Surrogate endpoints or markers—which are laboratory measurements, radiographic images, physical signs, or other measures that are thought to predict clinical benefit but are not themselves a measure of clinical benefit—have been used by the FDA for the approval of drugs through an accelerated approval pathway. This pathway is utilized when it is deemed that a new drug can treat a serious condition and fill an unmet medical need.
The reliability of surrogate markers has been called into question regarding their ability to consistently predict outcomes at the clinical trial level. This is especially problematic when there is a discrepancy between a clinically relevant endpoint and its ability to capture the whole effect of treatment at the patient and trial levels.
One such surrogate marker used in breast cancer (BC) trials that has come under great scrutiny is the pathologic complete response (pCR), which the FDA defines as “absence of residual invasive, cancer on hematoxylin and eosin evaluation of the complete resected breast specimen and all sampled regional lymph nodes following completion of neoadjuvant systemic therapy.” pCR has been used in neoadjuvant randomized, controlled trials and has demonstrated a strong association with clinically relevant survival endpoints such as disease-free survival (DFS), overall survival (OS), and prolonged survival.
Unfortunately, data from the clinical trial level are not as robust. In 2014, the FDA acknowledged that this limitation resulted in a lack of meaningful association at the trial level. The agency tried to assuage concerns by stating that this was due to the limited number of trials analyzed and to the modest spread of treatment effects across the trials, which they described as the small differences in pCR rates between treatment arms since these studies mainly involved older cytotoxic agents.
In a special communication published in the Journal of the American Medical Associate Oncology, researchers provided a critical analysis of the strengths, weaknesses, and misinterpretations of pCR as a surrogate marker in trials of neoadjuvant therapy in early BC. In addition to calling out the discrepancy between patient and trial-level outcomes, they also sought to explain the reason why pCR has limited value as a surrogate marker.
Among the biological and methodological hypotheses that have been offered to explain the weak association between pCFR, DFS, and OS at the trial level are the following: that pCR measures the effect of therapy only on the primary tumor and not on micrometastatic systemic disease; that experimental treatments only increase the pCR rate for patients with sensitive disease who would have otherwise had a good prognosis and fails to overcome resistance in tumors of patients with a poor prognosis; that those patients who do not achieve a pCR maybe still achieve impressive survival benefit without reaching the surrogate marker; and that patients receive additional adjuvant systemic postsurgical treatments that can ultimately affect the prognosis of patients, including those who do not achieve pCR. Since all these hypotheses do not completely account for all discrepancies observed, a search has been on to identify alternative surrogate endpoints in early BC that are more reliable and produce consistent results at the patient and trial levels.
Among the proposed new surrogate endpoints are the use of major pathologic response (mPR), which is the presence of less than 10% of residual vital tumor cells; the residual cancer burden (RCB), which is a continuous index that quantifies the presence of residual disease by combining pathologic findings from primary tumor with those of regional lymph nodes; the use of multiple surrogate endpoints (e.g., pCR or RCB and disease recurrence events); and the use of circulating tumor DNA (ctDNA). To date, mPR and RCB have not been validated as reliable surrogate markers for BC at the trial level. Combinations of surrogate endpoints may result in uncertainty when determining measures, such as event-free survival due to the variation in the precision of the markers utilized. ctDNA is promising in that can exploit data from a large percentage of the study population and avoids some of the limitations of pCR, such as the effect of treatment on primary tumors and micrometastases. Further, it can be assessed serially during follow-up. These advantages include capturing the response of micrometastases and avoiding the diluting effect of postsurgical treatments. However, further research is needed before any of these potential surrogate markers are incorporated into clinical practice.
Pharmacists should be aware of the limitations of the use of surrogate markers when assessing the value of therapy in patients with early BC. They should also keep current with the latest attempts to identify reliable surrogate endpoints that consistently capture data at the patient and clinical trial levels.
The content contained in this article is for informational purposes only. The content is not intended to be a substitute for professional advice. Reliance on any information provided in this article is solely at your own risk.
« Click here to return to Breast Cancer Update.
Published December 13, 2022