1543MO - Analysis of concordance between trial and target populations for cancer drugs in the US, EU, and Switzerland (2014-2023)

Background

Clinical trial results are crucial to understand the benefit-harm profile of new drugs. Extrapolating these findings to patients beyond the trial population lacks certainty, and patients could experience more intense side effects. The objective was to analyze the accordance between trial and target populations for cancer drugs approved in the US, EU, and Switzerland.

Methods

We used public databases (FDA, EMA, Swissmedic) to identify all new cancer drugs approved between 2014 and 2023 in the US, EU, and Switzerland. The following patient characteristics were extracted from the pivotal trial population and target population (clinicaltrials.gov, approval labels): Age, disease severity (e.g., locally advanced, metastatic), disease subtype (e.g., mutations), prior therapy (e.g., prior treatment lines). A drug can have multiple pivotal trials; each trial was assessed separately. Using descriptive statistics, we assessed the concordance between trial and target populations for the outlined patient characteristics.

Results

79 cancer drugs were approved in the US, EU, and Switzerland 2014-2023. When comparing patient population in pivotal trials and approval label, the FDA approved for 56/79 (71%) drugs a broader target population compared to the trial population, mostly in terms of age (38%), followed by prior therapy (35%), severity (23%), disease subtype (15%). The EMA approved for 44/79 (56%) drugs a broader target population, mostly in terms of prior therapy (30%), followed by severity (19%), disease subtype (16%), age (4%). Swissmedic approved for 39/79 (49%) drugs a broader target population, mostly in terms of prior therapy (24%), followed by age (19%), disease subtype (16%), severity (9%). For 41/79 (52%), 36/79 (46%), and 48/79 (61%) cancer drugs, the FDA, EMA, and Swissmedic, respectively, were more restrictive in the approved target population compared to the trial population, in approx 1/3 of the cases due to prior therapy and 1/4 due to severity.

Conclusions

Many cancer drugs - especially in the US but also in the EU and Switzerland - were approved for patients outside the trial population. Approval bodies should devote particular scrutiny to close the gap between trial and target populations for cancer drugs.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Swiss National Science Foundation (SNSF).

Disclosure

K.N. Vokinger: Financial Interests, Funding: Swiss National Science Foundation (SNSF). M. Serra-Burriel: Financial Interests, Funding: EU Horizon Grants. A. Kesselheim: Financial Interests, Funding: Arnold Ventures. All other authors have declared no conflicts of interest.