The relationship between progression-free and post-progression survival in treating four types of metastatic cancer
Introduction
Trying to predict overall survival on the basis of either disease-free or progression-free survival in cancer studies is important because, for a new type of cancer treatment, data for disease-free or progression-free survival will often become available far sooner that data for overall survival. For this reason, the relationship between progression-free or disease-free survival and overall survival in clinical trials of cancer treatments has been studied by a number of researchers. Sargent et al. [1] studied the relationship between disease-free and overall survival for the adjuvant treatment of colon cancer. For the general treatment of metastatic colon cancer, the relationship between progression-free and overall survival was studied by Louvet et al. [2], for both metastatic colon and non-small-cell lung cancer this relationship was studied by Johnson et al. [3] and for metastatic breast cancer this relationship was studied by Sherrill et al. [4]. All of these studies found a positive correlation between either disease-free or progression-free survival and overall survival.
This study will analyse the relationship between progression-free and overall survival for cancer treatments in a way that is not explored in the above studies by attempting to answer the following question in relation to four different types of metastatic cancer:
Does an increase in progression-free survival generally lead to an increase, a decrease or no change in the time between disease progression and death, i.e. the post-progression survival time?
By attempting to answer this question, it is hoped that a clearer understanding will be obtained about the extent to which increases in progression-free survival that appear to be caused by new types of cancer treatment will be generally reflected in increases in overall survival.
The four types of metastatic cancer that will be studied are metastatic breast cancer, colorectal cancer, hormone-refractory prostate cancer and non-small-cell lung cancer. The four general cancer types of breast, colorectal, prostate and lung cancer were selected because within the UK, apart from non-melanoma skin cancer, they are by far the most common types of cancer in terms of incidence and on the basis of the most recent data (Cancer Research UK 2005 Data) they are the four most common causes of death from cancer within the UK.
Similar to the study design used by Johnson et al. [3] but unlike the other studies mentioned above, the effect on overall survival of differences in progression-free survival between treatment and control groups will be examined rather than progression-free survival being simply compared with overall survival for each individual treatment arm. In particular, we will analyse results from randomized controlled trials (RCTs) where one type of chemotherapy has been compared to another type of chemotherapy. This restriction on the type of cancer treatment considered, although not imposed by Johnson et al. [3], was put in place as an attempt to reduce potential biases caused by heterogeneity between studies.
Note that Edwards et al. [5] studied the relationship between the time to relapse and the time from relapse to death for individual patients pooled from five trials of adjuvant therapy for early stage breast cancer. However, comparing progression-free survival with post-progression survival for individual trial subjects rather than comparing mean or median differences between treatment arms would not be very useful in attempting to answer the question of interest outlined above. Moreover, the existence of a positive relationship at the patient level between these two types of survival time could simply be due to differences in the resistance to the spread of the cancer for the patients concerned.
Section snippets
Materials and methods
The literature search was conducted for any systematic reviews of randomized controlled trials (RCTs) that study the effectiveness of chemotherapy in treating either metastatic breast cancer, hormone-refractory prostate cancer, colorectal cancer or non-small-cell lung cancer. The search was conducted over Medline and the Cochrane Database of Systematic Reviews using the date range of January 1990 to June 2007. The reviews had to be written in English. The search strings used can be found in
Results
For metastatic breast, colorectal, hormone-refractory prostate and non-small-cell lung cancer, the number of systematic reviews that were found by the literature search was 42, 30, 13 and 31, respectively, and from amongst these reviews, the number of treatment comparisons that were found that satisfied the given inclusion criteria was 34, 42, 25 and 17, respectively, which were contained within, respectively, 33, 38, 23 and 13 RCTs. For the analysis of these RCT results that will now be
Discussion
On the basis of Fig. 1, Fig. 3, Fig. 4, it is important to note that for breast, hormone-refractory prostate and non-small-cell lung cancer, most of the changes in post-progression survival are negative rather than positive. In particular, for hormone-refractory prostate and non-small-cell lung cancer, the outcome of a sign test for the direction of changes in post-progression survival is significant at the 5% and 1% levels of significance, respectively. This suggests that for
Acknowledgement
This work was supported by the Engineering and Physical Sciences Research Council of the UK through the MATCH Programme (Grant GR/S29874/01).
References (5)
- et al.
Response rate or time to progression as predictors of survival in trials of metastatic colorectal cancer or non-small-cell lung cancer: a meta-analysis
Lancet Oncology
(2006) - et al.
End points in the analysis of breast cancer survival: relapse versus death from tumor
Surgery
(1998)
Cited by (18)
A systematic review evaluating the relationship between progression free survival and post progression survival in advanced ovarian cancer
2012, Gynecologic OncologyCitation Excerpt :However there is very little literature on the effects of gains in PFS on OS in ovarian cancer and none in advanced primary ovarian cancer. In addition, only two previous papers [8,9] (both from our group), have attempted to calibrate the size of any relationship between PFS and OS. This is important because a six-month gain in PFS may translate into a 9, 12 or 15 month gain in OS and the magnitude of this effect is critical for patients and for health economic models.
Validation of surrogate endpoints in advanced solid tumors: Systematic review of statistical methods, results, and implications for policy makers
2014, International Journal of Technology Assessment in Health CareEstimating changes in overall survival using progression-free survival in metastatic breast and colorectal cancer
2011, International Journal of Technology Assessment in Health CareUtilizing restricted mean duration of response for efficacy evaluation of cancer treatments
2022, Pharmaceutical StatisticsSample size determination for the current strategy in oncology Phase 3 trials that tests progression-free survival and overall survival in a two-stage design framework
2018, Journal of Biopharmaceutical StatisticsRestricted mean survival time: An obligatory end point for time-to-event analysis in cancer trials?
2016, Journal of Clinical Oncology