From the Journals

Studies eye adjuvant therapy in locally advanced rectal cancer

View on the News

Propensity scores can overestimate effects

These retrospective studies probably overestimate the benefits of adjuvant chemotherapy in ypT0N0 rectal cancer, argued George J. Chang, MD, MS.

“If there is a true benefit associated with adjuvant chemotherapy, it is likely to be small,” he wrote in an editorial accompanying the studies in JAMA Oncology.

The problem comes down to methodology – both studies used propensity-score matching, which can overestimate treatment effects, continued Dr. Chang of the University of Texas MD Anderson Cancer Center, Houston.

Investigators should beware of this method for large registry studies like these, he added. “There are many examples in the literature of propensity-score analysis leading to biased estimates, despite apparently rigorous methodology.”

Neoadjuvant chemoradiation treatment has a favorable prognosis in ypT0N0, so patients should carefully weigh the risk of side effects from adjuvant chemotherapy, such as chronic neuropathy, Dr. Chang wrote. He recommended pursuing alternatives to propensity-score matching that are less subject to confounding.

Dr. George J. Chang had no conflicts of interest. These comments paraphrase his editorial (JAMA Oncol. 2018 Apr 19. doi: 10.1001/jamaoncol.2018.0227 ).


 

FROM JAMA ONCOLOGY

For adults whose stage T3 or greater rectal cancer shows a pathological complete response to neoadjuvant chemoradiation and resection, perioperative adjuvant therapy was associated with significantly improved outcomes, according to the results of two observational cohort studies of the National Cancer Database.

In the first study, Fahima Dossa, MD, of the University of Toronto, and her associates compared 667 patients with T1 to 2N+ or T3 to 4N0/+ rectal adenocarcinoma who received adjuvant chemotherapy with 667 patients who did not. They used propensity-score matching to account for measurable differences between groups that might influence the receipt of adjuvant treatment. They also conducted a subgroup analysis to determine if treatment effects depended on whether patients initially had node-positive disease.

After a median follow-up of 3.1 years (interquartile range, 1.9-4.4 years), adjuvant therapy was associated with significantly improved overall survival (hazard ratio, 0.44; 95% confidence interval, 0.28-0.70), Dr. Dossa and her coinvestigators said. However, the effect was limited to patients with pretreatment, node-positive disease (HR, 0.24; 95% CI, 0.10-0.58).

“Although this study suggests a beneficial effect of adjuvant treatment on survival in patients with pathologic[al] complete response, these results are limited by the presence of potential unmeasured confounding in this nonrandomized study,” Dr. Dossa and her associates wrote in JAMA Oncology. A pathological complete response signifies a good prognosis in locally advanced rectal cancer, and any additional survival benefit from adjuvant chemotherapy has to be weighed against the risk of chemotoxicity, they stressed. (In their study, 5-year overall survival rates were 95% in the adjuvant group and 88% in the observation-only group).

In the second study, Patricio M. Polanco, MD, of the University of Texas, Dallas, and his associates also used propensity-score matching to compare 741 patients with clinical TNM stages T1 to 2N+ or T3 to 4N0/+ rectal adenocarcinoma who received adjuvant chemotherapy with 741 patients who underwent observation only. Once again, after a median follow-up of 3.2 years, adjuvant chemotherapy was associated with significantly improved overall survival (HR, 0.50; 95% CI, 0.32-0.79). Estimated 5-year overall survival rates were 95% and 88%, respectively (P = .05). Subgroup analyses showed that adjuvant chemotherapy had the strongest apparent effect in the setting of stage T3/T4 and node-positive disease.

Current U.S. guidelines recommend treating stage T3 or higher or node-positive rectal cancer with neoadjuvant chemoradiation therapy plus 6 months of perioperative chemotherapy. Each of these studies attempted to address the lack of evidence clearly supporting adjuvant chemotherapy for these patients, the researchers noted.

Dr. Fossa and her associates reported no specific funders or conflicts of interests. Dr. Polanco disclosed support from the VA North Texas New Investigator Program, Department of Veterans Affairs. He and his associates reported no conflicts of interest.

SOURCES: Dossa F et al. JAMA Oncol. 2018 Apr 19. doi: 10.1001/jamaoncol.2017.5597; Polanco PM et al. JAMA Oncol. 2018 Apr 19. doi: 10.1001/jamaoncol.2018.0231.

Recommended Reading

Gastrointestinal cancers: new standards of care from landmark trials
MDedge Hematology and Oncology
Accuracy of colon cancer lymph node sampling influenced by location
MDedge Hematology and Oncology
MDedge Daily News: Could gut bacteria trigger lupus?
MDedge Hematology and Oncology
MDedge Daily News: Antibiotic resistance leads to ‘nightmare’ bacteria
MDedge Hematology and Oncology
No increased complication risk with delaying resection for LARC
MDedge Hematology and Oncology
CtDNA agrees (mostly) with tissue analysis in mCRC
MDedge Hematology and Oncology
MDedge Daily News: Diabetes patients ignore a deadly risk
MDedge Hematology and Oncology
Enhanced recovery led to fewer complications for major oncologic procedures
MDedge Hematology and Oncology
Healthy lifestyle linked to better colon cancer survival
MDedge Hematology and Oncology
MDedge Daily News: Can a nasal spray reverse suicidality?
MDedge Hematology and Oncology