Publication
Article
To claim CE credit for this activity, please visit https://www.pharmacytimes.org/courses/the-clinical-and-economic-benefits-associated-with-novel-multi-cancer-early-detection-tests-conference-highlights-from-the-2022-ispor-annual-meeting
Introduction
Despite decades of advancements in cancer prevention and treatment, cancer remains the second most common cause of death in the United States. In 2022, approximately 609,360 people are expected to die from cancer, although that number is predicted to climb given the impact of the COVID-19 pandemic on preventive screenings and early diagnosis and treatment. Estimates based on 2018 and 2019 statistics predict 1.9 million new cases of cancer will be diagnosed in 2022.1
Cancer carries a high economic cost, with overall expenditures in 2020 of more than $200 billion.2 Yet while the United States spends more per capita on cancer care, a comparison with 22 other countries found it had a higher mortality rate than 9 others.2
The overall age-adjusted cancer death rate from fell 32% between 1991 and 2019. However, just 4 cancers—lung, breast, colorectal, and prostate—all of which have screening protocols either on a population basis or for certain high-risk groups, account for most of that decline. Improvements in survival rates for other cancers, particularly ovarian, pancreatic, and liver, have been much smaller.1
Cancers without recommended screening tests account for 71% of cancer deaths in the United States.3 The incidence rate of those cancers increased 1.2% between 2000 and 2018, while there was no significant change in cancers with guideline-recommended screenings. Among those aged 50 years and older, incidence rates of cancers without guideline-recommended screenings remained constant, while rates fell 0.1% per year for those with screenings.4
Detecting cancers as early as possible, when treatment is more likely to be successful, is an important way to reduce overall cancer mortality.5 Earlier diagnoses can also result in significant economic savings.One study found an absolute mean cost for women diagnosed with stage I ovarian cancer of $73,428; by stage IV, the cost increased to $227,899, a more than 4-fold difference. Costs for colorectal cancer doubled, from $129,645 when diagnosed in stage I to $271,216 when diagnosed in stage IV.6 Thus, earlier diagnosis could have economic as well as lifesaving benefits, provided it is not associated with a high burden of overdiagnosis and overtreatment.7,8
Patients with cancer shoulder a significant part of the economic burden, with an estimated $16 billion in out-of-pocket expenses to diagnose and treat their cancer in 2020.9 The more advanced the presentation of cancer, the higher those costs. A study presented at the 2022 International Society for Pharmacoeconomics and Outcomes Research (ISPOR) Annual Meeting assessed out-of-pocket costs (co-insurance, co-pay, and deductible) for the first year after diagnosis for 6028 commercially insured patients with cancer, the majority of whom had breast cancer. Those diagnosed with stage I breast cancer had mean costs of $8979 while those diagnosed with stage IV encountered mean out-of-pocket costs of $11,348 (P <.001) The results were similar for ovarian, lung, and colorectal cancers (Table 110). The authors suggested that earlier diagnosis could reduce patient financial burden by requiring less intensive treatment and management.10
The stage at which cancer is diagnosed affects employment status in patients who are still working. An analysis of claims from more than 1.2 million commercially insured patients aged 18 to 64 years found a median time to reduction in employment status of 14.7 months in those diagnosed in the earlier stages (n = 141,440) and 4.3 months in those diagnosed with metastatic cancer (n = 17,875 [P <.001]). These differences were observed regardless of cancer type.11
Cancer Detection: Current Screening
The US Preventive Services Task Force (USPSTF), an independent, volunteer panel of national experts in disease prevention and evidence-based medicine that makes evidence-based recommendations about clinical preventive services, currently recommends population-based screening for breast, colorectal, and cervical cancers, with prostate cancer screening based on a discussion of the potential harms and benefits.12-15 It also recommends screening certain smokers for lung cancer.12Table 27-13 contains a summary of cancer screening tests and recommendations from the USPSTF.
The Task Force’s decisions on screening recommendations are based on 4 factors (Figure 115):
There are different well-recognized guidelines and recommendations but there may be variations when compared with the USPSTF recommendations; one example pertains to prostate cancer, for which recommendations from the American Cancer Society are also widely accepted and recommendations vary based on age and risk.1 Despite available guidelines, screening uptake in the United States remains low. Numerous barriers exist, including financial, educational, psychosocial, and logistical. Others include a lack of provider recommendations as well as a lack of trust in the healthcare system.16-19 Pre-pandemic screenings for certain cancers had already been declining, even among insured individuals.20 Those declines were even higher among those with low socioeconomic status and no insurance. For example, women without health insurance had a 26% to 39% lower prevalence of breast cancer screening compared with those with insurance.20 During the pandemic, screening rates for all cancers plummeted, leaving providers warning of a coming tsunami of later-stage cancer diagnoses.1,19,21,22
False-positive results add another layer of complexity to the decision to recommend population-based screening. For instance, false-positive rates for low-dose computed tomography (LDCT) screening in smokers have been reported as high as 26.6%.23-25 Nearly half of women who undergo an annual mammography will have at least 1 false-positive test over 10 years, with some studies estimating rates between 29% and 77%.26,27
False-positive results elicit significant anxiety in patients, particularly those with a high risk of cancer.28,29 It can also lead to significantly higher medical costs due to biopsies and scans to rule out cancer.30
Multi-Cancer Early Detection Tests
The use of blood tests to screen maternal blood for DNA evidence of fetal abnormalities ushered in the potential of similar screening tests for cancer when the pregnancy tests also picked up DNA evidence of maternal neoplasms.28This spurred the development of several multi-cancer early detection (MCED) blood tests that could be used to identify dozens of cancers at earlier stages, including those without routine screening tests.31-33 The tests often combine artificial intelligence (AI) and machine learning with assays to detect a variety of circulating analytes, typically cell-free DNA (cfDNA), for early indications of cancer (Figure 234).
MCED tests offer several advantages, including potentially greater accessibility than more complicated, organ-specific tests. There is even some thought that they could help reduce cancer-related racial disparities by eliminating the need for access to specialists for screening and ensuring consistent quality and accuracy of test results. However, the opposite could also be true if uptake is low in minority communities and/or there is a lack of insurance coverage.35
One study estimating the impact of adding MCED screening to standard population-based screenings for breast, cervical, colorectal, and lung cancers predicted it would result in a diagnosis of an additional 105,526 cancers with a 25% uptake and an additional 189,498 cancers with a 100% uptake. The estimated true-positive/false-positive (TP/FP) rate with current screening was 1.43 (43 false positives for every person diagnosed with cancer) at a cost of $89,042 to confirm there was no cancer versus 1:1.8 using the MCED test at a cost of $7060 (Figure 336).
While MCED tests have low false-positive rates, there is concern they could increase the diagnosis of indolent cancers that may never impact the mortality of the patient or precancers that never develop into cancer or cause symptoms.31,36,37 Large clinical trials have been conducted for 3 MCED tests.
Galleri
Galleri examines cfDNA methylation patterns across a very large number of methylation sites and uses AI analysis to recognize patterns associated with the presence of cancer as well as the likely organ of origin of the suspected malignancy. It has been shown to detect more than 50 cancers.38 Figure 438 depicts its sensitivity in individual tumors by stage.
Galleri is currently marketed in the United States under a Clinical Laboratory Improvement Amendments (CLIA) waiver and is intended for use in those older than 50 years with and without additional risk factors for cancer in conjunction with currently recommended screenings.3,39 It has received Breakthrough Device Designation from the Food and Drug Administration (FDA).40 Clinical trials assessing the clinical utility of the test are ongoing and there are plans to seek FDA approval in 2023.
Galleri has been evaluated in several studies, including the Circulating Cell-free Genome Atlas (CCGA), a prospective, multicenter, case-control, observational study conducted with specimens from 15,254 participants, 56% of whom had suspected cancer,41,42 and the PATHFINDER trial, a prospective study in 6629 adults aged 50 years and older designed to evaluate the role of the test in clinical practice.31 Publications from the CCGA trial report up to 100% specificity (no false positives) in a group of confirmed, cancer-free patients and a 46.7% cancer detection accuracy among patients with confirmed cancer. The test detected 78.9% of cancers in stage II with tumor of origin predicted in all samples with a 97.1% accuracy.41 A sub-study in 6689 participants (2482 with cancer) found a 99.3% specificity overall for more than 50 cancers.43 Participants in CCGA continue to be followed.
The PATHFINDER study provided a real-world scenario to evaluate the potential use of MCED tests. In this study, the participant and the physician received the results and the physician determined the diagnostic pathway. The goal was to identify the most effective approach to achieve diagnostic certainty after a cancer “signal” from the test. Interim results showed a cancer signal in 92 participants (1.4%), of which 29 had a true positive and 36 had a false positive after imaging and procedures, resulting in a positive predictive value of 45%. The cancer signal origin prediction was 100% when predicting the first or second stage of cancer of origin in those with an elevated cancer risk (smoking history, history of prior cancer treatment completed more than 3 years ago, or genetic predisposition to cancer; 95% CI, 83.2-100) and 87.5% in those without a higher cancer risk (CI, 52.9-99.4). It took about 50 days to diagnostic resolution from the time the test results were returned. The majority of cancers detected were diagnosed at stages I-III. The study has closed accrual with final analysis still pending.31
A follow-up study, PATHFINDER 2, is a currently accruing multicenter clinical trial. With an estimated sample size of 10,000 participants older than 50 years, the investigators aim to evaluate the safety of the MCED test in terms of diagnostic testing triggered by the MCED test result as well as the performance of the MCED test in individuals eligible for cancer screening.44
CancerSEEK
The CancerSEEK test detects the presence of prespecified DNA mutation and protein markers associated with cancer. It is designed as a 2-step process, with a second test to validate the results of the first in the case of an initial positive result.
In the DETECT-A trial, 9911 healthy women aged 65 to 75 years with no personal history of cancer and a high adherence to recommended screening guidelines received the CancerSEEK MCED test coupled with positron emission tomography/CT imaging if both the baseline and confirmatory CancerSEEK test were positive. The study investigators chose to enroll only women to enrich for ovarian cancer, a malignancy that currently lacks standard-of-care (SOC) screening and typically has a favorable prognosis only when detected early.45
Of 9911 women screened, 490 (4.9%) had positive baseline testing, and 127 (1.3%) had confirmation on the second test. After the second test, 64 women had imaging concerning for cancer, with ultimately 26 cancers confirmed by biopsy (6 ovarian). All patients were followed closely for 1 year, during which time an additional 67 cancers were observed in women who had a negative baseline CancerSEEK test (22 detected with SOC screening and 45 detected by other means) (Figure 545).
The DETECT-ASCEND 2 study is an actively enrolling clinical trial aiming to further refine the CancerSEEK MCED test in 21,000 patients older than 50 years in 2 different cohorts of healthy volunteers with no cancer history and those with recently diagnosed but untreated cancers.
DELFI
The DELFI test is based on fragmentomes, disordered DNA packaging in cancer cells, and uses machine learning algorithms to assess patterns. It was evaluated in the 7-year LUCAS trial, which prescreened 365 individuals at high risk for lung cancer to determine if they should undergo LDCT screening. The combination of the test plus LDCT detected 90% of lung cancers, including 80% of stage I cancers, and reduced the number of unnecessary LDCTs by half.33
An earlier study analyzed fragmentation profiles of 236 patients with breast, colorectal, lung, ovarian, pancreatic, gastric, or bile duct cancer and 245 healthy individuals. It found a sensitivity ranging from 57% to 99% among the 7 cancers with a 98% specificity. Overall, it detected cancer in 91% of the patients with cancer.46 Several trials are also underway using MCED screening for colorectal cancer.42,47,48
Managed Care Implications of MCED Tests
Managed care providers must consider several issues, including economic, clinical, and psychosocial, when determining coverage of MCED tests. Clinical trial participation is strongly recommended if a patient is interested in one of these MCED tests. However, if a patient chooses not to participate in a clinical trial, the cost is expected to be about $1000, with Galleri currently sold in the United States at $949 per test.49
An estimated 120 million people in the United States are 50 years or older.50 If half of these individuals were tested, the cost for the tests alone would approach $60 billion, with additional costs for testing and treatment in those with identifiable cancers. It is also not clear how often the tests might be given or their impact on current screening uptake. In addition, because follow-up diagnostic tests will be required to confirm initial test results, patients may face significant out-of-pocket costs.51,52
Nonetheless, bipartisan bills in the US House of Representatives and Senate would require Medicare coverage of the tests once they are approved by the FDA.53 The economics of MCED tests are based on modeling since the tests are not yet in wide use. They also must use certain assumptions about the potential outcomes of the tests to reduce cancer-related deaths and treatment as no studies have yet evaluated this issue.
An abstract presented at ISPOR 2022 suggested using a mixture cure model, often used to determine cure fractions due to cancer pharmacologic therapies, to assess the economic impact of earlier cancer diagnosis. One advantage is that it allows for more robust calculation of costs attributable to cancer by isolating end-of-life costs from patients who do not die from cancer. Using this approach, the authors aggregated reductions in cancer-attributable costs after incorporating cure fraction versus no cure fraction of −26.5%, −9.45%, and −0.5% for local, regional, and distant, respectively.54
There are concerns that making MCED tests part of Medicare’s annual screening benefit could overwhelm the healthcare system due to the current cost per test plus the cost of follow-up scans and biopsies.55 An abstract presented at ISPOR 2022 modeled the potential of a value-based price (VBP) for an MCED test in a Medicare population. The model predicted that the proportion of cancers detected at stage IV would decline from 20.4% to 10.7%; patients diagnosed with an MCED test as well as SOC screening would have on average $5397 less in cancer-related treatment and diagnosis costs versus SOC screening alone. However, the model did not include the cost of the MCED test (which ranged from $730 to $1771 per test) or additional workup costs to confirm the diagnosis. At willingness-to-pay (WTP) thresholds of $50,000/QALY to $150,000/QALY, the authors concluded that VBP in this population was sensitive to the number of clinically significant cancers detected.56
A Markov model comparing an annualMCED test plus SOC screening in US adults aged 50 to 79 years predicted that 10.7% of cancers would be detected at stage IV by adding an MCED test compared with 20.4% with SOC screening. The combined screening method led to 0.17 more QALYs and reduced cancer-related treatment and diagnosis costs by $5208. At a WTP threshold of $50,000/QALY, the VBP would be $923 per test; at a WTP threshold of $150,000/QALY, the VBP would be $2069.11,57
Another health economic model comparing the clinical outcomes of adding an MCED test to current cancer screening versus current screening alone over a lifetime in people aged 50 to 79 years found those receiving an MCED test had an incremental gain of 0.35 life-years and 0.34 QALYs per person, primarily due to earlier detection of cancers and improved mortality. At WTP thresholds of $100,000/QALY to $150,000/QALY, that translated to a cost benefit of $34,400 to $51,600 per person, respectively.58
A study estimating the impact of adding MCED screening for breast, cervical, colorectal, and lung cancers to current screening modalities estimated it would detect an additional 105,526 of these and other cancers with a 25% uptake and 422,105 with a 100% uptake. The estimated TP/FP ratio, which demonstrates how many patients without cancer should undergo diagnostic evaluation to detect 1 person with cancer, was 1:43 ($89,042 per cancer detected) with the 4 SOC tests and 1:1.8 with MCED ($7060 per cancer detected). Combining both yielded a TP/FP ratio of 1:14 in the United States, with diagnostic costs per cancer of $32,461.36
The authors noted that while reported sensitivities for MCED tests tended to be lower than currently used screening tests, potential advantages could be that they cover multiple cancers and have a relatively low false-positive rate. They also noted that a single test could help overcome the high rates of nonadherence seen in the United States.36
Another consideration is that the lower false-positive rates of MCEDs could reduce patient anxiety as well as costs.30,59 The results of a vignette-based utility elicitation study presented at ISPOR 2022, in which participants completed the time-tradeoff utility interviews, found a true negative had the highest utility at 0.978 (SD 0.028), as might be expected. Among scenarios with false positives, the scenario in which the test delivered a false-positive result for breast cancer but no biopsy was performed had the highest utility at 0.958 (SD 0.016), while a false-positive result for lung cancer with a follow-up CT scan at 6 months had the lowest utility (0.847 [SD 0.122]).60
There are other considerations beyond the cost of the tests.61 For instance, it should be considered whether the test be applied only to those at high risk of developing certain cancers or with known risk factors, such as obesity. Discussions should include when they begin and how often they should be received. Additionally, it must be determined how the tests will integrate with existing screening protocols. Assessment of potential psychological effects of early multi-cancer detection should be addressed and involve detailed patient monitoring. Other considerations include follow-up protocols for patients with positive tests; use of the tests in patients whose cancer is in remission; and the need for large-scale, real-world studies to determine their utility in reducing deaths from cancer and their impact on patients.61
The National Cancer Institute plans to launch a large trial within the next 2 years to evaluate these and other questions using several MCED tests. The trial will be designed using feedback from a 2021 workshop the agency held with research leaders.62
Conclusions
Cancer continues to be a leading cause of mortality in the United States, second only to cardiovascular disease. While the overall cancer-related mortality has fallen by one-third since 1991, much of that decline occurred in just 4 cancers, lung, breast, colorectal, and prostate, for which there are screening protocols in place. The majority of other cancers are diagnosed in later stages, at which time treatments are typically less effective. It is clear from decades of research that early detection is associated with improved prognosis and survival. However, the USPSTF recommends widespread screening for just 4 types of cancers listed previously. There are no recommended screening tests for cancers that cause 71% of all cancer-related deaths. In addition, screening rates remain below national targets, with several barriers to population-based screening.
Clinical studies show that MCED tests can identify multiple cancers early, including those that lack screening tests. This has the potential to reduce mortality rates and overall direct and indirect healthcare costs. The use of noninvasive MCEDs to screen for multiple cancers simultaneously may improve early diagnosis of many cancers. The evidence is strong as to their specificity and sensitivity; yet, despite the clinical efficacy of these tests, at this time there is no evidence that they can prevent deaths or reduce cancer-related costs. These data, along with policy discussions regarding coverage, frequency of use, accessibility, and other issues, are needed.
Author affiliation: Nima Nabavizadeh, MD, is Associate Professor, Department of Radiation Medicine atOregon Health & Science University, Portland, OR.
Funding source: This activity is supported by an educational grant from GRAIL, Inc.
Author disclosure: Dr Nabavizadeh has no relevant financial relationships with commercial interests to disclose.
Author information: Substantial contributions to analysis and interpretation of data, critical revision of the manuscript for important intellectual content, and supervision.
Address correspondence to: nabaviza@ohsu.edu
Medical writing and editorial support provided by: Debra Gordon, MS
REFERENCES