The molecular tumor board—a key element of precision oncology

Cancer is a highly heterogenous disease—encompassing a large variety of clinical presentations, biological behaviors, treatment responses and outcomes within the same malignant disease. The introduction of tumor profiling with next generation sequencing (NGS) opened up opportunities for individualized, patient-tailored cancer care and is the main catalyst of precision oncology.

Nevertheless, standardized interpretation of large amounts of partly highly complex molecular results and their translation into the clinical context remains challenging [1].

Molecularly focused interdisciplinary meetings—molecular tumor boards (MTBs)—have been established internationally in many academic institutions to meet this challenge. They have become a valuable tool in converting complex molecular data into clinical cancer care [2]. Cancer patients, especially in later-line treatment settings, have been shown to benefit from the MTB discussions: MTBs were able to provide molecularly driven treatment recommendations in a notable number of MTB patients with metastatic disease after several lines of standard of care (SOC) treatment [3] and have led to improved progression-free survival in this setting [4].

However, due to the fairly recent introduction of the MTB format and its unique character, when compared to organ-system-centered tumor boards, MTBs are currently not conducted in a standardized way across institutions [5].

In this short review, we address key questions regarding MTB structure and set-up, highlighting aspects of interest and challenges ahead based on the current literature and our experience at the Comprehensive Cancer Center Zurich (CCCZ).

The aims of an MTB are to provide a multidisciplinary approach to potential therapeutic strategies, based on molecular characteristics of a patient’s tumor. This includes identification of potential treatment targets, interpretation and discussion of treatment strategies based on these specific alterations in order to provide a personalized treatment recommendation. Here, an MTB can provide recommendations on choosing between clinical standards (in-label use), off-label use and allocation to a trial. Furthermore, MTBs also serve an important role in the diagnostic process in cancers of unknown primary and can also identify significant pro-oncogenic germline alterations. In doing so, the MTB aims to be beneficial to a broad spectrum of patients across cancer entities and to provide treatment recommendation in a clinically useful timeframe.

The cases undergoing MTB discussion are selected following local guidelines and all cancers are potentially eligible for MTB discussion. However, one main patient group that is usually discussed in the MTB are patients who have undergone SOC treatments and are in need of further systemic treatment (beyond SOC scenario). Here, the molecular assessment and the MTB discussion aim for the identification of a targetable alteration to facilitate a further treatment line.

Also, rare cancers are frequently discussed in the MTB. Rare malignancies collectively represent a notable amount of cancers, often lacking standardized therapeutic options. Here, next generation sequencing (NGS) and MTB discussion have become an integrative part of the diagnostic and therapeutic approach at many cancer centers.

However, molecular profiles of other patient groups pose questions beyond the identification of targetable alterations and are also addressed in the MTB:

With NGS having become a part of clinical routine, also longitudinal testing in form of new (liquid or solid) biopsies with molecular assessments, e.g., at time of tumor progression, have become more and more frequent. This offers insight into tumor evolution and clonal response to therapy, which is of great importance especially in the context of resistance mechanisms.

Standardized NGS panels by different providers are commonly used and allow comparability, reproducibility and result interpretation across institutions. Also, custom-made panels are employed, which, however, can harbor challenges in case of cross-institutional case discussion due to unique technical specifications which are a crucial part of the result interpretation.

Currently, NGS panel selection largely follows local guidelines. A standardization across institutions would facilitate cross-institutional case discussions and shared MTBs and could potentially enable and simplify data exchange even on an international level. However, in addition to different infrastructures, differences in health insurance and cost coverage as well as differing availability of drugs in different countries also remain challenges.

In general, appropriate diagnostic approaches should be chosen to answer relevant questions for a specific disease in question while also considering turn-around time and financial aspects. A combination of NGS panels of different sizes, RNA-based assays and immunohistochemistry can be used to assess the relevant genomic and functional properties of a tumor. While, for example, a smaller NGS hotspot panel might allow a faster return of results it might not include all variants relevant for treatment selection. This can be reflected in a bespoke testing algorithm for specific cancer types (e.g., second step comprehensive NGS in colorectal cancer after results of RAS and BRAF wildtype status in small NGS panel).

At our institution, diagnostic algorithms for specific cancers have been implemented following guidelines, current evidence and local multidisciplinary assessment by disease experts. The algorithms comprise suitable NGS panels, at this time for example Foundation One (Foundation Medicine, Inc. (FMI), Cambridge, MA, USA) or the Oncomine Precision Assay (OPA; Thermo Fisher, Waltham, MA, USA), and—depending on the entity—in situ hybridization and/or immunohistochemical stainings. The algorithms undergo regular review and are constantly revised to reflect new SOC and potential treatment options.

New sequencing panels are constantly developed and need to be reviewed by MTB teams to include technological advances into the molecular assessment and the MTB workstream. One technological advancement of great interest is whole exome sequencing (WES), which might become available on a broader level. Assessing integration of WES data into the MTB highlights the importance of the bioinformatics pipeline annotation, facilitating interpretation of relevant molecular findings for individual patients. Furthermore, turn-around time and financial sustainability will influence the utility of WES in the MTB setting.

An MTB is multidisciplinary discussion involving a diverse group of healthcare professionals with expertise in various disciplines, as first described in 2014 [6]. This includes medical oncologists, pathologists, molecular biologists, human geneticists, bioinformaticians and other specialist relevant to the specific case. The participation of experts of these various disciplines allows for a collaborative and comprehensive approach to analyzing and interpreting the molecular results, to identification of treatment strategies and to making informed treatment recommendations.

Larger academic institutions are ideally positioned to facilitate the multidisciplinary approach required for meaningful interpretation of molecular results. Physicians treating cancer patients at institutions that do not comprise the multitude of disciplines mentioned above can present patient cases for interpretation of NGS results at MTBs of a near-by or relevant larger academic hospital. Thus, MTBs as a valuable tool of cancer care are available for the broad cancer patient community.

For some cancer types, substantial progress in molecularly driven treatment was achieved leading to clear therapeutic strategies based on specific oncogenic drivers, e.g., in CML (BCR-ABL) or ALK and EGFR positive NSCLC (ALK rearrangement; classic EGFR alterations). Based on recent evidence collected from OnkoKB, a curated knowledge base of molecular alterations, oncogenic effects and treatment implications, up to 37% of all cancer patients harbor at least one potentially targetable alteration [7]. However, interpretation of molecular alterations can be challenging and judging pathogenicity and actionability of a genomic alteration is complex and challenging [8]. Once a molecular analysis of a biological sample is conducted, findings are compared to a reference genome and are first assessed for reproducibility and quality and second for functional relevance. Bioinformatics pipelines are used to facilitate this large-scale review of data and allow a multilevel review of detected molecular findings before they are released for further discussion and interpretation in the MTB. Furthermore, these software tools can also allocate clinical trials potentially fitting to a molecular alteration found in the specific tumor sample.

The clinical interpretation of molecular results in the MTB comprises available insights into cancer driving oncogenes as well as potential resistance mechanisms. Result interpretation should be based on the highest level of currently available evidence and incorporate the regularly updated standardized international classification systems ranking levels of clinical actionability of molecular alterations, such as (i) the European Society for Medical Oncology (ESMO) scale for clinical actionability of molecular targets (ESCAT), which distinguishes six main levels of actionability based the type of evidence available; (ii) the US-based OncoKB (precision oncology knowledge base) that distinguishes four levels of actionability and two levels of resistance, and, next to the available evidence levels also incorporates US Food and Drug Administration (FDA) approval status.

Incorporation of this standardized actionability assessments facilitates evidence-based medicine and enables the treating physician to assess the MTB recommendation with some ease when evaluating an optional treatment for their patient.

The MTB discussion is reflected in a comprehensive report containing the patient’s identification, clinically relevant details, pathology results, the molecular profile and its interpretation in the clinical context, allowing a concise recommendation.

Molecular parameters discussed at the MTB are—depending on the conducted molecular assessment: (i) point mutations/copy number alterations/fusions and rearrangements; (ii) genomic signatures, such as tumor mutational burden (TMB), loss of heterozygosity (LOH), microsatellite status, and homozygote repair deficiency (HRD).

If the molecular assessment introduces a potential rationale for a certain treatment within a clinical trial, the National Clinical Trial (NCT) number, a unique clinical trial identifier, which is officially assigned after release of the trial protocol, is an integral part of the MTB report.

Furthermore, additional aspects such as potential germ line mutations detected in context of the NGS are considered and, depending on the findings, genetic counselling is recommended.

MTBs are a crucial tool of precision oncology and have become an integrative part of modern oncology, translating molecular results into clinical care.

Among many potential future developments within the MTB format, structured data collection and integration of new technologies are two opportunities to be mentioned:

With the increasing amount of NGS conducted, the structured collection of clinical and molecular characteristics, MTB treatment recommendations and patient outcomes becomes an unparalleled opportunity to create insight into molecular phenotypes on a large level [2]. Data exchange and cross-institutional collaboration could allow insights into molecular phenotypes and their clinical outcomes, providing useful information for patient care and, thus, potentially enabling improved patient outcomes. However, data collection in a structured format is crucial to facilitate data exchange and collaboration can be enabled further by using standardized coding systems for data collection (e.g., snomed, mcode).

Furthermore, new tumor-profiling technologies are under development and comprehensive multi-omic approaches assess whether in-depth analysis of individual cancers could enhance treatment recommendations for cancer patients [9]. We expect an introduction of new tumor-profiling technologies into the clinical routine in the upcoming years and MTBs will be the institutional tools to integrate them based on a strong linkage and collaboration with the scientific technology experts.

MTBs, therefore, remain a center-piece of precision oncology and have an important role fueling scientific progress in this field due to their unique interdisciplinary and forward-looking nature.