CHICAGO – Delaying adjuvant chemotherapy following curative resection of colorectal cancer worsens overall and disease-free survival, according to the findings of a meta-analysis and systematic literature review.
Using 4 weeks postoperatively as a reference, Dr. James J. Biagi and his colleagues found that each additional 4 weeks between surgery and initiation of adjuvant chemotherapy was tied to a 14% decline in overall survival. A patient medically eligible for adjuvant chemotherapy at 4 weeks but whose adjuvant chemotherapy was delayed for any reason had a 14% higher mortality risk if treatment started at 8 weeks, and an almost 30% higher risk if treatment started at 12 weeks.
A second major finding was that some benefit of adjuvant chemotherapy remained beyond 12 weeks, Dr. Biagi and his associates reported at the annual meeting of the American Society of Clinical Oncology (ASCO).
Although most protocols do not recommend adjuvant chemotherapy beyond 3 months, "it is possible that a reasonable limit may be more on the order of 4 to 5 months," Dr. Biagi and his colleagues wrote in a report (JAMA 2011;305:2335-42) released to coincide with the presentation of results at the meeting.
The findings were consistent for disease-free or cancer-specific survival as well, reported Dr. Biagi, acting head of oncology at Queen’s University Cancer Research Institute, Kingston, Ontario, and his colleagues.
Prior to the study, the optimal time from surgery to initiation of adjuvant chemotherapy had not been identified. Dr. Biagi and his coinvestigators searched MEDLINE, as well as abstracts from recent proceedings of the annual meeting of ASCO and the European Society for Medical Oncology. They found 10 relevant studies that assessed overall survival related to adjuvant chemotherapy wait times with a total 15,410 patients. Six of these studies also addressed disease-free, relapse-free, or cancer-specific survival among 12,584 patients.
The studies of overall survival yielded a combined hazard ratio of 1.14 for a relative increase in mortality for each 4 weeks of waiting time past 4 weeks. The combined hazard ratio was the same, 1.14, for the six studies with additional survival end points. In addition, three studies specifically addressed cancer-specific survival; these yielded a combined hazard ratio of 1.15.
Nine studies were population- or registry-based, and one was a secondary analysis of a randomized controlled trial. Because it is unlikely that a prospective study will be conducted to evaluate the relationship between adjuvant chemotherapy wait times and survival in this patient population, the authors wrote that "[they] believe the level of evidence from this study provides sufficient evidence of causality."
The effect of delays might be substantial, Dr. Biagi and his colleagues noted. An estimated 49,000 people are newly diagnosed with stage III colorectal cancer each year. This represents 35% of the approximately 140,000 or so new cases of colorectal cancer in the United States.
The influence of postoperative performance status on wait times was not evaluated, which is a potential limitation of the study, Dr. Biagi and his coauthors wrote. In addition, the researchers were unable to determine the effect of starting adjuvant chemotherapy in the initial weeks after resection (before 4 weeks) or what percentage of patients completed adjuvant chemotherapy. Moreover, the studies were largely from a period of fluoropyrimidine adjuvant chemotherapy prior to widespread use of oxaliplatin, so extrapolation to the current era in which oxaliplatin is often added to fluoropyrimidine is uncertain.
Dr. Biagi previously presented the results of this meta-analysis and systematic review at a meeting on gastrointestinal cancers sponsored by ASCO, and it was reported by this news organization.
Dr. Biagi had no relevant financial disclosures. A study coauthor, Dr. William J. Mackillop, disclosed that he provided expert testimony in a 2009 class action suit regarding delays in postlumpectomy radiotherapy for breast cancer and the probability of local disease control.