SITC Clinical Immuno-Oncology Network (SCION) commentary on measurement and interpretation of essential biomarkers in early clinical trials

Proposing a framework for prioritization of biomarker studies in immunotherapy clinical trials

Remarkable advances in cancer immunotherapy as well as sophisticated biospecimen profiling technologies have synergistically transformed clinical trial design. Biomarker studies associated with high-quality clinical data have the potential to drive both scientific and clinical progress by capturing the complexity and evolution of the tumor and the host immune response over space, time, and treatment course. Moreover, these observations will be foundational for the anticipated artificial intelligence (AI) revolution. Members of the Society for Immunotherapy of Cancer (SITC) continue to be at the forefront of this exciting and rapidly evolving field. SITC strives to provide ongoing leadership to support best practises and train the next generation of clinicians and scientists. In 2021, SITC initiated the Clinical Immuno-Oncology Network (SCION) in conjunction with the popular Cancer Immunotherapy Winter School. SCION is an annual 1-week intensive workshop for clinicians and scientists from academia and industry focused on immunotherapy clinical trial protocol development. Each group of six early career scholars is supported by a team that includes a clinical oncologist, a scientist/immunologist, a biostatistician, and a patient advocate. SCION is committed to the principle that the cancer immunotherapy community must learn from every trial, regardless of whether it is successful in establishing a new therapeutic option. Integration of well-chosen correlative biomarkers in trial design is crucial if we are to meet this goal. However, navigating an ever-expanding menu of potential biomarker correlatives with limited resources is an ongoing practical challenge. Synthesis of the best available scientific evidence is needed to direct clinical trial biomarker selection, especially for new clinical investigators. With this commentary, and follow-up publications, SCION faculty hope to address the need for better guidance for biomarker selection and prioritization in early-phase immunotherapy clinical trials.

The current commentary is authored by a team representing SCION faculty as well as leaders of the SITC Pathology and Biomarker Committees. In the context of immunotherapy clinical trial design, we outline the challenges and opportunities of biomarker studies and propose a process and framework for biomarker prioritization (see text boxes 1 and 2). Our goal is to solicit feedback from the SITC and JITC readership community, which will inform the development of recommendations for essential (“Level 1”) biomarker data collection and reporting in immunotherapy clinical trials. Acknowledging the limitations of current data and the rapid evolution of the field, we anticipate iterative revisions of essential biomarker recommendations in response to emerging mechanistic and clinical insights. Please provide your guidance and suggestions directly to SITC staff, Ellie Rickman (erickman@sitcancer.org) and Peter Intile (pintile@sitc.org) who will compile them by the end of the next quarter to be addressed in a full article by SCION faculty.

Box 1. Proposed immunotherapy biomarker evaluation criteria, process, and deliverables.

Biomarker evaluation criteria

• Utility in stratification of patients with treatment response and/or clinical outcomes (clinical relevance).

• Provides biological insights into tumor biology, the tumor microenvironment, and/or the patient’s physiologic or immunologic capacity (biological relevance).

• Relevance across cancer types and stages.

• Standardized assays and analysis approaches.

• Human data prioritized but informed by compelling other sources, including murine and primate models, work with organoids, etc.

Biomarker evaluation process

• Literature review and summary of evidence.

• Consensus/Delphi process to include Society for Immunotherapy of Cancer (SITC) Clinical Immuno-Oncology Network (SCION) faculty and members of the SITC Biomarker and Pathology Committees.

Deliverables (planned publication)

• Checklist of recommended biomarker tests for consideration for all immunotherapy clinical trials, with ranked categorization based on level of evidence and expert consensus (see text box 2).

Future directions

• Annual review and consideration during SCION clinical trials workshop, curated by the current leadership.

• Plan for regular updates/revisions as needed based on evolving data.

Box 2. Proposed framework for immunotherapy biomarker prioritization.

Criteria for level one biomarkers: essential

• Strong evidence for association with immunotherapy response and/or patient outcomes.

• Broad relevance across tumor types and treatment approaches.

• High feasibility (standardized testing methods, routine or readily available testing).

Criteria for level two biomarkers: emergent and exploratory

• Emerging evidence for association with immunotherapy response and/or patient outcomes.

• Relevance demonstrated in multiple studies and tumor types.

• Evolving methodology and interpretation standards.

Criteria for level three biomarkers: contextual

• Strong evidence for relevance in the particular tumor type or treatment under investigation.

• Prior data to inform assay methodology and data interpretation.

Biomarkers provide valuable snapshots of the underlying biology

The Food and Drug Administration (FDA) BEST Resource (https://www.ncbi.nlm.nih.gov/books/NBK338448/%23IX-S, Biomarkers, Endpoints, and other Tools) defines a biomarker as a “defined characteristic that is measured as an indicator of normal biological processes, pathogenic processes, or biological responses to an exposure or intervention, including therapeutic interventions”, and describes the following categories of biomarkers: (1) a susceptibility/risk biomarker; (2) a diagnostic biomarker; (3) a monitoring biomarker; (4) a prognostic biomarker; (5) a predictive biomarker; (6) a response biomarker; or (7) a safety biomarker (table 1). Biomarkers may include molecular, histologic, radiographic, or physiologic indicators, and some may fall into more than one category (eg, tumor histology can be diagnostic, prognostic, and predictive, as when used for patient selection). With the passage of the 21st Century Cures Act, the FDA has provided additional guidance regarding biomarkers accompanying a therapeutic agent (https://www.fda.gov/drugs/biomarker-qualification-program/about-biomarkers-and-qualification). They established a three-stage submission process to develop a biomarker for regulatory use. Independent of regulatory requirements, the REMARK initiative (Reporting recommendations for tumor MARKer Prognostic Studies) outlined principles for testing new biomarkers and reporting performance to facilitate evaluation and compare results across studies.1 Despite these thoughtful resources and recommendations, the question of when to use individual biomarkers in research and clinical practice is only settled gradually, through the accumulation of data from research publications and statistical validation across multiple studies.

Table 1.

Biomarker goals in immuno-oncology

Biomarker goal	Definition	Example(s)
Risk determination	Measuring risk or susceptibility for a particular disease.	BRCA mutations for breast or ovarian, pancreatic cancer risk.
Disease diagnosis	Determining whether a patient has a specific disease or disease subset.	M-protein for multiple myeloma; Her2/Neu expression in breast cancer; pathology/cytology.
Patient selection	Selecting patients that are more likely to benefit from a given therapy.	Tumor PD-L1 expression for PD-1-based therapy.
Prognosis	Assessing outcome from baseline or early treatment samples independent of therapy.	Neutrophil/lymphocyte ratio or LDH levels for multiple cancers.
Prediction	Assessing outcome from baseline or early treatment samples to a specific therapeutic intervention.	Tumor mutation burden or immunoscore for checkpoint inhibitor response.
Monitoring	Following tumor burden or therapeutic progress over time.	PSA as measure of prostate cancer burden; PCR for minimal residual disease in leukemia.
Response	Verifying or correlating with a clinical response.	Pathologic response; radiomic imaging.8
Safety	Alerting to a potential adverse event.	Cytokine levels for immune-related adverse events.
Mechanistic	Defining pathways associated with therapeutic mechanism of action or therapeutic resistance.	Methylation assays for HDAC inhibitors.

BRCA, BReast CAncer; CA, cancer; CEA, carcinoembroyonic antigen; EGFR, epidermal growth factor receptor; Her/Neu, human epidermal growth factor receptor 2/Neu protooncogene; LAG3, lymphocyte activation gene 3; LDH, Lactate dehydrogenase; MSI, Microsatellite instability; PD-1, Programmed Death 1; PD-L1, Programmed Death-Ligand 1; PSA, prostate specific antigen; PSMA, prostate specific membrane antigen; scRNA-seq, single cell ribonucleic acid sequencing.

Immunotherapy clinical trials present unique challenges that beg for a more thoughtful and systematic approach to biomarker testing. A successful antitumor immune response is dependent on the convergence of the immune repertoire, emergent tumor antigens, and the balance of pro-inflammatory and anti-inflammatory signals from the cancer cells, the host cells within the tumor microenvironment, and the patient’s systemic and secondary/tertiary nodal sites. A new paradigm of bedside-to-bench research is quickly emerging as we seek to understand antitumor immune responses that are dynamically influenced by interactions among (1) tumor cells (eg, tumor mutational burden, Lactate dehydrogenase (LDH) levels2 3; (2) the tumor microenvironment (eg, immune checkpoint molecule expression, lymphocyte count2 4; and (3) host factors (eg, patient microbiome) (see table 2). Well-established and emerging biomarkers provide valuable snapshots of these signals. While many of these tests are useful as stand-alone predictive and/or prognostic tests, the lack of standardized data collection and reporting has limited our ability to establish the nuanced integrated interpretations likely required to optimize cancer therapy (eg, through meta-analyses and AI). Given the burden biospecimen collection places on our patients, physicians, and staff, biomarker studies must be prioritized based on the anticipated value of the resulting data. Recommendations to guide uniform strategies for specimen collection, biomarker prioritization, and data reporting can address these challenges from a scientific perspective.

Table 2.

Examples of biomarkers that detect features of cancer cells, the tumor immune microenvironment, and systemic factors impacting immunotherapy response

Tumor tissue biomarkers		Patient-level/systemic immunity and tumor burden
Cancer cells	Tumor immune microenvironment
Pathology (eg, histology, grade)9	Checkpoint molecule expression (eg, PD-L1, Lag3)10 11	Patient demographics (age, sex, etc)
Targetable driver mutations (eg, Her2/Neu, EGFR)	Tumor-infiltrating immune cells, dendritic cells12	Tumor stage
Tumor mutational burden,11 13 Microsattelite Instability, MSI status	Interferon gamma signature	Metabolic studies/nutritional status (serum albumin, lactate, etc), microbiome14
Serum tumor markers (eg, CA19-9, CEA, CA-125, Prostate Specific Antigen, PSA)15	Radiomics8	Neutrophil lymphocyte ratio; Systemic Inflammatory Immune Index16
Circulating tumor DNA17	Digitized Imaging with machine learning outputs18	Lactate dehydrogenase
Targetable antigen expression (eg, CD19, CD20, mesothelin, Prostate specific membrane antigen, PSMA)	Single Cell RNA sequencing, multiomics19	Peripheral blood immune profiling (eg, T-cell repertoire,20 21 T-cell exhaustion)
		Stool microbiome 12

CA, cancer; CEA, carcinoembroyonic antigen; EGFR, epidermal growth factor receptor; Her2/Neu, human epidermal growth factor receptor 2/Neu protooncogene; LAG3, lymphocyte activation gene 3; LDH, Lactate dehydrogenase; MSI, Microsatellite instability; PD-L1, Programmed Death-Ligand 1; PSA, prostate specific antigen; PSMA, prostate specific membrane antigen; scRNA-seq, single cell ribonucleic acid sequencing.

The role of biomarkers in an evolving regulatory environment for immunotherapy clinical trials

The US FDA has recently launched two major initiatives to guide the design of clinical trials. Project Pragmatica (https://www.fda.gov/about-fda/oncology-center-excellence/project-pragmatica) seeks to improve the operational efficiency and patient centricity of clinical trials by enhancing the flexibility of trial design and aligning trials more closely with standards of routine clinical practice. In the face of efforts to simplify clinical trials, requirements for biospecimen collection and biomarker testing must be clear and well-justified. Project Optimus (https://www.fda.gov/about-fda/oncology-center-excellence/project-optimus) focuses on modernizing practises for optimal dosing of novel therapies based on consideration of biological efficacy as well as safety and tolerability. This represents a major paradigm shift from the focus on the maximum tolerated dose that dominated oncology practise for the development of cytotoxic and molecular targeting therapies. Recent studies suggest that the most biologically active dose of immunotherapy agents can be much lower than the maximum tolerated dose.5 Biological activity can be estimated from pharmacokinetic and pharmacodynamic biomarkers that measure the presence/persistence and activity of immunotherapeutic agents in tumor tissue. The routine incorporation of these studies into early immunotherapy clinical trials is essential for successful therapeutic development.

Prioritizing correlative biomarkers for clinical trials

We propose grouping biomarkers into three priority levels based on the strength of evidence, breadth of relevance, and feasibility of testing. Level one biomarkers should be highly informative about a fundamental question of cancer immunology, with a well-validated assay, to justify inclusion in all immunotherapy trials (eg, established biomarkers). One would have to justify not including a Level one marker in a trial. Level two biomarkers will have early evidence that supports inclusion dependent on suitable funding and scientific validity (eg, emerging biomarkers). Level three biomarkers are of specific importance for individual trials.

A call to action: learning something from every immunotherapy clinical trial

It is imperative that we learn something from every immunotherapy clinical trial.6 7 Enthusiasm for rapidly advancing novel agents must be balanced with humility and caution as we perturb a complex and powerful immune system that can cure cancer as well as mediate life-threatening autoimmunity. The design of highly impactful clinical trials is a science, an art, and a moving target. As a community, we can offer leadership by developing an evidence-based resource to guide biomarker incorporation into immunotherapy clinical trials. In this document, we have outlined an approach to prioritization. We invite feedback from the SITC and JITC readership community as we seek to bring the highest-value biomarkers into focus.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

Not applicable.