Overview: progression-free survival as an endpoint in clinical trials with solid tumors

Abstract

Progression-free survival (PFS) is increasingly used as an important and even a primary endpoint in randomized cancer clinical trials in the evaluation of patients with solid tumors for both practical and clinical considerations. Although in its simplest form, PFS is the time from randomization to a predefined endpoint, there are many factors that can influence the exact moment of when disease progression is recorded. In this overview, we review the circumstances that can devalue the use of PFS as a primary endpoint and attempt to provide a pathway for a future desired state when PFS will become not just a secondary alternative to overall survival but rather an endpoint of choice.

Conflict of interest statement

Conflicts of Interest: None

©2013 AACR

Figures

Figure 1. Lesion Response: Change In Tumor Density Versus Size

The figure above shows the response behavior of a solitary hepatic metastatic lesion on a contrast-enhanced CT scan in a subject with pancreatic carcinoma during experimental therapy. Note the tumor nodularity within the lesion (*) at screening surrounded by fluid (dark appearance on CT). The tumor nodular has disappeared on 90-day scan while the lesion has become more fluid containing (average density within the lesion went from 59HU to 42HU or 29% decrease as indicated in the chart), suggesting intervening tumor necrosis. However, the lesion has not changed substantially in size. This lesion would be considered stable by RECIST criteria but would be a responding lesion by CHOI (21). By 180 days, the lesion has remained stable in size but the nodularity is beginning to reappear (double-head arrow) suggesting tumor recurrence. Thus, a PFS of 180 days would not have been captured by RECIST criteria, leading to a positive bias in favor of experimental therapy.

Figure 2. Non-Pulmonary Volumetric Analysis As A Measure of Treatment Responses

Volumetric analysis of tumor burden was performed (inset top left) at baseline and end of cycle 2 using contrast enhanced CT. Even though the bidimensional measurements of the tumor (green lines) did not change significantly during with therapy, the tumor volume decreased by 73%, suggesting a favorable response to therapy. Quantitative measurements of tumor volume change might be a more sensitive method of assessing tumor response than unidimensional or bidimensional measurements.