Screening of Adolescents for Testicular Cancer — A Nationwide Retrospective Cohort Study

Screening 300,793 adolescents caught 4 testicular cancers in 6 months — and 86% of all cases were stage I anyway

Journal: American Journal of Epidemiology | Published: 2026-03-17 | Type: Retrospective cohort study | PMID: 41181889 Authors: Zloof Y, Erlich T, Braun M, et al. (Medical Corps, Israel Defense Forces; Hebrew University of Jerusalem; Sheba Medical Center) Funding/COI: Funding not listed; authors declare no conflicts of interest

Summary

Israeli military pre-service screening gave researchers an unusually clean natural experiment: over 300,000 adolescent males, all examined by physicians before mandatory conscription, followed for nearly a decade. The conclusion is unflattering to routine screening. Sensitivity dropped from 67% at six months to 40% at twelve, meaning most cancers evaded detection despite the exam. And since the vast majority of testicular cancers present at stage I with or without screening, catching them early via physical exam didn't demonstrably change outcomes.

Claims

• 43 testicular cancer cases identified across 1,172,603 person-years of follow-up in 300,793 adolescents; incidence rate 3.67 per 100,000 person-years
• Sensitivity: 66.7% at 6 months, dropping to 40% at 12 months; specificity held at 99.7%
• Positive predictive value: 0.5% — of 808 adolescents flagged as suspicious, only 4 had cancer within 6 months
• Number needed to screen: 75,198 adolescents per cancer detected; each detection came with 176 ultrasound tests and 112 specialist consultations
• 86% of cases were stage I at diagnosis (data available for 31 of 43 cases), consistent with the 75–80% stage I rate reported in unscreened populations
• No testicular cancer mortality observed in the cohort
• Tumor breakdown: 10 seminomas, 32 non-seminomatous germ cell tumors, 1 rhabdomyosarcoma; none preceded by cryptorchidism

Study Quality

This is one of the largest testicular cancer screening studies ever conducted — previous efforts (1,600 men in Germany, 1,504 US Army personnel) were too small to yield meaningful sensitivity estimates for a rare disease. The nationwide, population-based design with near-complete cohort capture is the study's primary strength. The retrospective structure is appropriate given the pre-existing military screening program, which provided a consistent, standardized examination protocol across five medical centers over ten years.

The authors correctly restricted sensitivity analysis to the first six months post-exam, acknowledging the rapid doubling time of testicular germ cell tumors — cancers appearing after twelve months likely didn't exist at the time of screening, making a longer window methodologically unsound. This is a thoughtful design choice that many screening studies get wrong.

Red Flags

• Funding source is not disclosed — for a study drawing data from military medical records, knowing whether this was government-funded or industry-adjacent matters
• Population generalizability is limited: Jewish ultra-orthodox males and Arab citizens are exempt from Israeli military service and explicitly underrepresented; this excludes potentially distinct risk subgroups
• Stage data missing for 12 of 43 cases (28%) — the 86% stage I figure is based on 31 cases, not the full cohort
• Follow-up capped at end of mandatory service — cancers arising after discharge are not captured, potentially underestimating incidence
• No comparison arm: this is not a randomized trial. The study can describe screening performance but cannot directly prove that screening changes mortality or stage distribution

Strengths

• Largest testicular cancer screening cohort in the literature by a wide margin (300,793 participants)
• Standardized, consistent examination protocol across a defined screening interval
• 10-year observation window with near-complete national coverage of the eligible male adolescent population
• Incidence rate (3.67 per 100,000 person-years) aligns with European and North American benchmarks, suggesting the cohort is not anomalous
• Quantifies downstream burden explicitly: ultrasounds, specialist consultations, and false-positive workup per cancer found

Verdict

This is a well-executed retrospective cohort study that does exactly what previous underpowered efforts could not: produce credible sensitivity estimates for testicular cancer screening. The numbers are not in screening's favor. A 40% sensitivity at twelve months, a positive predictive value under 1%, and no detectable stage-shifting benefit add up to a strong empirical argument against routine physical exam screening in asymptomatic adolescents — which aligns with current guidance from the US Preventive Services Task Force. The missing funding disclosure is an irritant, the population exclusions are worth noting, and randomized evidence for mortality impact remains absent. But as natural experiments go, this one is about as clean as the literature is likely to produce for a rare malignancy in a young, healthy cohort.