28O - Comparison of prognostic scores in oncology patients participating in phase I clinical trials with immunotherapy

Background

Phase I studies are designed to evaluate the toxicity and maximum tolerated dose of a new drug but, for the last decade, antitumoral activity is also measured in this type of trials. Usually, an expected overall survival of at least three months is a mandatory requirement for accessing these trials. However, up to 15% of these patients may die before this mark. Different prognostic scales have been developed to identify those patients with a worse prognosis and thus less chances to benefit from phase 1 trials. RMH score, the GRIm score, the PIPO score, the ROPRO score, and the LIPI score are among the most widely used tools to predict survival. The aim of this study is to compare these prognostic scales in terms of best discrimination capacity.

Methods

A retrospective study was performed in a cohort of 844 patients included in phase 1 clinical trials with immunotherapy in Clinica Universidad de Navarra from January 2016 until March 2023. Relevant data needed for calculating the scores were obtained retrospectively from medical OS was calculated from the first dose of the drug to death or last documented clinic contact. Each prognostic score punctuation was calculated according to published data. C-index and dynamic AUC were calculated for each of the scores. The statistical analysis was conducted using R-4.3.1®and Python software. Specific code is available upon request.

Results

Among the 844 patients, 53.08% were males with a mean age of 59 years old. The most frequent tumor was lung cancer followed by colorectal cancer. Most phase 1 clinical trials were with checkpoint inhibitors. Median OS was 9.08 months. AUC at 3 months was 0.56 for GRIm and RMH scores; 0.63 for PIPO score; and 0.67 and 0.76 for LIPI and ROPRO, respectively.

Conclusions

In our cohort, the RMH, PIPO and GRIm scales are very similar to prognosticate survival in patients in phase 1 trials. LIPI and ROPRO scores show a better prognostic capacity. ROPRO score maintained the highest AUC value through all the observation periods. However, further prospective studies are needed.

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.