The value of virtual molecular tumor boards for informed clinical decision-making

Abstract

Genomic profiling and other new technologies have increased the volume and complexity of information available for guiding clinical decision-making in precision oncology. Consequently, there is a need for multidisciplinary expert teams, in the form of molecular tumor boards (MTBs), who can translate this information into a therapeutic plan, including matching patients to suitable clinical trials. Virtual MTBs (vMTBs) can help to overcome many of the challenges associated with in-person MTBs, such as limited time availability, access to appropriate experts or datasets, or interactions between institutions. However, real-world experience from vMTBs is lacking. Here, we describe oncologists’ vMTB experiences and the value of working with multicenter and/or multinational vMTBs. We also address knowledge gaps and barriers that could affect the implementation of vMTBs in routine clinical practice. Case studies from Argentina, Turkey, and Portugal illustrate the value of informed clinical decision-making by vMTBs, including expansion of therapeutic options for patients, faster time to treatment, and the resulting improvement in patient outcomes or impact of vMTB discussions on patients. With the uptake of comprehensive genomic profiling and the evolution of some cancers now being conceptualized as a collection of rare diseases with small patient populations based on molecular profiling, the importance of MTBs has increased in modern cancer management. However, an adjustment in clinical decision-making by healthcare professionals is required and evidence of the added value of vMTBs is lacking. Existing vMTBs and recommendations from participating oncologists could point toward a structured evaluation and analysis of this new platform.

This review focuses on the value of virtual molecular tumor boards for informed clinical decision-making in the care of patients with cancer.

Implications for Practice.

This review focuses on the value of virtual molecular tumor boards for informed clinical decision-making in the care of patients with cancer. As precision medicine has led to an increase in the volume and complexity of information available for guiding clinical decision-making, virtual molecular tumor boards provide an opportunity to overcome many of the challenges associated with in-person molecular tumor boards. Real-world experience, including case studies from Argentina, Turkey, and Portugal, illustrates the value of this type of intervention in modern cancer management, and could help to strengthen the virtual molecular tumor board platform and further improve patient outcomes.

Introduction

An ongoing revolution in cancer care is aiming to improve treatment outcomes for patients with cancer via precision and personalized medicine.¹ This relies on the generation, collection, integration, and accurate interpretation of complex clinical information originating from various sources. Recent developments have greatly increased the volume of information available for guiding cancer care, including next-generation sequencing of solid/liquid tumor biopsies, digital pathology, modern real-world data registries, and integration of artificial intelligence for data analysis (Figure 1).

Figure 1.

The role of complex clinical information generation in increasing data volume in cancer care.

Next-generation sequencing-based comprehensive genomic profiling has become an integral part of precision oncology, in which tumor-specific genomic information (insertions and deletions, gene rearrangements, copy number alterations, substitutions, and genomic signatures) enables decision-making regarding molecularly guided therapeutics.² The US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) have already approved numerous molecularly guided therapeutics for specific cancer types, thereby expanding treatment options for patients.² Some molecularly guided therapeutics have also demonstrated clinical activity across multiple tumor types that share the same molecular alteration,³ resulting in broad and tumor-agnostic approvals.^4-8 Importantly, real-world evidence has been used increasingly to inform regulatory decisions as some cancers have evolved from being conceptualized as single diseases to a collection of rare diseases based on molecular profiling.² However, the main challenge relates to the integration and interpretation of complex datasets for evaluating which patients can benefit from these treatments.

The volume and complexity of the information generated from comprehensive genomic profiling and other new technologies complicates clinical decision-making in oncology and has led to a requirement for multidisciplinary molecular tumor boards (MTBs), during which experts can interpret this information to help define a treatment plan that ensures the best possible patient outcomes. These MTBs integrate molecular information, imaging and pathologic data, patient characteristics, and knowledge of available treatment options and corresponding evidence levels, as well as real-world evidence, into a therapeutic plan for the patient, including those who have exhausted their therapy options or who present with complex genomic profiles.⁹ MTBs can also assess patient suitability for clinical trials of novel molecularly matched therapies based on their genomic information, although this may be limited by the choice of clinical trial registry for matching patients against (eg, clinicaltrials.gov may exclude patients outside the US).

Although MTBs have traditionally been conducted in person, this can present challenges, such as: experts having limited time available to attend on-site meetings; smaller cancer centers or individual healthcare professionals not having access to the relevant experts, or to real-world evidence and other large datasets to enable clinical decision-making; limited interaction between centers both nationally and internationally; and COVID-19-related social distancing measures or travel/meeting restrictions. Virtual MTBs (vMTBs) allow meetings to take place remotely (or to use computational systems to obtain feedback from medical and scientific experts)¹⁰ and can help to overcome many of the challenges associated with in-person MTBs. Given that many oncologists became increasingly involved in delivering cancer care to patients through virtual clinics during COVID-19, vMTBs represent a method of collaborating with other experts that oncologists are already familiar and comfortable with.

Here, we describe oncologists’ experiences of working with multicenter and/or multinational vMTBs and highlight how vMTBs may help with clinical decision-making, the value of vMTBs to local healthcare systems, and other relevant learnings that may inspire oncologists to become involved in the implementation of vMTBs and benefit from their educational aspects. We aim to illustrate the value of informed clinical decision-making by vMTBs with specific examples of their tangible impact in clinical cancer care, including expansion of therapeutic options and faster time to treatment. We will also discuss knowledge gaps and barriers that could affect vMTB implementation in routine clinical practice.

Oncologists’ experiences and learnings from vMTBs

An overview of the vMTB environment is shown in Figure 2. Details of individual case studies discussed during vMTBs, the recommended treatment plans, and the outcomes and impact of the vMTBs on patients are provided in Table 1.

Figure 2.

Oncologists’ experiences and learnings from participating in vMTBs.

Table 1.

Details of case studies.

Country	Case	Driver mutations	Treatment decision	Outcome
Argentina	• 53-year-old female never smoker • Nonspecific abdominal pain and weight loss • Abdominal US showed multiple liver lesions; PET/CT scan shows 41 mm x 20 mm mediastinal mass • Percutaneous liver biopsy confirmed squamous cell carcinoma-of-unknown-primary-origin (IHC: TTF-1- and synaptophysin-negative; Ki67: 60%; IC PD-L1: 30%) • Three cycles of carboplatin + paclitaxel given with no clinical or radiologic response • Second-line treatment with doxorubicin + cyclophosphamide + cisplatin with stable disease • Comprehensive genomic profiling	• MSI status undetermined • KIT D820H; VAF 24.2% • CDKN2A p16INK4a D108fs*14; p14ARF R122fs*41; VAF 0.31% • VUS: GNAS P292S	• Avapritinib started under a compassionate use program	• Patient developed painful hepatomegaly and moderate ascites • After 20 days of treatment, pain and ascites decreased • After 3 treatment cycles, dose reduced due to grade 2 liver toxicity and diarrhea • Disease progression and death 2 months later
Turkey	• 68-year-old female • Transverse colon mass without metastasis • T4 N1 M0 G3, lymphovascular invasion-positive • Six-course adjuvant XELOX administered • Two masses detected in lung 8 months after completion of treatment • Metastasectomy performed on one lesion; SBRT on another; mutation status obtained from lung lesion	• MSS • TMB = 4 • KRAS, NRAS wt • CTNNB1 N426D (subclonal) • FBXW7 R689W • APC R876 and Q1328 • BRAF D594G • TP53 splice site 994-40_994-2del39	• Cetuximab + trametinib	• Radiologic complete response and no ctDNA detected with SignateraTM minimal residual disease assay (Natera, Austin, TX) after 16 months’ follow-up
Portugal	• Infantile desmoplastic ganglioglioma revised to embryonal tumor with CIC-NUT mutation • vMTB provided important discussion with information integrated from FMI and IHC from Barcelona Hospital • Due to the rearrangement and other information, grading changed from benign to high-grade disease • Decision made that BSC was the best option • Parents were given clarity on the case and decision	• CIC-NUTM1 rearrangement	• Treatment upgrade to Sarcoma Protocol	• Patient died

Abbreviations: BSC, best supportive care; CT, computerized tomography; ctDNA, circulating tumor DNA; EAP, expanded access program; FMI, Foundation Medicine, Inc.; IC, immune cell; IHC, immunohistochemistry; MSI, microsatellite instability; MSS, microsatellite stable; (v)MTB, (virtual) molecular tumor board; PD-L1, programmed cell death-ligand 1; PET, positron emission tomography; SBRT, stereotactic body radiation therapy; TMB, tumor mutational burden; TTF-1, thyroid transcription factor 1; US, ultrasound; wt, wild type.

Argentina: unveiling carcinoma-of-unknown-primary-origin and utilization of compassionate use programs

A patient with squamous cell liver carcinoma-of-unknown-primary-origin received first-line treatment with carboplatin and paclitaxel and second-line treatment with doxorubicin, cyclophosphamide, and cisplatin. After 5 cycles of second-line treatment, the patient experienced disease progression in the liver and underwent comprehensive genomic profiling, based on the recommendation of the institutional MTB. The results were then assessed by the vMTB to allow input from a broader range of experts, who also considered the patient’s limited response to previous histology-based therapies. The vMTB’s interpretations complemented the diagnosis by identifying the c-Kit mutation usually found in gastrointestinal stromal tumors. Using their knowledge of newly available tyrosine kinase inhibitors that specifically target c-Kit, the vMTB helped to identify compassionate use programs. The patient was subsequently placed on avapritinib treatment via one such program. After 20 days, pain and ascites decreased. Three treatment cycles were completed before the patient died.

In Argentina, the only drugs available at the time were sunitinib, imatinib, or sorafenib. The cross-border collaboration facilitated by the vMTB helped to identify other drugs that were available through compassionate use programs already ongoing in other Latin American countries, such as Paraguay.

Turkey: understanding molecular mechanisms of disease may facilitate drug access

Following completion of adjuvant capecitabine and oxaliplatin for transverse colon cancer, a patient presented with multiple lung metastases, one of which was treated with metastasectomy, one with stereotactic body radiation therapy, and 2 that could not be reached due to their paracardiac localization. The patient’s mutation status was established from the lung metastases.

The vMTB discussed the mechanistic details of the different types of BRAF mutations. The patient had a type 3 BRAF (D594G) mutation. Tumors with type 3 BRAF mutations have low or no kinase activity and they signal in a RAS-dependent manner.¹¹ Due to this mechanism, the vMTB recommended a treatment strategy that would inhibit the pathway both upstream and downstream of RAS. The patient was switched from treatment with FOLFIRI (folinic acid, fluorouracil, and irinotecan) and cetuximab (an EGFR inhibitor) to treatment with cetuximab plus trametinib (a MEK inhibitor). Access to this drug was granted based on information obtained through discussions within the vMTB. The patient’s quality of life improved due to the more favorable toxicity profile; the patient also achieved a radiologic complete response. Another patient with stage IV cholangiocarcinoma bearing a type 3 BRAF mutation also achieved a near-complete response following treatment with cetuximab and trametinib. Based on the data from these 2 patients and additional vMTB explanations, the Turkish Medicines and Medical Devices Agency approved and reimbursed cetuximab and trametinib treatment of cholangiocarcinoma with BRAF type 3 mutations.

With regards to the patient with type 3 BRAF-mutated colon cancer, the treating physicians had limited knowledge of the different types of BRAF mutations before the case was discussed with the vMTB. The decision by the treating physician to switch treatment strategies has led to this patient being disease-free for the last 3 years and treatment-free for the last year. Participation of Turkish healthcare professionals in the vMTB has helped to provide a model for the formation of local MTBs with regards to developing constructs for submitting cases, slide preparation, case summaries, sequence of discussions, time management, and the need for bioinformatics, geneticists, clinical cancer scientists, biologists, and physicians with sufficient genomic experience. It has also improved healthcare professional awareness of precision oncology both across Turkey and in other countries, whose patients are now able to travel to Turkey for testing and treatment.

Portugal: adjusting the diagnosis in pediatric patients and deciding not to treat

The vMTB has provided important integrated discussion of the clinical and molecular aspects regarding a patient with infantile desmoplastic ganglioglioma revised to embryonal tumor with CIC-NUTM1 gene rearrangement. Due to this rearrangement and other information obtained from immunohistochemistry, the tumor grading was changed from benign to high-grade and the recommendation for treatment was best supportive care. Although the decision was made not to explore further treatment options, the patient’s parents were given clarity on both the patient’s condition and the treatment decision. This case is paradigmatic of the importance of genetic analysis, vMTB involvement, and timing. In this case, the vMTB served to restrict treatment options and prevent the patient from being exposed to unnecessary toxicities from conventional chemotherapy.

In the case of central/peripheral nervous system tumors, where treatment delays can result in major neurologic deficit, efficient decision-making is critical. Discussions with the vMTB were critical in gaining the expertise for developing a treatment plan for that patient and any future patients with a similar genetic landscape. CIC-NUT-mutated tumors are rare, so access to information and the ability to share experiences with the vMTB was useful. Misdiagnosing the desmoplastic ganglioglioma case led to the family struggling to accept the diagnosis due to the change from benign to high-grade disease. This ability of the vMTB to clarify diagnoses, define prognostic implications, and offer advice on treatment plans strengthens the case for multidisciplinary teams to be involved in diagnosis and treatment decisions as early as possible, particularly for rare tumors. As the vMTB is in collaboration with Israel, it also supports connectivity with other countries as a method of bridging knowledge gaps and providing new insights into the challenges being faced.

Outlook and conclusions

MTBs are integral in defining and substantiating personalized treatment plans for patients. Their value in guiding clinical decision-making will likely grow with the increased uptake of comprehensive genomic profiling for patients with cancer in routine clinical practice. Given the volume of information available for guiding treatment decisions, it becomes challenging for clinicians to access and keep up with developments in related scientific evidence. MTBs play an important role in interpreting molecular testing results in patients with advanced cancers. Many studies have shown improvements in oncologic endpoints for patients whose cases were discussed in MTBs.^12,13 Comprehensive genomic profiling reports provide a list of genetic mutations and genomic signatures (eg, tumor mutational burden and microsatellite status) found within a tumor tissue or liquid biopsy sample. However, they don’t provide information on associations between genomic signatures and multiple, non-driver-like mutations. Discussions of these associations in MTBs has expanded treatment options for patients. MTB evaluation can provide information regarding the primary tumor source and etiologic treatments for patients with carcinoma-of-unknown-primary-origin, or can direct patients to appropriate clinical trials. A wider benefit of regular and consistent recommendations from MTBs is the increased use of more personalized treatments, which has economic benefits for the healthcare system and health benefits for patients.

Conducting MTBs virtually has the advantage that oncologists can attend at anytime from anywhere in the world to discuss cases with experienced molecular oncologists, which benefits patient health and improves clinician education. The ability to participate in multiple vMTB sessions in different countries also enables international input on both molecular oncology and on the healthcare system. Joint scientific studies can be completed faster with teamwork, and rare diseases can be evaluated on a global scale. Larger centers already integrate genomic and pathologic information because they include molecular and clinical teams that work closely together. vMTBs will enable a similar process to be carried out in smaller centers, thereby improving patient care.

Several recommendations have been proposed to optimize the implementation of vMTBs, including harmonization of sequencing techniques, definition of the minimum expert panel and functional requirements, and policy development for dealing appropriately with unsolicited findings.¹⁴ To overcome local barriers to vMTB implementation, it may be necessary to prove their added value in clinical decision-making in a more structured way via comparative analyses. However, this would require consideration of suitable comparator arms for vMTB-led outcomes (eg, other local vMTBs or absence of a vMTB). Consistency is also important when bringing specialists together for a vMTB, to ensure that discussions are aligned across similar patients. An important factor in the success of vMTBs is consideration of the level of expertise required or which experts should participate (Figure 3). For example, interpretation of molecular reports obtained through comprehensive genomic profiling reports is complex and requires understanding of gene interactions from experienced genomicists, which can change treatment strategies for patients and thus gives genomicists a critical role in vMTBs.

Figure 3.

Overview of specialist participation in vMTBs. ^aDepending on the cases discussed, a vMTB is likely to comprise a mixture of specialists and, in the case of vMTBs that discuss large-scale sequencing results, a minimal team is recommended.¹⁴

The main limitation of vMTBs is that they can only discuss a small number of cases per month. Therefore, even if genetic tests were carried out routinely on large numbers of patients, assistance from vMTBs for interpreting results cannot always be guaranteed. Cases for discussion at vMTBs must be selected carefully based on urgency, complexity of genomic findings (eg, tumors with BRAF type 3 mutations), and their wider impact on other patients within the oncology community. Other limitations may include: language barriers when participating in international vMTBs; the time and money required to prepare reports and to perform the necessary tests (although this becomes easier and quicker over time as healthcare professional experience of vMTBs increases); the ability of healthcare professionals to meet deadlines for submitting reports, given their time constraints; the lack of basic education of healthcare professionals with regards to genomic aberrations; and the need for physicians to want to be involved and to incorporate genomic profiling and vMTBs into routine clinical practice.

vMTBs may sometimes be inconsequential, due to unrealistic expectations, incorrect choice/prioritization of cases, or the limits of the vMTB to provide appropriate answers to clinicians. However, successful vMTBs can be used as examples of best practice to inform different institutions on how to evolve their own vMTBs by adopting the most suitable model for these types of complex analyses.

vMTBs are generally recognized as supportive tools for navigating the complexity of the increasing levels of clinically relevant information and informing subsequent clinical decision-making. vMTBs are particularly relevant following the increasing uptake of comprehensive genomic profiling in patients with cancer, and complementary analysis from vMTBs can benefit any case where there is no clear correlation between genomic aberrations and corresponding therapy options. The exchange of experience in vMTBs (education of clinicians about tumor genetics and molecular biologists about the patient) is a dynamic learning process that is expected to improve cancer care. Over time, vMTBs could further evolve by using artificial intelligence-supported systems, for example, to improve completeness of real-world data collection and estimations of efficacy outcomes.^15,16

Author contributions

All authors: Conception and design, provision of study material or patients, collection and/or assembly of data, data analysis and interpretation, manuscript writing, and final approval of manuscript.

Funding

This work was supported by F. Hoffmann-La Roche Ltd, Basel, Switzerland (no grant number). Support for third-party writing assistance for this manuscript, furnished by Katie Wilson, PhD, of Health Interactions, was provided by F. Hoffmann-La Roche Ltd.

Conflicts of Interest

Martín Angel: Honoraria: Bristol Myers Squibb. Other (travel, accommodation, expenses): Bristol Myers Squibb, F. Hoffmann-La Roche Ltd. Research Funding: F. Hoffmann-La Roche Ltd (funding for third-party medical writing support for this manuscript). Mutlu Demiray: Research Funding: F. Hoffmann-La Roche Ltd (funding for third-party medical writing support for this manuscript). Umut Dişel: Consulting/advisory relationship: P2Padvice.org. Honoraria: Infogenetik, Roche Türkiye. Research Funding: F. Hoffmann-La Roche Ltd (funding for third-party medical writing support for this manuscript). João Passos. Consulting/advisory relationship: AstraZeneca/Alexion, F. Hoffmann-La Roche Ltd. Honoraria: AstraZeneca/Alexion. Other (travel, accommodation, expenses): AstraZeneca/Alexion. Research Funding: F. Hoffmann-La Roche Ltd (funding for third-party medical writing support for this manuscript). No other potential conflicts of interest were reported.

Data availability

No new data were generated or analyzed in support of this research.