Customize your JAMA Network experience by selecting one or more topics from the list below.
Pinsky PF, Schoen RE. Colorectal Cancer Incidence by Age Among Patients Undergoing Surveillance Colonoscopy. JAMA Intern Med. 2015;175(5):858–860. doi:10.1001/jamainternmed.2015.0344
A recent study from the Kaiser Permanente–Southern California region reported that, among patients undergoing surveillance colonoscopy, the rate of subsequent colorectal cancer (CRC) diagnosis was significantly lower for those aged 75 years or older than for those aged 50 to 74 years.1 The CRC incidence was 0.24 per 1000 person-years in older patients vs 3.64 per 1000 person-years in younger patients; in a multivariable proportional hazards model, the hazard ratio for CRC in elderly patients was 0.06 (95% CI, 0.02-0.13). The finding of a 15-fold lower CRC rate among older patients is striking considering that the CRC incidence is higher among those 75 years or older than those 50 to 74 years, according to the Surveillance, Epidemiology, and End Results Program database.2 We attempted to replicate this finding in the ancillary Study of Colonoscopy Utilization (SCU), a study examining the use and yield of surveillance colonoscopy, from the Prostate, Lung, Colorectal, and Ovarian (PLCO) cancer screening trial.3
The PLCO was a randomized trial of prostate, lung, colorectal, and ovarian cancer screening. Approximately 155 000 patients aged 55 to 74 years were randomized at 10 screening centers between 1993 and 2001 to an intervention or control arm.4 For colorectal cancer, those in the intervention arm received flexible sigmoidoscopy (FSG) at baseline and year 3 or 5. In 2006, the SCU was conducted in patients who had abnormal results on a baseline FSG followed by diagnostic colonoscopy, with eligible patients randomly sampled at different frequencies based on baseline adenoma status (patients with advanced adenoma were oversampled; those with CRC were excluded).3 Patients participating in the SCU were queried by telephone about surveillance colonoscopies. Colonoscopy reports and pathologic findings were verified with medical records. Cancer incidence was ascertained in all PLCO participants, primarily through annual study updates. Study participants were observed for up to 13 years from randomization.
We defined surveillance colonoscopy as any colonoscopy occurring within 10 years of the diagnostic colonoscopy that followed the abnormal results from the baseline FSG; we excluded colonoscopies performed for symptoms or as follow-up to abnormal results on the FSG screening at year 3 or 5. The rates of incident CRC diagnosed at or after the first surveillance colonoscopy stratified by age were computed incorporating sampling weights. Colorectal cancer incidence by age was also analyzed using a multivariable proportional hazards model controlling for sex. All PLCO cancer centers provided institutional review board approval for the study. All participants provided written informed consent.
Of 10 876 PLCO participants with abnormal results on the baseline FSG and follow-up colonoscopy, 3876 were selected for the SCU; 3627 (93.6%) of those selected completed the SCU interview and 2398 underwent a surveillance colonoscopy. Table 1 displays the characteristics of the surveillance colonoscopy cohort. During follow-up (mean length, 8.2 years), 21 cases of incident CRC were observed.
Colorectal cancer incidence rates (per 10 000 person-years) were 6.6, 9.5, and 11.4 for those aged 55 to 69 years, 70 to 74 years, and 75 to 80 years, respectively (Table 2). The proportional hazards model for CRC incidence showed an adjusted hazard ratio of 1.5 (95% CI, 0.7-3.5) for those aged 70 years or older compared with those younger than 70 years.
Owing to the small numbers, we cannot estimate the hazard ratio with precision for CRC in those 70 years or older undergoing surveillance colonoscopy. However, the 95% CI of our hazard ratio (0.7-3.5) includes the approximate 2.5-fold increased incidence in those 70 years or older vs those aged 55 to 69 years estimated from the Surveillance, Epidemiology, and End Results database.
Although our study involved participants enrolled in the PLCO trial, diagnostic follow-up of positive screening results, including the initial diagnostic colonoscopy and any surveillance colonoscopy, was performed outside the trial auspices in community settings in 9 metropolitan areas and 1 rural area. As such, these findings are representative of clinical practice. Limitations of our study include the fact that PLCO participants were volunteers in a screening trial, that the initial screening test was FSG and not colonoscopy, and that some patients (6.4%) were nonresponders to the SCU. In contrast, within a single health maintenance organization, institutional factors, including the age and comorbidity of patients selected for surveillance, or the quality of colonoscopy (approximately 40% of malignant neoplasms in the younger age group were found at the initial surveillance examination), could have contributed to the unusual observed outcome. More research is needed to estimate the risks and benefits of surveillance colonoscopy by age.
Corresponding Author: Paul F. Pinsky, PhD, Division of Cancer Prevention, National Cancer Institute, 9609 Medical Center Dr, Bethesda, MD 20892 (email@example.com).
Published Online: March 30, 2015. doi:10.1001/jamainternmed.2015.0344.
Author Contributions: Dr Pinsky had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Pinsky.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: All authors.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Pinsky.
Obtained funding: Schoen.
Conflict of Interest Disclosures: None reported.
Funding/Support: The Prostate, Lung, Colorectal and Ovarian trial and Study of Colonoscopy Utilization were funded through contracts from the National Cancer Institute.
Role of the Funder/Sponsor: The National Cancer Institute had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.