Abstract
PURPOSE
With the advent of precision medicine, molecular tumor boards (MTBs) were established to interpret genomic results and guide decision making for targeted therapy in oncology patients. There are currently no universal guidelines for how MTBs should operate and thus variance can be seen depending on which MTB is reviewing the case. This study assesses the concordance of MTB recommendations when a participant case is reviewed by two different MTBs, establishes potential reasons for discordance, and advocates for the establishment of standard MTB operating guidelines.
PATIENTS AND METHODS
Participants with advanced cancer, who had exhausted all standard treatment options were screened for the Targeted Agent and Profiling Utilization Registry (TAPUR) Study. Cases were submitted for MTB review if the treatment proposal was outside the protocol genomic matching rules, or if multiple treatment options were identified. Of the 306 cases submitted for review by the TAPUR MTB from 2016 to 2018, 107 were randomly selected for secondary review by a different MTB group. Recommendations from the original review were not disclosed. Concordance between MTB group recommendations was assessed. Concordance was defined as agreement between MTB reviews on the genomic alteration and study drug match proposed by the clinical site. Thematic qualitative analysis was conducted for the discordant cases to assess reasons for discordance.
RESULTS
Complete or partial concordance was observed in 79% of cases (95% CI, 70 to 86; one-sided P = .25). Most discordant analyses were due to disagreements on the strength of evidence regarding efficacy of the proposed treatment (32%).
CONCLUSION
When presented with identical participant cases, different MTB review groups make the same or similar treatment recommendations approximately 80% of the time.
Over 300 cases have been submitted to TAPUR Study molecular tumor board (MTB). When presented with the same study participant cases, different MTB review groups make the same or similar treatment recommendations most of the time in the context of a precision medicine study for patients with advanced cancer.
INTRODUCTION
Although there is evidence that molecular tumor board (MTB) treatment recommendations contribute positively to patient outcomes,3 there are currently no established operational guidelines for MTBs. The lack of operational guidelines may contribute to variation in MTB recommendations generated by different groups of reviewers analyzing the same case. The Targeted Agent and Profiling Utilization Registry (TAPUR) Study established an MTB to assist treating physicians in identifying potential treatment options within the TAPUR Study. The TAPUR MTB is composed of nearly 70 committee members with diverse backgrounds and expertise.
CONTEXT
Key Objective
Identify the consistency in molecular tumor board (MTB) recommendations between different MTBs reviewing identical cases.
Knowledge Generated
When different MTBs reviewed the same cases from the TAPUR Study, their identification of actionable genomic alterations and potential treatment options was fully or partially concordant in 79% of cases, which was not statistically significantly different from the null hypothesized concordance rate of 75% (one-sided P = .25).
Relevance
When presented with the same study participant cases, different MTB review groups make the same or similar treatment recommendations most of the time in the context of a precision medicine study for patients with advanced cancer.
The TAPUR Study is a phase II, prospective clinical trial that enrolls participants with advanced cancers harboring a genomic alteration known to be a target of, or to predict sensitivity to, at least one of the US Food and Drug Administration (FDA)-approved anticancer drug therapies available in the study.1 Eligible participants are matched to one of these treatments depending on the genomic alterations present in their tumor. The primary objective of the TAPUR Study is to assess the efficacy and safety of commercially available, targeted anticancer drugs used outside of their FDA-approved indications.1 Participants are placed into multiple parallel cohorts of up to 28 patients, defined by tumor type, genomic alteration, and matched study treatment. Cohorts are constructed and analyzed using a Simon's two-stage design. Participating clinical sites submit cases to the TAPUR MTB if there is uncertainty regarding the appropriate drug match.
There are many factors that may influence the outcome of an MTB review; a participant's case with the same treatment proposal might receive different recommendations depending on the expertise of the MTB reviewers present, on the accumulated knowledge at the time the MTB review occurs, and the treatments available on the TAPUR Study at the time of the MTB review. For example, new scientific evidence can either increase or decrease the evidence that supports the actionability of a specific alteration that may affect the MTB reviewers' recommendations. Since the TAPUR MTB convenes weekly but includes different experts at each meeting, we sought to determine if the MTB recommendation on the same case would be the same regardless of the composition of the MTB that reviews the case and when it is reviewed. Thus, cases originally reviewed at one MTB meeting were randomly selected and submitted to a different MTB meeting for secondary review. The purpose of this study was to determine the consistency of the recommendations between the original MTB review and the secondary review, and the factors that may contribute to lack of concordance between the two reviews.
MTB Structure and Review
The TAPUR MTB formed at the inception of the study in March 2016, with the intention to provide guidance to treating physicians on the interpretation of genomic test reports and the identification of potential treatments available in the study. The study created an MTB charter that includes guidelines for volunteers, as well as additional resources including but not limited to training materials, for example, Standards and Guidelines for the Interpretation and Reporting of Sequence Variants in Cancer,4 and TAPUR MTB Variant Determination Flowchart (Fig 1). Furthermore, a designated TAPUR staff member acts as the MTB coordinator assisting with overall project management.
FIG 1.
TAPUR MTB variant determination flowchart. MTB, molecular tumor board; TAPUR, Targeted Agent and Profiling Utilization Registry.
The TAPUR MTB is composed of clinicians from TAPUR sites with expertise in precision oncology, patient advocates, molecular pathologists, and pediatric oncologists. MTB meetings are held weekly and up to six cases are reviewed at each meeting. Each MTB meeting includes at least one pathologist, one patient advocate, and three medical oncologists.
For each patient registered, the treating physician at the TAPUR Study site is required to propose a study treatment for the patient based on local interpretation of the genomic test report. This proposal may be based on prespecified genomic matching rules in the protocol (which do not require additional MTB review). However, if multiple possible drug matches are identified or the possible drug match does not follow the protocol matching rules, the treating physician has the option to submit the treatment proposal to the TAPUR MTB. Furthermore, if the treatment proposal is not a direct match or if the proposed genomic target is reported as subclonal, equivocal, or a variant of unknown significance, then the participant's case must be presented to the MTB for review.
Requests for MTB review include the participant's tumor type, genomic test report, pathology report on the specimen tested, previous cancer-related treatment(s), and the treatment proposal for MTB consideration. Representatives of the clinical site are invited to present the participant's case to the MTB. In reviewing a case, the MTB may consult data available about the targeted alteration in publicly available genomic databases such as Catalogue of Somatic Mutations in Cancer. After review and discussion of the case, the MTB will then approve or deny the treating physician's treatment proposal or provide alternative recommendations.
After the MTB has concluded their review, a summary of the discussion and potential treatment options is sent to the submitting clinical site within 5-7 business days. The TAPUR site may proceed with enrolling the participant to an MTB-recommended treatment option if the participant meets all eligibility criteria.
PATIENTS AND METHODS
The primary objective of this MTB re-review study was to evaluate the concordance between the original MTB review and the secondary review as it relates to the treatment proposal submitted by the treating physician. The secondary objective was to review discordant cases and identify the reasons for disagreement between the original and secondary review meetings.
End Points
For the purposes of this study, full concordance was defined as the MTB recommendation from both the original and the secondary review agreeing regarding both the target alteration and the proposed study drug, either both approving the treatment proposal as an acceptable treatment option, or both disapproving the treatment proposal.
Partial concordance was defined as the original and secondary review agreeing on part of the treatment proposal, either both approving the treatment as an acceptable option but not the target alteration, or both agreeing the alteration was actionable but not agreeing on the proposed study treatment.
Discordance was defined as the original and secondary review disagreeing on their recommendations regarding both the proposed drug and targeted alteration with one meeting approving the treatment proposal as an acceptable treatment option and the other meeting denying the treatment proposal.
Study Design and Eligibility
All cases submitted to the MTB for review on or before 2018 were considered eligible. From those cases, a random sample of 107 cases were selected for secondary review. All MTB members were eligible to participate in the secondary review, based on their availability. When scheduling meetings, it was ensured that the cases had not previously been reviewed by any of the MTB members scheduled to participate.
MTB Re-Review Structure and Review
The same information submitted from the clinical site for the original MTB review was provided to MTB members performing the secondary review. Secondary MTB meetings setup slightly differed to the original MTB meetings. However, identical composition (ie, one pathologist, one patient advocate, and three medical oncologists) was not always possible. Clinical site representatives were not invited to secondary meetings and the clinical sites were not informed of the outcome of these meetings. At the start of each secondary review meeting, the MTB members were informed that they were participating in the secondary review of a participant's case but were not informed of the recommendations of the original review.
Statistical Analysis
Number and percentage of cases with full, partial, and discordant agreement are presented. For the primary analysis, full or partial concordance was considered concordance. A one-sided exact binomial test with alpha level of 0.05 was used to test that the concordance rate was >75%. Agreement rates as low as 50% and as high as 85% have been previously published.13,14 In general, previously reported data on agreement between MTBs average around 67%.13,14 The null hypothesis value of 75% was chosen to be higher than this, given that the previously published literature was not based on a clinical trial with prespecified genomic matching rules and possible treatment options defined in a protocol. This design, with a sample size of 107, has a power of 84% to reject the null hypothesis of 75% concordance when the true agreement rate is at least 85%. For the effective sample size (N = 89), the power is 74%. No hypothesis testing was performed for the secondary objectives.
Qualitative Analysis
Thematic qualitative analysis was conducted for the discordant cases to assess reasons for discordance. Two researchers (J.P. and J.K.) reviewed the MTB case reviews independently and identified keywords within each individual case summary that reflected common themes in the case summaries. After the individual review, the two researchers then met to discuss findings. If the researchers did not agree with the reason for discordance, another researcher was consulted until consensus was reached.
Human Research Statement
The Advarra institutional review board (IRB) is the IRB of record for the TAPUR Study and provided approval for this study. The participants in each re-review meeting were informed of the objectives and procedures of the substudy and, by virtue of their participation, provided their consent to participate.
RESULTS
A total of 1,551 participants were registered in the TAPUR Study between 2016 and 2018, of which 306 were reviewed by the MTB. Of the 107 participants randomly chosen from the 306, 89 participants had data available for the secondary comparison; 18 cases did not include a treatment proposal. The MTB secondary meetings began in 2018 and were completed in 2020.
Of the 89 cases that underwent secondary review, 70 cases (79%) were concordant in the recommendations of the two reviews; 62 (70%) cases were fully concordant, 8 (9%) were partially concordant, and 19 cases (21%) were discordant. The percentage of concordance was not significantly higher than 75% (P = .25).
There were six reasons identified for discordance in the two MTB reviews: cohort availability, alteration disagreement (actionability of alteration), concern for efficacy of drug, alternative treatment available on label, match approved despite concern for actionability of alteration, and match approved despite concern for efficacy of drug (refer to Fig 2). A summary of results and definitions is found in Table 1. The majority of the cases were discordant due to concern for efficacy of drug (32%) and alteration disagreement (actionability of alteration) (26%).
FIG 2.
Reasons for discordance.
TABLE 1.
Summary Table of Discordance Reasons for 19 Cases
| Primary Reasons for Discordance | No. (%) | Definitions |
|---|---|---|
| Concern for efficacy of drug | 6 (31.5) | Disagreement about the efficacy of the drug. Data are available to support both positions |
| Alteration disagreement (actionability of alteration) | 5 (26.3) | Disagreement over whether or not a particular alteration is actionable. Data are available to support both positions |
| Approved match despite concern for actionability of alteration | 3 (15.8) | Both groups agreed that the alteration was not actionable; however, one meeting approved the match, the other did not. They are approving despite the lack of supporting data. MTB permissiveness |
| Approved match despite concern for efficacy of drug | 3 (15.8) | Both groups agreed that the drug was not likely to be efficacious; however, one meeting approved the match, the other did not. They are approving despite the lack of supporting data. MTB permissiveness |
| Cohort availability | 1 (5.3) | The MTB took into account that the cohort was suspended at the time of MTB review |
| Treatment available on label | 1 (5.3) | Whether or not the MTB took into consideration the availability of an approved treatment that could be used on label, and whether or not it was available at the time of the meeting |
Abbreviation: MTB, molecular tumor board.
Of the 19 discordant cases, eight (42%) had the treatment proposal denied during the original MTB meeting and then subsequently approved during the secondary meeting. Conversely, 11 (58%) cases had the treatment proposal approved during the original MTB meeting and then subsequently denied during the secondary meeting. The average amount of time between the original and secondary meetings for discordant cases was 458 days and for concordant cases was 500 days. In all case summaries, the MTB noted relevant preclinical and clinical data as well as findings from review of genomic databases. None of the summaries for the discordant cases noted any newly published data or FDA drug approvals contributing to the MTBs recommendations. The initial case summary, secondary case summary, and reason for discordance for each participant case are detailed in Appendix Table A1.
Furthermore, there were six (32%) of 19 discordant cases that fell within the two categories of approved match despite concern for actionability of alteration and approved match despite concern for efficacy of drug. Within these categories, there was a level of permissiveness identified in the case summaries, where one of the MTBs approved a match despite concerns about the strength of the supporting data (ie, both groups agreeing that there was minimum evidence that the genomic alteration was actionable or both groups agreeing that there was minimum evidence that the drug was likely to be efficacious, however, one of the groups approved the match). Additionally, there was one case where the treatment proposed was no longer available for the patient (ie, cohort availability). Although the primary analysis included this discordant case, a sensitivity analysis was performed removing it from the analysis; inferences remained the same (P = .20, based on a null concordance rate of 75%).
DISCUSSION
When both MTB review groups were presented with the same participant case, there was full or partial concordance regarding the treating physician's treatment proposal for 79% of cases. This level of concordance demonstrates that treating physicians are consistently interpreting genomic reports and identifying acceptable treatment options most of the time. Although there were a number of discordant cases, the 21% discordance rate observed was lower than expected based on other published results.5,6
During review of the discordant cases, we looked to assess whether there were any patterns for differentiating treatment proposal recommendations or whether the time interval between MTB meetings had any impact. New information published between MTB reviews could be a reason for discordance in MTB recommendations. However, if this were the case, it would be expected that either most of the original meetings approved the treatment proposal and the secondary meeting denied it, or vice versa. This pattern was not seen with the discordant cases. There was an even mix for directionality with 42% of the treatment proposals for discordant cases being denied then approved during secondary review, and 58% of treatment proposals for discordant cases being approved and then denied during secondary review. In terms of time elapsed between MTB meetings, if newly published information or new FDA approvals had an impact on MTB recommendations, it would be expected that, on average, more time would have elapsed between meetings for discordant cases than for concordant cases. However, on average, more time had elapsed between concordant cases (500 days) than between discordant cases (458 days).
A key theme identified in the discordant cases was the amount of data or consistency of the published data available related to the treatment proposal. Discordance was most often due to the uncertainty about efficacy of drug and actionability of alteration. Previous research evaluating the level of concordance between MTB reviews has shown that discordance was higher when there was a lower level of evidence.5 A large proportion of the discordant cases (42%) in this study had summaries that cited weak evidence, or lack of evidence regarding the actionability of the proposed alteration or efficacy of the proposed study drug. For example, 8 (43%) of the treatment proposals among discordant cases were for temsirolimus based on a PIK3CA mutation. Mutations in PIK3CA have been reported in 13%-15% of solid tumors and have most commonly been identified in breast cancer, bladder cancer, colorectal cancer, and squamous cell carcinoma of the head and neck.15,16 Published research discussing the conflicting impact that PIK3CA has on prognosis and response to PI3K-directed therapy has concluded that further research is needed to determine the role that PIK3CA mutations play in solid tumors.15 The conflicting findings published on PIK3CA and PI3K inhibitors could account for why a large number of the discordant cases were for this treatment proposal.
In cases where there was limited supporting data, some MTBs decided against approving the proposed drug-alteration match, noting unlikely benefit to the patient. Other MTBs decided on providing the patient the opportunity to attempt the treatment, given that all other standard-of-care treatments had been exhausted. Of the discordant cases, 36% displayed a level of MTB permissiveness where one of the MTB reviews approved the treatment proposal, despite the lack of strong evidence. Since the goal of the TAPUR Study is to fill the knowledge gap when there is a lack of or inconclusive data, it is expected that there would be a level of permissiveness reflected in the MTB deliberations. From one perspective, by not allowing participants to enroll when there is a lack of data, we lose the opportunity to gather evidence and contribute information either supporting or refuting the actionability of an alteration.
Permissiveness may be influenced by including patient representatives in the TAPUR MTB meetings advocating for the patient. Patient advocates are known to provide a valuable and diverse perspective and add a human face to the participant's case being reviewed.8,9 There seems to be a gap in knowledge regarding the impact patient advocates have within MTBs. This is an area that would benefit from further research and could help contribute to establishing MTB membership guidelines.
In cases where there is limited evidence, it would be helpful to establish standard guidelines for MTBs on how to make recommendations in these situations. The TAPUR Study developed the TAPUR MTB Variant Determination Guide. However, there are many genomic databases that contain nonoverlapping information.17,18 The information available from various genomic databases as well as the challenges of interpreting genomic results has been shown to lead to differences of interpretation between reviewers.19 In the future, it could be helpful to develop a tool for MTBs that outlines the different resources that are available and the purpose of each resource. This could reduce reviewer discordance that results from different information available in different genomic databases.
A major limitation in applying the findings of this study to a general MTB is that MTB reviewers were asked only to assess whether the treatment proposal made by the site's treating physician was an acceptable match on the TAPUR Study, rather than to evaluate overall agreement for recommendations such as in other studies.14 This could account for the slightly higher rate of concordance compared with that observed in other MTB studies as potential treatment options on the TAPUR Study were constrained both by the availability of drugs in the study and the study matching rules. One would also expect a higher rate of concordance, given that both the clinical sites and the TAPUR MTB committee members were specifically selected to participate because of their experience with precision medicine.
In conclusion, this study may help contribute to benchmarking standards and guidelines for MTBs, including MTB member composition, and standardization of genomic reporting and database annotation that can improve concordance of MTB recommendations across different review groups. Doing so would help reduce the confusion/anxiety that patients could experience when receiving discordant MTB reviews from different groups of experts within an institution or when seeking second opinions.
ACKNOWLEDGMENT
The authors acknowledge all current members of the TAPUR Study team for their support and contributions to the TAPUR Study, and past members who contributed time and effort to the re-review study: Kaitlyn R. Antonelli, BA, Nicole L. Butler, MPH, Samiha Islam, MPH, Sasha Warren, Funmi Moore, MPH, and John Lybarger, MPH.
APPENDIX
TABLE A1.
Summaries of Reasons for Discordance and Outcome
| Case (No.) | Case Details | Reason for Discordance | Rationale |
|---|---|---|---|
| Case 1 | Treating physician proposed treatment | Pertuzumab + trastuzumab/ERBB3 mutation | Alteration disagreement (actionability of alteration) Rationale: The original meeting noted that the data on the ERBB3 T389K mutation being a driver mutation was weak and did not approve the treatment proposal for pertuzumab + trastuzumab. The re-review meeting indicated that although there is little clinical evidence to support using pertuzumab + trastuzumab without known HER2 amplification or mutation, there is some preclinical work suggesting the approach may be beneficial. Based on this, the re-review meeting approved the treatment proposal to pertuzumab + trastuzumab |
| Initial recommendation | Denied | ||
| Secondary recommendation | Pertuzumab + trastuzumab/ERBB3 mutation | ||
| Outcome | Not enrolled | ||
| Case 2 | Treating physician proposed treatment | Sunitinib/VHL mutation (VUS) | Approved match despite concern for actionability of alteration Rationale: Both the original meeting and re-review meeting agreed that the VHL E186K mutation (VUS) was unlikely to be functional, however the original meeting still approved the treatment proposal for sunitinib, the re-review meeting did not |
| Initial recommendation | Sunitinib/VHL mutation (VUS) | ||
| Secondary recommendation | Denied | ||
| Outcome | Enrolled on sunitinib. SD through week 28 and PD at week 40 | ||
| Case 3 | Treating physician proposed treatment | Temsirolimus/PIK3CA mutation | Treatment available on label Rationale: The original meeting approved the treatment proposal for temsirolimus based on the PIK3CA E545K mutation, stating that this mutation may predict responsiveness to an mTOR inhibitor. The re-review meeting noted that there was an mTOR inhibitor available to the patient as standard of care. They agreed that the patient should receive this if the treating physician wants to treat the patient with an mTOR inhibitor. The treatment proposal for temsirolimus was not approved |
| Initial recommendation | Temsirolimus/PIK3CA mutation | ||
| Secondary recommendation | Denied | ||
| Outcome | Enrolled on temsirolimus. PD before the first follow-up visit | ||
| Case 4 | Treating physician proposed treatment | Temsirolimus/PIK3CA mutation | Concern for efficacy of drug Rationale: The original meeting noted that mTOR inhibitors such as temsirolimus target PIK3CA only indirectly, and that a number of retrospective correlative analysis of clinical trials with mTOR inhibitors failed to show that patients with PIK3CA mutated tumors derive greater benefit than PIK3CA wild type tumors. The re-review meeting agreed that the clinical evidence for the predictive power of PIK3CA mutations to mTOR inhibition was weak, but they approved the match to temsirolimus |
| Initial recommendation | Denied | ||
| Secondary recommendation | Temsirolimus/PIK3CA mutation | ||
| Outcome | Not enrolled | ||
| Case 5 | Treating physician proposed treatment | Cetuximab/EGFR mutation | Approved match despite concern for efficacy of drug Rationale: Both meetings agreed that there was not sufficient evidence that single agent cetuximab has activity in EGFR mutated cancers. Despite their agreement on this, the original meeting approved the treatment proposal for cetuximab. The re-review meeting denied the request |
| Initial recommendation | Cetuximab/KRAS, NRAS and BRAF wildtype | ||
| Secondary recommendation | Denied | ||
| Outcome | Not enrolled | ||
| Case 6 | Treating physician proposed treatment | Palbociclib/CDK4 amplification (equivocal) | Approved match despite concern for actionability of alteration Rationale: The original meeting discussed the equivocal CDK4 amplification and concluded that there is not much data providing evidence that targeting these amplifications will result in meaningful clinical activity. Despite this, the original meeting approved the treatment proposal for palbociclib. The re-review meeting concluded that CDK4 amplification may not be a potent predictor of sensitivity to palbociclib. They did not approve the treatment proposal for palbociclib |
| Initial recommendation | Palbociclib/CDK4 amplification (equivocal) | ||
| Secondary recommendation | Denied | ||
| Outcome | Not enrolled | ||
| Case 7 | Treating physician proposed treatment | Temsirolimus/PIK3CA mutation | Concern for efficacy of drug Rationale: The original meeting noted that previous trials of mTOR inhibitors in colorectal cancer did not show a lot of benefit. They did not approve the treatment proposal for temsirolimus. The actionability of the PIK3CA G106V mutation was not discussed. The re-review meeting discussed that the PIK3CA mutation was actionable and approved the match to temsirolimus |
| Initial recommendation | Denied | ||
| Secondary recommendation | Temsirolimus/PIK3CA mutation | ||
| Outcome | Not enrolled | ||
| Case 8 | Treating physician proposed treatment | Temsirolimus/PTEN mutation | Concern for efficacy of drug Rationale: The original meeting reviewed the available data and noted that targeting PTEN variants (loss of PTEN) with mTOR inhibitors has not yielded sufficient evidence of clinical responsiveness. The treatment proposal for temsirolimus was not approved. The re-review meeting noted that the PTEN splice site 80-2_82delAGATA is activating of the PIK3/mTOR/AKT pathway and based on this approved the treatment proposal for temsirolimus |
| Initial recommendation | Denied | ||
| Secondary recommendation | Temsirolimus/PTEN mutation | ||
| Outcome | Not enrolled | ||
| Case 9 | Treating physician proposed treatment | Temsirolimus/PIK3CA mutation (subclonal) | Concern for efficacy of drug Rationale: The original meeting approved temsirolimus as an acceptable treatment option based on the subclonal PIK3CA E542K, E545K mutations. The re-review meeting did not approve the treatment proposal for temsirolimus citing negative data with mTOR inhibitors in colorectal cancer |
| Initial recommendation | Temsirolimus/PIK3CA mutation | ||
| Secondary recommendation | Denied | ||
| Outcome | Enrolled on temsirolimus. SD at 8 weeks and PD at 16 weeks | ||
| Case 10 | Treating physician proposed treatment | Regorafenib/RAF-1 rearrangement | Alteration disagreement (actionability of alteration) Rationale: The original meeting concluded that the RAF-1 rearrangement was likely to be activating and based on this approved the treatment proposal for regorafenib. The re-review meeting concluded that the RAF-1 rearrangement was not likely to be activating and therefore did not approve the treatment proposal for regorafenib |
| Initial recommendation | Regorafenib/RAF-1 rearrangement | ||
| Secondary recommendation | Denied | ||
| Outcome | Not enrolled | ||
| Case 11 | Treating physician proposed treatment | Pertuzumab + trastuzumab/ERBB3 mutation | Approved match despite concern for efficacy of drug Rationale: The original meeting noted that there was insufficient evidence for HER2 targeted therapies in patients whose tumor harbors an ERBB3 mutation. The treatment proposal for pertuzumab + trastuzumab was not approved. The re-review meeting also agreed that there was little clinical data to support the use of this combination in activating ERBB3 mutations, however they approved the match anyways |
| Initial recommendation | Denied | ||
| Secondary recommendation | Pertuzumab + trastuzumab/ERBB3 mutation | ||
| Outcome | Not enrolled | ||
| Case 12 | Treating physician proposed treatment | Temsirolimus/PIK3CA mutation | Concern for efficacy of drug Rationale: The original meeting concluded that the PIK3CA H1047R mutation was actionable and that there was evidence supporting the relationship between PIK3CA mutations and sensitivity to inhibitors of the PIK3 pathway. They approved the treatment proposal for temsirolimus. The re-review meeting noted that the patient has already received the mTOR inhibitor everolimus and did not expect a great response to temsirolimus since it has a similar mechanism of action. They did not approve the treatment proposal for temsirolimus |
| Initial recommendation | Temsirolimus/PIK3CA mutation | ||
| Secondary recommendation | Denied | ||
| Outcome | Not enrolled | ||
| Case 13 | Treating physician proposed treatment | Temsirolimus/PIK3CA mutation | Approved match despite concern for efficacy of drug Rationale: Both the original meeting and re-review meeting agreed that the PIK3CA H1047R mutation was activating. Both meetings also cited negative data in studies with temsirolimus and colorectal cancer. The original meeting approved the treatment proposal for temsirolimus despite this information. The re-review meeting did not approve the treatment proposal for temsirolimus |
| Initial recommendation | Temsirolimus/PIK3CA mutation | ||
| Secondary recommendation | Denied | ||
| Outcome | Not enrolled | ||
| Case 14 | Treating physician proposed treatment | Temsirolimus/PIK3CA mutation | Concern for efficacy of drug Rationale: Both the original meeting and the re-review meeting agreed that the PIK3CA K111del was pathogenic. The original meeting approved the treatment proposal for temsirolimus based on the PIK3CA mutation. The re-review meeting did not approve the treatment proposal for temsirolimus citing a concern for lack of efficacy |
| Initial recommendation | Temsirolimus/PIK3CA mutation | ||
| Secondary recommendation | Denied | ||
| Outcome | Enrolled on temsirolimus. SD at 8 weeks and PD at 16 weeks | ||
| Case 15 | Treating physician proposed treatment | Palbociclib/CDK4 amplification | Cohort availability Rationale: The original meeting did not approve the treatment proposal for palbociclib, citing that the cohort was suspended. The re-review meeting approved the treatment proposal for palbociclib based on the CDK4 amplification, citing a direct match per the TAPUR study protocol. The status of cohort availability was not discussed |
| Initial recommendation | Denied | ||
| Secondary recommendation | Palbociclib/CDK4 amplification | ||
| Outcome | Not enrolled | ||
| Case 16 | Treating physician proposed treatment | Pembrolizumab/Intermediate tumor mutational burden (7 muts/mb) | Alteration disagreement (actionability of alteration) Rationale: The original meeting concluded that a tumor mutational burden of 7 muts/mb did not meet the necessary threshold for a match to pembrolizumab. The treatment proposal for pembrolizumab was not approved. The re-review meeting concluded that an intermediate tumor mutational burden of 7 muts/mb was an appropriate match for pembrolizumab and approved the clinical sites treatment proposal |
| Initial recommendation | Denied | ||
| Secondary recommendation | Pembrolizumab/Intermediate tumor mutational burden (7 muts/mb) | ||
| Outcome | Not enrolled | ||
| Case 17 | Treating physician proposed treatment | Nivolumab + ipilimumab/BRCA1 mutation (VUS) | Alteration disagreement (actionability of alteration) Rationale: The original meeting did not approve the treatment proposal for nivolumab + ipilimumab. They concluded that the BRCA1 C9445 VUS mutation is not likely to be deleterious. The re-review meeting approved nivolumab + ipilimumab, but for high tumor mutational burden and not the BRCA1 VUS mutation. The focus of discussion was on the high tumor mutational burden and the actionability of the BRCA1 mutation was not discussed |
| Initial recommendation | Denied | ||
| Secondary recommendation | Nivolumab + ipilimumab/High tumor mutational burden | ||
| Outcome | Enrolled on pembrolizumab. Expired before the first follow-up visit | ||
| Case 18 | Treating physician proposed treatment | Temsirolimus/PIK3CA mutation | Alteration disagreement (actionability of alteration) Rationale: The original meeting approved temsirolimus based on the PIK3CA G106_R108 deletion. They noted that for this patient's disease type, mTOR inhibitors are FDA approved in combination with endocrine therapy. The actionability of the mutation was not discussed. The re-review meeting did not approve the treatment proposal for temsirolimus. They noted that the PIK3CA mutation was not well characterized in terms of functionality and that the mTOR inhibitor everolimus was already approved for this patient's indication |
| Initial recommendation | Temsirolimus/PIK3CA mutation | ||
| Secondary recommendation | Denied | ||
| Outcome | Enrolled on temsirolimus. PD at 8 weeks | ||
| Case 19 | Treating physician proposed treatment | Dasatinib/KIT amplification | Approved match despite concern for actionability of alteration Rationale: The original meeting approved the treatment proposal for dasatinib based on the KIT amplification, despite noting that there was a low-level amplification of a wide number of genes which the MTB concluded, likely meant that none of them are oncogenic drivers. The re-review meeting did not approve the treatment proposal for dasatinib, noting that the role of KIT amplification in prostate adenocarcinoma is unclear |
| Initial recommendation | dasatinib/KIT amplification | ||
| Secondary recommendation | Denied | ||
| Outcome | Enrolled on dasatinib. SD at 8 weeks then ended study with tumor progression before the 16-week visit |
Abbreviations: EGFR, estimated glomerular filtration rate; MTB, molecular tumor board; mTOR, mammalian target of rapamycin; PD, progressive disease; SD, stable disease; VUS, variant of unknown significance.
SUPPORT
Funding for the TAPUR Study is provided by AstraZeneca, Bayer, Boehringer Ingelheim, Bristol Myers Squibb, Eli Lilly and Company, Genentech, Merck & Co. Inc, Pfizer, Seagen and Taiho Oncology, Inc.
CLINICAL TRIAL INFORMATION
AUTHOR CONTRIBUTIONS
Conception and design: Jacqueline K. Perez, Jeanene Kleber, Pam Mangat, Richard L. Schilsky
Collection and assembly of data: Jacqueline K. Perez, Jeanene Kleber
Data analysis and interpretation: Jacqueline K. Perez, Jeanene Kleber, Michael Rothe, Elizabeth Garrett-Mayer
Manuscript writing: All authors
Final approval of manuscript: All authors
Accountable for all aspects of the work: All authors
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/po/author-center.
Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).
Richard L. Schilsky
Leadership: Clarified Precision Medicine, Leap Therapeutics
Stock and Other Ownership Interests: EQRx, Leap Therapeutics
Honoraria: Toray Industries
Consulting or Advisory Role: Cellworks, Scandion Oncology, Bryologyx, Illumina, EQRx, Syapse, Zephyr AI, AADi, Flatiron Health
Research Funding: AstraZeneca (Inst), Bayer (Inst), Bristol Myers Squibb (Inst), Genentech/Roche (Inst), Lilly (Inst), Merck (Inst), Pfizer (Inst), Boehringer Ingelheim (Inst), Seagen (Inst), Taiho Oncology (Inst)
Open Payments Link: https://openpaymentsdata.cms.gov/physician/1138818
No other potential conflicts of interest were reported.
REFERENCES
- 1. Mangat PK, Halabi S, Bruinooge SS, et al. Rationale and design of the targeted agent and profiling utilization registry study. JCO Precis Oncol. doi: 10.1200/PO.18.00122. 10.1200/PO.18.00122 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Reference deleted. [Google Scholar]
- 3. Miller RW, Hutchcraft ML, Weiss HL, et al. Molecular tumor board–assisted care in an advanced cancer population: Results of a phase II clinical trial. JCO Precis Oncol. 2022 doi: 10.1200/PO.21.00524. 10.1200/PO.21.00524 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Li MM, Datto M, Duncavage EJ, et al. Standards and guidelines for the interpretation and reporting of sequence variants in cancer: A joint consensus recommendation of the Association for Molecular Pathology, American Society of Clinical Oncology, and College of American Pathologists. J Mol Diagn. 2017;19:4–23. doi: 10.1016/j.jmoldx.2016.10.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Naito Y, Sunami K, Kage H, et al. Concordance between recommendations from multidisciplinary molecular tumor boards and central consensus for cancer treatment in Japan. JAMA Netw Open. 2022;5:e2245081. doi: 10.1001/jamanetworkopen.2022.45081. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Walters MK, Ackerman AT, Weese JL, et al. Quantifying the value of the molecular tumor board: Discordance recommendation rate and drug cost avoidance. JCO Precis Oncol. doi: 10.1200/PO.22.00132. 10.1200/PO.22.00132 [DOI] [PubMed] [Google Scholar]
- 7. Reference deleted. [Google Scholar]
- 8. Perlmutter J, Bell SK, Darien G. Cancer research advocacy: Past, present, and future. Cancer Res. 2013;73:4611–4615. doi: 10.1158/0008-5472.CAN-12-4553-T. [DOI] [PubMed] [Google Scholar]
- 9. A Spears P. Patient engagement in cancer research from the patient's perspective. Future Oncol. 2021;17:3717–3728. doi: 10.2217/fon-2020-1198. [DOI] [PubMed] [Google Scholar]
- 10. Reference deleted. [Google Scholar]
- 11. Reference deleted. [Google Scholar]
- 12. Reference deleted. [Google Scholar]
- 13. Koopman B, Groen HJM, Ligtenberg MJL, et al. Multicenter comparison of molecular tumor boards in The Netherlands: Definition, composition, methods, and targeted therapy recommendations. Oncologist. 2021;26:e1347–e1358. doi: 10.1002/onco.13580. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Rieke DT, de Bortoli T, Horak P, et al. Feasibility and outcome of reproducible clinical interpretation of high-dimensional molecular data: A comparison of two molecular tumor boards. BMC Med. 2022;20:367. doi: 10.1186/s12916-022-02560-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Alqahtani A, Ayesh HSK, Halawani H. PIK3CA Gene mutations in solid malignancies: Association with clinicopathological parameters and prognosis. Cancers. 2019;12:93. doi: 10.3390/cancers12010093. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Karakas B, Bachman KE, Park BH. Mutation of the PIK3CA oncogene in human cancers. Br J Cancer. 2006;94:455–459. doi: 10.1038/sj.bjc.6602970. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Pallarz S, Benary M, Lamping M, et al. Comparative analysis of public knowledge bases for precision oncology. JCO Precis Oncol. doi: 10.1200/PO.18.00371. 10.1200/PO.18.00371 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Wagner AH, Walsh B, Mayfield G, et al. A harmonized meta-knowledgebase of clinical interpretations of somatic genomic variants in cancer. Nat Genet. 2020;52:448–457. doi: 10.1038/s41588-020-0603-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Rieke D, Lamping M, Schuh M, et al. Comparison of treatment recommendations by molecular tumor boards worldwide. JCO Precis Oncol. doi: 10.1200/PO.18.00098. 10.1200/PO.18.00098 [DOI] [PubMed] [Google Scholar]