development. Am J Manag Care. 2020;26(suppl 14):S292-299 https://www.ajmc.com/view/novel-blood-based-early-cancer-detection-diagnostics-in-development
8 Hyunsoon Cho et al.; JNCI Monographs, Volume 2014, Issue 49, November 2014, Pages 187–197; When Do Changes in Cancer Survival Mean Progress? The Insight from Population Incidence and Mortality https://academic.oup.com/jncimono/article/2014/49/187/898294
9 Roxanne Nelson; Medscape May 17, 2021; Blood Test for 50 Cancers Coming to US Clinics Soon; https://www.medscape.com/viewarticle/951268
11 Peter Russell; Medscape September 13, 2021; NHS to Trial Galleri Cancer Blood Test https://www.medscape.com/viewarticle/958600
12 A 4-part series of insurance-focused articles on cancer advances, PartnerRe 2021: (1) Global cancer trends https://www.partnerre.com/opinions_research/global-cancer-trends-a-fast-changing-risk-landscape/; (2) Early detection/liquid biopsy https://www.partnerre.com/opinions_research/cancer-diagnostics-earlier-detection-impact-on-incidence-mortality/; (3) Targeted cancer therapy https://www.partnerre.com/opinions_research/targeted-cancer-therapy-road-to-precision-medicine/; and (4) Genomic sequencing.
13 Earl Hubbell et al.; Cancer Epidemiol Biomarkers Prev. 2021 Mar;30(3):460-468 Modeled Reductions in Late-stage Cancer with a Multi-Cancer Early Detection Test https://cebp.aacrjournals.org/content/30/3/460
14 Alexander M Aravanis et al. Cell 2017 Feb 9;168(4):571-574 Next-Generation Sequencing of Circulating Tumor DNA for Early Cancer Detection https://pubmed.ncbi.nlm.nih.gov/28187279/