Inappropriate Use of Tumor Markers in Asymptomatic Patients: A Systematic Review and Proposed Follow-Up Algorithm.
Serum tumor markers such as CA 125, CA 19-9, carcinoembryonic antigen (CEA), alpha-fetoprotein (AFP), and CA 15-3 are widely available, inexpensive, and easily measured, which has led to their inappropriate use for cancer screening in asymptomatic individuals. Despite guideline recommendations against this practice, these tests remain common, contributing to false positives, unnecessary anxiety, and avoidable invasive procedures.
We conducted a systematic review of PubMed, Cochrane Library, and LILACS from inception to May 10, 2025, identifying studies evaluating the use of the aforementioned tumor markers in asymptomatic populations. Eligible studies reported diagnostic accuracy or clinical outcomes. Two reviewers independently screened, extracted data, and assessed risk of bias using study design-specific tools (ROBINS-I, Newcastle-Ottawa, QUADAS-2, RoB-2). The GRADE assessment was used for certainty of evidence.
Of 1,179 records, 38 studies were included (10 retrospective, 18 prospective, 3 cross-sectional, 3 RCTs, others). CA 125 was the most frequently assessed marker, followed by CA 19-9, CEA, CA 15-3, and AFP. Across studies, sensitivity and specificity varied, but positive predictive value (PPV) was consistently low in general asymptomatic populations. False-positive results frequently led to imaging, invasive procedures, and surgery, with complication rates up to 15%. No mortality benefit was demonstrated in large RCTs. Certain subgroups (e.g., new-onset diabetes with elevated CA 19-9) showed improved PPVs, but evidence was insufficient to support population screening.
Routine use of serum tumor markers for cancer screening in asymptomatic individuals is unsupported by evidence and may cause harm. We propose a structured follow-up framework to guide the evaluation of incidentally elevated tumor markers, aiming to reduce unnecessary interventions and optimize resource use.
We conducted a systematic review of PubMed, Cochrane Library, and LILACS from inception to May 10, 2025, identifying studies evaluating the use of the aforementioned tumor markers in asymptomatic populations. Eligible studies reported diagnostic accuracy or clinical outcomes. Two reviewers independently screened, extracted data, and assessed risk of bias using study design-specific tools (ROBINS-I, Newcastle-Ottawa, QUADAS-2, RoB-2). The GRADE assessment was used for certainty of evidence.
Of 1,179 records, 38 studies were included (10 retrospective, 18 prospective, 3 cross-sectional, 3 RCTs, others). CA 125 was the most frequently assessed marker, followed by CA 19-9, CEA, CA 15-3, and AFP. Across studies, sensitivity and specificity varied, but positive predictive value (PPV) was consistently low in general asymptomatic populations. False-positive results frequently led to imaging, invasive procedures, and surgery, with complication rates up to 15%. No mortality benefit was demonstrated in large RCTs. Certain subgroups (e.g., new-onset diabetes with elevated CA 19-9) showed improved PPVs, but evidence was insufficient to support population screening.
Routine use of serum tumor markers for cancer screening in asymptomatic individuals is unsupported by evidence and may cause harm. We propose a structured follow-up framework to guide the evaluation of incidentally elevated tumor markers, aiming to reduce unnecessary interventions and optimize resource use.
Authors
Matsas Matsas, Bueso Pineda Bueso Pineda, Saowapa Saowapa, Ortiz Maldonado Ortiz Maldonado, Molina Llata Molina Llata, Payne Payne, Buscemi Buscemi, Farooq Farooq, Karri Karri, Del Giglio Del Giglio
View on Pubmed