Research teams at Memorial Sloan Kettering Cancer Center and Epic Sciences have found that greater diversity among circulating tumor cells (CTCs) in the blood of advanced prostate cancer patients predicts not only shorter overall survival, but also the development of resistance to key anti-androgen therapies. Furthermore, high diversity among CTCs was not associated with resistance to taxane-based chemotherapy. This suggests that patients likely to fail anti-androgen therapies but potentially benefit from chemotherapy can be identified.
These findings, which are being presented at the American Society for Clinical Oncology Genitourinary Cancers Symposium this week, mark the first direct view of how the diversity of CTCs identified from patient blood samples, also called heterogeneity, affects clinical outcomes.
"Heterogeneity is the next frontier for oncology. Quantitating heterogeneity and understanding how it affects the sensitivity to specific classes of drug will enable the choice of one treatment over another to be based on the unique features of individual patients' disease and has significant implications for the development of new drugs," said Howard Scher, M.D., who led the study and is the Chief of the Genitourinary Oncology Service at Memorial Sloan Kettering Cancer Center.
"Highly heterogeneous cancers are tougher to beat," explained Murali Prahalad, Ph.D., President and CEO of Epic Sciences. "Heterogeneous cancers have a diversity of cell populations, each with distinct genomic variations, cellular morphologies, and protein expression. Since the populations of tumor cells are so different, it's more likely that a single therapy can only treat a small part of the cancer and that highly aggressive CTC species remain to drive disease progression."
The researchers also studied rare subtypes of CTCs, many of which are undetectable by other liquid biopsy tests. They found that many CTC subtypes, even when composing just a minor fraction of the total CTC population, predict shorter overall survival and drug resistance. One newly identified rare CTC subtype, was discovered by an artificial intelligence algorithm, that classified CTCs based on 20 discrete morphologic and protein expression features, and was found in a subset of patients. Patients whose blood contained this type of CTC universally failed all therapies recorded in their medical records and experienced much shorter overall survival. Subsequent genome sequencing of this CTC subtype found that the cells shared a genomic signature distinct from other CTCs, suggesting that a CTC's genomic features may be inferred by visual analysis.
"Viewing cancer at the single cell level transforms our understanding of a patient's disease. This study, while focused on prostate cancer, presents a roadmap for other cancer types for how heterogeneity can inform patient selection in clinical trials, monitoring of resistant clones, and therapy triaging. The observation of widespread and high heterogeneity that affected outcomes in mCRPC patients also provides important insights as to the information being missed by cfDNA and traditional tissue biopsies, but that is visible through CTCs using Epic Sciences' proprietary approach," said Prahalad.
In the study, 9,225 CTCs from 221 metastatic castrate resistant prostate cancer (mCRPC) patients were detected and characterized by the Epic Sciences no cell left behind® liquid biopsy platform, which employs a selection-free methodology capable of identifying all types of CTCs — including unexpected or novel tumor cell subtypes. Each CTC was characterized at the single cell level by 20 discrete morphologic and protein expression features. A subset of CTCs (741) was further analyzed by single-cell genome sequencing for global gene copy number variations. These data enabled computers to generate a heterogeneity score for the subpopulations of CTCs observed in each patient and group CTCs into mathematically similar subtypes (independent of pre-existing categories specified by the research community).
The findings will be reported this Thursday in an oral presentation entitled, "Single CTC characterization identifies phenotypic and genomic heterogeneity as a mechanism of resistance to AR signaling directed therapies (AR Tx) in mCRPC patients".