As clinicians, we all strive to improve outcomes for our patients and we are always disappointed when a large phase III trial evaluating a novel agent reports negative results. As in life, we all learn from mistakes and cancer research is clearly no exception to this. Negative trials are as important as positive ones. It is crucial that the results of these trials are reported in order to understand what went wrong and what could have been done differently, to avoid the same mistakes when designing the next trial. For this reason, I thought would be interesting to review the recently published data from the phase III trial ATTRACT-1.
Lara and colleagues randomized unselected patients with advanced stage IIIB/IV NSCLC to receive standard chemotherapy with carboplatin and paclitaxel with or without the vascular disrupting agent ASA404 (1800 mg/m2) given for six cycles followed by maintenance ASA404/placebo. This trial was conducted at more than 200 sites in 20 countries and 1299 patients were enrolled. Primary end point was overall survival (OS); secondary end points included overall response rate (ORR) and progression-free survival (PFS).
A preplanned interim analysis was conducted in March 2010 (occurrence of 25% of the total deaths) and the independent data safety monitoring committee recommended stopping the trial for futility. The two arms were well balanced with regard to the demographic characteristics. Ninety-one percent of patients had stage IV disease and the majority had non-squamous tumour histology (75%), with adenocarcinoma (approximately 67%) being the most common subtype. There was no statistically significant difference in OS between the two arms (13.4 vs. 12.7 months, HR 1.01; 95% CI 0.85 - 1.19; p=0.535). There were also no differences in OS according to histology (non-squamous: HR 0.98; 95% CI, 0.80 - 1.19 and squamous: HR 1.10; 95% CI 0.79 - 1.52) or sex (male: HR 1.02; 95% CI 0.83 - 1.25 and female: HR, 0.98; 95% CI 0.72 - 1.34). Overall response rate (24.7% vs. 24.6%, p=1.0) and PFS (5.5 vs. 5.5 months, p=0.727) were similar. The rate of adverse events (AEs) was not different between the ASA404 and placebo arm, with grade 4 neutropenia (27% v 19%) and infusion site pain (10% v 0.5%) reported more frequently in the ASA404 arm. Notably, patients enrolled in the ASA404 group did not have increased vascular toxicity, such as bleeding or thrombosis with, even in the squamous cell cancer subset.
Another targeted agent fails to improve survival when added to chemotherapy, another trial to be added to a disappointingly long list. What went wrong? Were the negative results pre-announced? The elegant accompanying editorial by Patricia LoRusso and colleagues explains that the results of the ATTRACT-1 trial were not a surprise. The basis of the decision to embark in a phase III study was influenced by the results from the small (n=70) randomized phase II study of ASA404 plus paclitaxel/carboplatin, which showed longer OS in the investigational arm (14 vs. 8.8 months, HR 0.73) but with a non-statistically significant p value of 0.33. Furthermore, the mechanism of action of ASA404 it is still not completely understood and, as a result, no biomarkers were used to select a specific subgroup of patients for accrual.
This trial is a clear example of how critical it is to have a good understanding of the mechanism of action and preclinical activity of the new drug and how important is to have solid and reliable phase II data before embarking in to phase III development. Only careful evaluation of early-phase data and possibly selecting subsets of patients most likely to benefit will decrease the chances of failure and most importantly prolong patients’ survival. We must not forget that every time we fail, a large amount of funding and clinical trial’s team effort is wasted. Most importantly, we make inappropriate use of the most valuable and irreplaceable resource we have: our patients.