Testing for gene mutations that hinder the ability of colorectal cancer patients to respond to anti-EGFR drugs is a cost-effective way of administering such pricey treatments, according to a study based on simulated modeling that was published this week in the Journal of the National Cancer Institute.
However, the same study also found that anti-EGFR therapies – even with the help of genetic testing to guide treatment to best responders – do not cause patients to live much longer than those receiving other types of treatments. Given the steep price of EGFR inhibitors, the cost-effectiveness analysis, conducted by Ajay Behl and several others from the non-profit HealthPartners Research Foundation, highlights the complexities of assessing the value of medical interventions in the real world, where patients are not as tightly controlled as in clinical trials.
In the JNCI study, Behl and colleagues implemented a so-called decision analytic framework to gauge the cost-effectiveness of screening patients for KRAS and BRAF mutations in metastatic colorectal cancer patients who receive Bristol-Myers Squibb/Merck/Lilly's EGFR inhibitor Erbitux. The researchers simulated 50,000 metastatic colorectal cancer patients 10,000 times in order to gauge average costs and patient outcomes and then compared these data points for four treatment modalities: best supportive care (treatments other than EGFR inhibitors); EGFR inhibitors received without genetic testing; testing only for KRAS mutations before giving EGFR inhibitors; and testing for BRAF and KRAS mutations before giving EGFR inhibitors.
The analysis revealed that testing for KRAS mutations before administering anti-EGFR treatments could save approximately $7,500 per patient over the course of their treatment, while adding BRAF testing saves an additional $1,023 per patient.
However, overall survival for patients who received anti-EGFR treatment with KRAS and BRAF genetic testing in the simulation increased by around 12 days compared to those receiving best supportive care (chemotherapy, radiotherapy, surgery, VEGF inhibitors), at a cost of $22,033, which resulted in an incremental cost-effectiveness ratio of $650,000 per life-years saved. This, the authors noted, is "above the generally accepted threshold for acceptable cost-effectiveness ratio of $100,000 per quality-adjusted life year."
Furthermore, the researchers found that providing EGFR inhibitors with or without testing may make a difference in terms of cost but not in terms of patient survival. Administering EGFR inhibitors without KRAS testing to all metastatic colorectal cancer has an incremental cost-effectiveness ratio of $2.8 million per life-years saved compared to the $650,000 per life-years saved when a pharmacogenetic strategy is employed. However, patients in the anti-EGFR modality without testing experienced approximately one more day of survival than those whose treatments were guided by testing.
Although implementing genetic testing for KRAS and BRAF to guide anti-EGFR treatment can save payors around $8,000 per patient compared to administering these drugs without PGx testing, the study results indicate that the cheapest strategy would be to forgo EGFR inhibitors for metastatic colorectal cancer patients altogether. According to Behl et al., payors who "decide against the use of anti-EGFR therapy" can save approximately $20,000 per patient. While there is also a falloff in survival time under this scenario, strategies involving treatments other than EGFR inhibitros may be preferable for payors with a low willingness to pay. Payors with a higher willingness to pay threshold would be more likely to support PGx-guided anti-EGFR treatment, they note.
The cost-effectiveness analysis comes as payors are attempting figure out how best to evaluate the impact of molecular diagnostics on patient care, and price tests accordingly. Behl told PGx Reporter that he hopes the latest analysis provides payors with an opportunity to reconsider their use of EGFR inhibitors in metastatic colorectal cancer.
"In general, our results are less supportive for the use of anti-EGFR therapy than previous analyses, and they indicate lower cost savings from KRAS testing than previously reported," the study authors write in the JNCI paper. "Although we cannot confirm that anti-EGFR therapy is a cost-effective use of health care resources, we can confirm that KRAS testing is cost-saving. BRAF testing may offer additional savings."
In approximately 50 percent of colorectal cancer patients their disease spreads to other parts of the body. If caught early, colorectal cancer is curable through surgery for some patients, but 90 percent of metastases are initially non-resectable. Neoadjuvant treatments, including chemotherapy, radiotherapy, EGFR inhibitors, and VEGF inhibitors, are given to patients to increase the chance that their metastases becomes resectable, or these therapies are provided as palliative care for patients who cannot receive surgery.
The National Cancer Institute estimates that in 2012 there will be around 143,000 new cases and more than 51,000 deaths due to colorectal cancer. Based on the assumption that around 30,000 patients in the US are newly diagnosed with metastatic colorectal cancer each year, a study presented at the American Society of Clinical Oncology's annual meeting in 2009 by Veena Shankaran of the Fred Hutchinson Cancer Research Center and others reported that KRAS testing would save the healthcare system $740 million annually. In contrast, the analysis by Behl et al. found that testing metastatic colorectal cancer patients before prescribing anti-EGFR treatments would save around $103 million per year.
The US Food and Drug Administration has updated the labels of two EGFR inhibitors, Erbitux and Vectibix, to inform physicians and patients that retrospective analysis has shown that metastatic colorectal cancer patients harboring certain mutations in the KRAS gene have a limited response to these treatments. Earlier this year, the agency expanded the use of Erbitux as a first-line metastatic colorectal cancer treatment in combination with FOLFIRI only for KRAS mutation-negative, EGFR-expressing patients.
The analysis by Behl et al. only used data inputs for Erbitux, noting it was the most common anti-EGFR treatment prescribed in the US. Qiagen markets a companion KRAS test for Erbitux, which was recently approved by the FDA (see related article, in this issue).
The FDA doesn't yet recommend testing for BRAF mutations before administering EGFR inhibitors to colorectal cancer patients. However, published data suggest that people harboring BRAF mutations may have poorer disease prognosis and become resistant to anti-EGFR drugs after receiving them in the first-line setting. If BRAF mutations turn out to be a marker that is prognostic rather than predictive of patient response to anti-EGFR treatment, it might have implications for the cost-effectiveness analysis by Behl and colleagues.
According to Bristol-Myers Squibb, advanced colorectal cancer patients receiving Erbitux monotherapy for a median of seven weeks costs $18,000 and patients receiving Erbitux in combination with irinotecan for a median of 16 weeks costs $40,000. Qiagen officials have previously said that the price of the company's Therascreen KRAS companion diagnostic for Erbitux is around $200 and estimated the US market opportunity for the test in metastatic colorectal cancer to be $20 million.
"The cost savings potential for KRAS mutation screening in patients with colorectal cancer are evident," a Qiagen spokesperson said this week in a statement, which highlighted that past cost-effectiveness studies show the US healthcare system could save more than $600 million annually by using the test to identify metastatic colorectal cancer patients who are unlikely to respond to anti-EGFR treatments.
"Cost effectiveness studies need to take into account more than laboratory equipment costs alone, most of which is one-time infrastructure set-up expense," the spokesperson added. "Once in place and staff are properly trained, they provide efficiencies that equate to cost savings in the months and years ahead."
The analysis by Behl et al. had several limitations. The model focused on overall survival and did not provide progression-free survival outcomes for the different treatment strategies. Also, the analysis did not factor in differences in quality of life between the treatment options and did not factor in patient preferences for discontinuing Erbitux treatment due to adverse reactions.
In an editorial published alongside the JNCI paper, University of Washington's Josh Carlson and Fred Hutchinson Cancer Research Center's Scott Ramsey pointed out that by not factoring in quality of life considerations, Behl and colleagues fail to capture the "toxicity-sparing impacts of KRAS and BRAF testing." Additionally, by not accounting for the cost of treating adverse reactions, the researchers likely "underestimate the cost savings," but not by a significant margin since the toxicity profile of Erbitux is "modest."
Real World vs. Clinical Trial
In the same editorial, Carlson and Ramsey argued that the JNCI cost-effectiveness study established a sound economic case for KRAS testing to guide anti-EGFR treatment for metastatic colorectal cancer patients and deemed testing in this scenario to be a "no brainer." However, they attributed the discrepancy on cost savings found by Shankaran et al. and Behl et al. to the fact that the former modeled clinical trial scenarios, while the latter modeled real-world clinical scenarios.
For example, the pivotal trial that led to the conversion from accelerated to full approval for Erbitux in 2007 found that patients treated with the drug plus best supportive care had median overall survival of 6.1 months compared to 4.6 months for those receiving just best supportive care. Retrospective PGx analysis of patients in this study found that Erbitux-treated patients without KRAS mutations experienced median overall survival of 8.6 months compared to five months for those with these mutations.
The authors of the JNCI paper point out, however, that the survival is not as robust when Erbitux treatment is viewed in the context of real-world clinical scenarios, which include multiple metastases, various treatment modalities, and resections. "Our analysis is based on a general population of metastatic colorectal cancer patients, of whom roughly 50 percent have synchronous abdominal and peritoneal metastases with a maximum overall survival of 6.5 months," Behl and colleagues explained. "It is difficult to compare findings from such a patient base to those from randomized-controlled trials and other literature because of the variation in inclusion criteria for patients."
The study authors noted that if their modeling had excluded metastatic colorectal cancer disease with abdominal or peritoneal metastases from receiving anti-EGFR therapy that might have brought their cost-effectiveness findings closer to the findings in other studies. However, such a model would fail to account for the financial impact that various types of colorectal cancer metastases – to the liver, lung, and abdomen – have on the cost of patient care.
Between 15 percent and 25 percent of patients with colorectal cancer end up with synchronous liver cancer, and these patients tend to have worse outcomes than those with metachronous metastatic liver cancer, according to figures from the National Comprehensive Cancer Network. Around 50 percent of metastatic colorectal cancer patients experience synchronous abdominal or peritoneal metastases, in which case they can only receive palliative treatment.
In an effort to mirror real-world care, Behl and colleagues randomly assigned the patient cohort to different treatment paths based on simulated attributes of their disease, such as the nature of their metastases and weekly survival probabilities. The patients could enter different scenarios on a weekly basis, including anti-EGFR treatment, anti-VEGF treatment, oxaliplatin- or irinotecan-based treatments, surgery, surveillance, or death.
In clinical trials of different metastatic cancer treatments survival rates are hard to pin down. The JNCI study authors cite overall survival rates of 4.8 months and 20 months for studies involving treatment modalities other than Erbitux and between 8.1 months and 23.5 months for Erbitux-containing regimens. The main reason for the variation in survival findings in clinical trials "is likely to be the time of diagnosis of the metastases," they state.
Additionally, in contrast to other, more favorable cost-effectiveness analysis, Behl and colleagues delved into the specific interventions that add to the cost of care. For example, the cost of resection – the only intervention with curative potential – is lowest when no anti-EGFR treatment is given, at $8,600. The cost of resection with EGFR inhibitors without genetic testing is $11,500, and only $200 less when genetic testing is used to guide treatment decisions.
Behl told PGx Reporter that the resection cost increases with anti-EGFR therapy with or without genetic testing compared to when resections are performed in the context of best supportive care because more resection surgeries are likely occurring in patients given anti-EGFR treatments. "We want more successful resections, in order to increase the overall survival," he said.
The study pegs the total average costs for best supportive care at $34,291. The average cumulative treatment cost increases to $56,234 and $64,841 when EGFR inhibitors are added to the treatment paradigm, with and without genetic testing, respectively. "The vast majority of additional treatment costs associated with anti-EGFR therapy is attributable to chemotherapy, with comparatively little increase in resection costs and subsequent improvement in survival," the study authors write.
Given the complexities of treating patients in community practices, where the majority of cancer care is delivered in the US, doctors have the challenging task of figuring out based on data from clinical trials which interventions will not only improve patient outcomes but also not unduly add to the cost of care. Carlson and Ramsey in their JNCI editorial caution physicians to carefully consider the results of clinical trials involving molecular testing before incorporating such tests in their practices. "Molecular testing is as much about generating cost savings by identifying nonresponders as it is about improving survival by identifying responders," they wrote. "In this context, highly accurate tests are needed to mitigate the potential negative survival impacts of misclassification."
Carlson and Ramsey further warn that some molecular tests may not be as good as they first seem in clinical trials. "Community practice is messy," they said. "Tests will be inconsistently applied and evaluated. Patients discontinue therapies more quickly than they do in registration studies. Distant metastases are much more difficult to manage than they are in clinical trials."
Willingness to Pay
Behl could not provide an estimate for how much payors are currently spending on EGFR inhibitors compared to other types of treatments for metastatic colorectal cancer. Although their analysis found KRAS and BRAF testing to be a cost-effective intervention for administering Erbitux, "the paper should provide all stakeholders with an opportunity to rethink the use of anti-EGFR treatment in mCRC," Behl told PGx Reporter.
The study points out that for the payor with a willingness to pay threshold of less than $1 million per life-year saved, anti-EGFR treatments with PGx testing is "definitely preferable." But still, this is above the acceptable cost-effectiveness ratio of $100,000 per quality-adjusted life year threshold for most payors, the study authors note in the JNCI paper.
"At the lowest levels of willingness to pay, not using anti-EGFR therapy is preferred with certainty. With an increase in willingness to pay, the screening strategies are preferred," the researchers state. "After about $3 million per year willingness to pay, providing anti-EGFR treatment to everyone is most preferred."
Beyond the economic value of KRAS testing in metastatic colorectal cancer, Carlson and Ramsey note that the JNCI study highlights the challenge before payors in determining reimbursement for molecular diagnostics as they increasingly become part of mainstream care. "How would we consider a test that affords a similarly large savings in costs but yields substantially fewer life years (or quality-adjusted life years) because of poorer test performance?" they posited. "What if the treatment offers a moderate benefit to all but a much higher benefit to a molecularly targeted minority?"
Payors are currently struggling to figure out the metrics by which to measure the value molecular tests add to patient care. The Centers for Medicare & Medicaid Services has directed Medicare contractor Palmetto GBA to establish the MolDx program, a system of gauging molecular diagnostic utilization and setting reimbursement policy based on the clinical evidence showing that a test is useful in improving patient outcomes or lowering healthcare spending. Based on the clinical evidentiary data submitted by labs and the utilization information gathered by Palmetto, the contractor will determine CMS's reimbursement policy and establish "value based" pricing for the molecular diagnostics performed (PGx Reporter 11/16/2011).
Noting the rising spending on molecular diagnostics, insurer UnitedHealthcare is supportive of a national registry to track utilization of molecular diagnostics. In a working paper released in March, UHC reported that its health plan participants racked up nearly $500 million in genetic and molecular diagnostic testing costs in 2010, a 14 percent increase on a per-person basis since 2008. Overall, national spending on genetic tests and molecular diagnostics "may have reached around $5 billion" in 2010 and could reach as high as $25 billion by 2021, according to the report (PGx Reporter 3/14/2012).
Lee Newcomer, senior vice president of oncology services at UHC, has previously proposed a model for determining whether a test should be reimbursed by payors, suggesting that a test must either show 10 percent improvement in patient outcomes over the standard of care, or reduce the cost of care by 10 percent (PGx Reporter 11/7/2012).