top of page
  • Writer's pictureDr. Robert A. Nagourney, MD

Biomarkers in Oncology

Updated: Oct 25, 2021

The treatment of advanced human malignancies has progressed slowly since the first introduction of systemic chemotherapy in the 1950’s.

Our capacity to harness adaptive immunity using T-cell checkpoint inhibitors has added a new modality for treatment. It has long been recognized that the selection of candidates for treatment using biological markers could enhance outcomes, facilitate drug discovery, reduce costs and curtail futile care.

These predictive/prognostic biological markers are known as “biomarkers”.

Biomarkers are defined as “objectively measured indicators of biological processes or response to a therapeutic intervention.[1] The FDA uses four subgroups I) exploration, 2) demonstration, 3) characterization and 4) surrogacy, with only surrogacy accepted for drug approval. Recognized surrogates include serum cholesterol, HIV viral load and blood pressure. Biomarker validation including calibration, precision, specificity and sensitivity remains the principal developmental hurdle. In this expanding era of immunotherapy, the need for validated predictive biomarkers has rapidly grown.

Checkpoint inhibitors targeting PD1, PDL1and CTLA4 offer fertile ground for discovery. PDL1 (CD279) is highly expressed in T, B, NK-cells and macrophages and serves as a receptor for PDL1, PDL2, B7H3, and B7H4. PDL1 expression in tumor cells, immune cells or both has been proposed as a predictive biomarker yet technical variability, differing cutoffs (1%, 5%, 10%, 50%), frozen vs FFPE and varying clinical circumstances continue to complicate treatment-candidate selection.

Recently, BIM expression was reported as a new PD1 response biomarker.[2] Additional markers, including CD8+/CD20+ ratios, circulating cytokines and the use of tumor exomes are under investigation.[3] Anti-CTLA4 (ipilumab) has also been the subject of biomarker analyses with response shown to correlate with T-cell count, T-cell activation, inflammatory micro-environment and T-cell clonotypes. Whole exome sequencing has been used to screen candidate neo-antigens with immune signatures then shown to correlate with outcome (p=0.01).[4] However, these investigators noted, “no gene was universally mutated”.

The complexity of human immunity continues to challenge those seeking to identify specific mutations, splice variants or amplifications that can segregate responders from non-responders The NEJM study used phenotypic autologous T-cell response to identify relevant neo-antigens while the Mayo investigators defined a “response phenotype” using BIM expression Thus genotypic interrogation can be enhanced through the study of the human phenotype. Closing the gap between genotype and phenotype offers unique opportunities to advance cancer therapy and drug development.

To date, genomic prediction of clinical response to “molecularly targeted agents” has met with limited success. One study provided a 1.5% (1/68) objective response rate in colon cancer patients who received molecularly targeted therapy,[5] similar to the 4% (1/27) response rate observed in BRAF mutation (+) colon cancer patients selected for Vemurafenib.[6] A recent trial that randomized patients to “molecular selection” vs. “physician choice” showed no difference in time to progression (2.3 vs 2.0 months) with the authors concluding that “Off label use of molecularly targeted agents should be discouraged”.[7]

The decades-long focus upon altered cell proliferation over more modern concepts of altered cell survival (apoptosis), a focus upon the cancer cell and not its micro-environment and the promotion of genomics over functional platforms have contributed to slow progress in cancer research. Nonetheless, the application of laboratory models capable of interrogating the biologic basis of clinical response at the phenotypic level has the potential to inform and accelerate future developments

We have explored functional analyses that examine human tumor biology phenotypically. The Ex Vivo Analysis of Programmed Cell Death (EVA-PCD™) incorporates the modern tenets of drug induced programmed cell death in the context of human tumor primary culture microspheroids that {{cta('8abe8785-a4a0-40ee-b960-71ebb51d4a37','justifyright')}} recapitulate native state human tumors replete with stroma, vasculature, inflammatory cells, cytokines, and cell-cell interactions. Results have been shown to correlate with response, time to progression and survival and have been the subject of prior meta-analyses.[8],[9] Preliminary work supports their capacity to examine biologic response modifiers like VEGF inhibitors.[10] More recently this platform has been applied to the study of human ovarian carcinoma using cell death measures as correlates with metabolomic endpoints.[11]

We are witness to a growing appreciation of human tumor phenotypic analyses as important adjuncts to genomic, transcriptomic and proteomic platforms. Phenotypic analyses have the capacity to interrogate the complexities, redundancies and promiscuities of human tumor biology. The intelligent combination of phenotypic (functional) and analyte-based (molecular) platforms will facilitate patient selection and drug discovery.

[1] Firestein GS. A biomarker by any other name. Nature Clinical Practice Rheum Vol 2:12; 635, 2006

[2] Dronca RS, et al. BIM as a predictive T cell biomarker for response to anti-PD-1 therapy in metastatic melanoma Proc Int’l Canc Immunotherapy Conf, Ab A007, 2015

[3] Whiteside Theresa. Immune responses to cancer: are they potential biomarkers of prognosis? Frontiers in Oncology 3:107;1-8, May 2013

[4] Snyder A, et al. Genetic Basis for Clinical Response to CTLA4 blockade in melanoma. NEJM 371:2189-2199, 2014

[5] Dienstmann R, et al. Molecular Profiling of Patients with Colorectal Cancer and Matched Targeted Therapy in Phase I Clinical Trials.Molecular Cancer Therapeutics, 11(9)2062-2071, 2012

[6] Hyman David M., Puzanov Igor, Subbiah Vivek, et al. (2015) Vemurafenib in multiple nonmelanoma cancers with BRAF V600 Mutations. NEJM 373;8:726-736

[7] LeTournea Christophe, Delord Jean-Pierre, Goncalves Anthony, et al. (2015) Molecularly targeted therapy based on tumour molecular profiling versus conventional therapy for advanced cancer (SHIVA): a multicenter, open-label, proof-of-concept, randomized, controlled phase 2 trial. Lancet, 16-13;1324-1334

[8] Bosanquet Andrew G, Kasper Gertjan J, Larsson Rolf, et al. (2007) Individualized Tumor Response (ITR) Profiling for Drug Selection in Tailored Therapy: Meta-analysis of 1929 Cases of Leukemia and Lymphoma. Blood 110; abs 3471

[9] Apfel Christian, Souza Kimberly, Cyrill Hornuss, et al. (2013) Accuracy and clinical utility of in vitro cytometric profiling to personalize chemotherapy: Preliminary findings of a systematic review and meta-analysis. J Clin Oncol 31, 2013 (suppl; abs e22188)

[10] Weisenthal LM et al Cell Culture detection of microvascular cell death in clinical specimens of human neopalsms and peripheral blood. J Intern Med. 264 (3) 275-287, 2008

[11] D’Amora Paulo, Dale Ismael, Salzgeber Marcia, et al. A Phase II study in epithelial ovarian cancer (EOC) to correlate drug sensitivity and metabolomic signatures with objective response (OR), time to progression (TTP) and overall survival (OS). 19 Congresso Brasileiro de Oncologia Clinica, Oct 2015.


bottom of page