Incidence of Upper Tract Urothelial Carcinoma in Relation to Smoking, Analgesic Use, and Medical History in the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial

Principal Investigator

Name
Jonathan Hofmann

Degrees
Ph.D., M.P.H.

Institution
Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health

Position Title
Senior Investigator

Email
hofmannjn@mail.nih.gov

About this CDAS Project

Study
PLCO (Learn more about this study)

Project ID
PLCO-2085

Initial CDAS Request Approval
Sep 28, 2026

Title
Incidence of Upper Tract Urothelial Carcinoma in Relation to Smoking, Analgesic Use, and Medical History in the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial

Summary
Upper tract urothelial carcinoma (UTUC), or cancer developing in the renal pelvis or ureter, accounts for only 5-10% of urothelial carcinomas and occurs at an estimated annual incidence of 1-2 cases per 100,000 in the U.S. Incidence peaks in the early to mid-70s and is twice as common in the renal pelvis as in the ureter, and in men compared to women. Although UTUC is thought to behave similarly to bladder urothelial carcinomas (90-95% of cases) and share overlapping risk factors, its etiology remains comparatively understudied due to its rarity.

Tobacco smoking is the most well-established UTUC risk factor, with case-control studies reporting 3- to 6-fold increased risk among smokers relative to non-smokers, and even higher risk among those with long smoking histories or heavy cumulative exposure. Beyond incidence, smoking has also been linked to worse outcomes among UTUC patients, including younger age at diagnosis, more advanced stage, and higher recurrence and mortality risk. However, most prospective evidence on smoking and urothelial cancer mortality or incidence has either pooled UTUC with bladder cancer or examined all renal pelvis/ureter cancers rather than urothelial histology specifically, hindering UTUC-specific conclusions.

Investigation into other suspected risk factors is less conclusive. Phenacetin-containing analgesics (banned in 1983) have been linked to UTUC risk, but evidence for non-phenacetin analgesics such as acetaminophen, aspirin, and other NSAIDs is inconsistent, with prospective data being especially scarce. One prospective cohort reported no association between long-term NSAID use and bladder urothelial cancer but did not examine UTUC separately. Coffee and tea consumption have been studied more extensively in relation to bladder cancer, with some evidence of effect modification by smoking status, but few studies have examined these exposures specifically in relation to UTUC or renal pelvis/ureter cancer. Medical history factors including higher body mass index and prior or concomitant bladder cancer have also been implicated, but research on these exposures has yielded mixed results.

Most existing evidence on these risk factors derives from case-control studies, which may be susceptible to recall and selection bias, or from cohort studies of bladder cancer that do not report UTUC-specific estimates. Large prospective studies including the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial with standardized variables for suspected risk factors, and long follow-up for accrual of sufficient numbers of UTUC cases, are needed to clarify these associations. Incident cancers in the renal pelvis (site code: C659) and ureter (C669) have been identified in PLCO participants through registry linkage. To date, a total of 236 cases have been identified during follow-up, of which 155 (66%) were in the renal pelvis and 81 (34%) in the ureter. Analyses will focus on 222 (94%) cases characterized as urothelial carcinomas (histology codes: 8120, 8130, 8131, 8050). Participants with prevalent cancer at baseline (except the bladder cancer history analysis) and missing or inconsistent exposure or follow-up data will be excluded. Overall, the proposed study is expected to contribute to the limited body of evidence examining UTUC etiology and its associated risk factors.

Aims

To evaluate associations between UTUC risk in relation to tobacco smoking, analgesic use (acetaminophen, aspirin, ibuprofen, and other non-steroidal anti-inflammatory drugs), coffee and tea consumption, and medical history factors (body mass index, prior or concomitant bladder cancer) in the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial. Cox proportional hazard models will be used to estimate hazard ratios and 95% confidence intervals of UTUC in association with the risk factors of interest, using age as the time scale. As sample size permits, we will examine risk variation across histologic and molecular subtypes of UTUC in relation to the risk factors of interest.

Collaborators

Stella Koutros Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health
Jonathan Hofmann Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health
Grant Tore Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health