Patient selection for phase I trials in oncology is challenging. A typical inclusion criterion for phase I studies is 'life expectancy >3 months', however overall survival (OS) of patients with advanced solid malignancies are difficult to predict. Other studies have demonstrated that despite the use of traditional eligibility criteria for enrollment into a Phase I (P1) study, about 1/3 of patients fail to meet all the necessary eligibility criteria at screening and ∼15-20% die within the first 90 days of P1 trial entry. The P1 Unit at the Royal Marsden Hospital (RMH), UK has found that a prognostic score is a helpful tool in the process of patient selection for P1 trial entry. We aim to develop a similar prognostic score predicting OS in our P1 patient population here at the NCIS.
Patients in P1 studies completed between Oct. 2013 – Jan. 2016 at NCIS and had EMR available were reviewed. Patient characteristics and survival data were analysed.
120 pts from 10 completed P1 studies were included. Median age was 44.5 years (range: 23-79). Baseline variables were: age >60 (56%), male (53%), ECOG 0-1 (98%), comorbidities ≥2 (26%), no. of prior therapies ≥3 (65%), no. of metastatic sites ≥3 (40%). Univariate analysis revealed that albumin <35 g/L (HR: 3.23, 95% CI: 1.93 – 5.39, p < 0.001) and LDH >580 IU/L (HR: 1.95, 95% CI: 1.18 – 3.24, p = 0.009) were highly significant negative prognosticators for OS. The number of metastatic sites (>2 sites) was also at the threshold of significance (HR: 1.62, 95% CI: 0.99 – 2.64, p < 0.055). In the multivariable analysis of OS adjusting for tumour type, a high RMH prognostic score (2-3) vs. low score (0-1) was highly significantly for OS (HR: 3.67, 95% CI: 2.15 - 6.28), p < 0.001). OS according to the RMH prognostic risk score (high vs. low) for patients enrolled in P1 trials at the NCIS was 6 months and 16 months, respectively.
Risk scoring based on objective clinical parameters indicated that patients with a high score had a significantly shorter OS and this may assist in the selection of patient entry into P1 studies. A prospective study will be needed to test the validity of the prognostic score.
National University Cancer Institute, Singapore
National University Cancer Institute, Singapore
National University Cancer Institute, Singapore
All authors have declared no conflicts of interest.