1704P - Clinical benefit of cancer drugs approved by the US food and drug administration based on appropriateness of control arm and its change over time

Background

Value frameworks such as American Society of Clinical Oncology Value Framework (ASCO-VF) and European Society for Medical Oncology Magnitude of Clinical Benefit Scale (ESMO-MCBS) can be sensitive to control arms of randomized trials (RCTs). Standard of care changes over time. We explored clinical value based on appropriateness of control arm and its change over time.

Methods

We searched Drugs@FDA to identify new cancer drugs approved based on RCTs between January 2012 and December 2021, excluding trials with overlap between experimental and control arm. Appropriateness of control arm was based on published ESMO and NCCN guidelines both 1 year prior to start of accrual (as defined by ClinicalTrials.gov) and at time of drug approval. Control arm was defined as suboptimal if the strength of recommendation was 2B-3 for NCCN, III-V / C-E for ESMO guidelines and/or if prior RCT data showed that the control arm was inferior to an available alternative. Substantial clinical benefit was defined as ASCO-VF threshold score ≥45 and ESMO-MCBS grade A or B (curative intent) and 4 or 5 (palliative intent).

Results

We identified 55 RCTs supporting the approval of 32 drugs. Substantial benefit was observed in 65% and 55%, using ESMO-MCBS and ASCO-VF. Prior to accrual, appropriate control arm therapy was associated with non-significantly higher odds of clinical benefit using ESMO-MCBS with NCCN (OR 3.19, P=.23), but not ESMO guidelines (OR 1.29, P=.79). At the time of drug approvals associations were similar (OR 3.03, P=.06 and 1.17, P=.80). For ASCO-VF, appropriate control arm therapy was associated with non-significantly increased odds of clinical benefit (OR 5.52, P=.14) for both guidelines. These associations were not observed at the time of drug approval (OR 0.81, P=.69 for NCCN and 0.81, P=.72 for ESMO). Results were similar for RCTs performed exclusively in palliative intent.

Conclusions

ASCO-VF appears sensitive to change in appropriateness of control group therapy over time. ESMO-MCBS appears more consistent. Heterogeneity was observed based on scale, guideline and timepoint. Further exploration of impact of control group therapy on value frameworks is warranted.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.