Going to extremes: determinants of extraordinary response and survival in patients with cancer

Abstract

Research into factors affecting treatment response or survival in patients with cancer frequently involves cohorts that span the most common range of clinical outcomes, as such patients are most readily available for study. However, attention has turned to highly unusual patients who have exceptionally favourable or atypically poor responses to treatment and/or overall survival, with the expectation that patients at the extremes may provide insights that could ultimately improve the outcome of individuals with more typical disease trajectories. While clinicians can often recount surprising patients whose clinical journey was very unusual, given known clinical characteristics and prognostic indicators, there is a lack of consensus among researchers on how best to define exceptional patients, and little has been proposed for the optimal design of studies to identify factors that dictate unusual outcome. In this Opinion article, we review different approaches to identifying exceptional patients with cancer and possible study designs to investigate extraordinary clinical outcomes. We discuss pitfalls with finding these rare patients, including challenges associated with accrual of patients across different treatment centres and time periods. We describe recent molecular and immunological factors that have been identified as contributing to unusual patient outcome and make recommendations for future studies on these intriguing patients.

Fig. 1 |. Classification of patients with an extraordinary treatment response or survival.

Patients with cancer with an exceptional outcome can be classified based on either an atypically good or bad treatment response or on their unusual length of overall survival (OS). a | Rapid progression is observed in a proportion of patients who are expected to respond favorably to conventional or novel therapy. Hyper-progression has been observed in some patients treated with immune checkpoint inhibitors, with apparent accelerated tumour growth on treatment. An example of a patient with primary refractory high-grade serous ovarian cancer, where progression occurs on or within 4 weeks of the end of treatment, is depicted. b | Exceptionally favorable responses can reflect the duration, depth, or proportion of patients responding to therapy, and is most commonly a feature of new treatment approaches. An unusual response (n = 1) can occur when there is a durable response in the context of very few other patients responding to a novel treatment. Alternatively, some patients never relapse: an example of a patient with ovarian cancer in which surgery failed to clear all disease, and who therefore would be expected to relapse in 12–18 months, but who remained disease free for many years after a single line of chemotherapy is depicted. Multiple responders are a clinically distinct subgroup of exceptional responders, showing repetitive profound responses to several lines of chemotherapy. Some but not all exceptional responders may become long-term survivors. c | Most information on short-term and long-term cancer survival relates to conventional therapy in which data from a large number of patients, collected over many years, are available. PFS, progression-free survival; RECIST, Response Evaluation Criteria in Solid Tumours.

Fig. 2 |. Contrasting patterns of survival in breast and lung cancer.

a | Kaplan-Meier survival curves for patients with lung (blue) and breast (red) cancer, showing distinctly different patterns of patient survival over a 10-year period. Survival time of those patients who were alive at last follow-up was censored at their last follow-up. b | Distribution of disease-specific deaths, showing the proportion of patients who die within specific time intervals. In patients with lung cancer, the peak death rate occurs sharply within the first few months after diagnosis, whereas breast cancer-specific deaths occur over an extended time period. Survival distributions vary between cancer types and show that overall survival is not normally distributed. Patients who were alive at last contact were excluded from this analysis. c | Conditional survival analyses indicating the probability of surviving an additional year if a patient has already survived a certain amount of time since diagnosis. In breast cancer, the risk of death remains remarkably constant, whereas in lung cancer, the chance of surviving increases substantially over time. The shape of the curves associated with disease-specific deaths and conditional survival for lung cancer suggest that the greatest difference in determinants of outcome may be found between patients that survive less than 2 years with those who are alive after 4 or more years. Data for this figure were extracted from Surveillance, Epidemiology and End Results Program (SEER; www.seer.cancer.gov) Research Data (1973–2015), National Cancer Institute, DCCPS, Surveillance Research Program released April 2018, on the basis of the November 2017 submission. All patients diagnosed with lung (n = 266,779) or female breast cancer (n = 521,857) between 1995 and 2005 and available follow-up data were included to allow a >10-year follow-up period.

Fig. 3 |. Examples of factors influencing exceptionally favourable response to therapy and/or long-term survival.

a | Germline genetics can influence response to therapy via imparting a high mutational load, for example through mutations in DNA mismatch repair or polymerase ε (POLE) genes, or by leading to the development of tumours that are vulnerable to platinum-based chemotherapy and poly(ADP-ribose) polymerase (PARP) inhibitors, such as with mutations in BRCA1, BRCA2, or other genes involved in homologous recombination (HR) DNA repair. b | Endogenous antitumour immune responses orchestrated by tumour infiltrating lymphocytes (TILs) are associated with longer survival. Both the tumour mutational load and characteristics of the microenvironment influence response to immunotherapy. c | Presence of a driver mutation or combinations of mutations can determine durable responses to targeted therapies. d | Optimal patient health, access to quality care, and certain co-medications may impart long-term survival in a patient who may otherwise have had a more typical disease trajectory.