🔬 Tumour Markers in Primary Care — Appropriate Use, Interpretation & Safety-Netting

Tumour markers are among the most misused tests in primary care — they are frequently requested for screening in asymptomatic patients (where they have poor positive predictive value) and their elevated results cause significant patient anxiety and cascade investigations. The fundamental principle that GPs must understand: no tumour marker is sufficiently sensitive and specific to be used as a standalone screening test in asymptomatic patients. The positive predictive value of any tumour marker depends critically on the pre-test probability: a PSA of 4 ng/ml in a 40-year-old with no symptoms has very different implications than the same result in a 70-year-old with nocturia and DRE abnormality. The two appropriate contexts for tumour marker use in primary care are: (1) investigation of symptomatic patients where a specific malignancy is suspected (CA125 for bloating in a post-menopausal woman), and (2) monitoring known malignancy for recurrence (CEA for colorectal cancer surveillance). The single exception to the no-screening rule is PSA for prostate cancer — there is no national screening programme, but a man without symptoms who is concerned can discuss PSA testing with his GP and decide after balanced information about the benefits and limitations (Prostate Cancer Risk Management Programme principle: offer an informed discussion; a man aged 50 or over who still wants a PSA test after counselling may have it. The UK National Screening Committee does not recommend population PSA screening.).

The PSA controversy in UK primary care is one of the most important clinical ethics issues in cancer screening — the NHS does not have a national prostate cancer screening programme (unlike bowel, breast, and cervical cancer) because the evidence for PSA screening reducing prostate cancer mortality is contested (screening trials such as ERSPC suggest a modest reduction in prostate cancer mortality, while ProtecT found similar prostate-cancer survival with active monitoring, surgery and radiotherapy, suggesting many PSA-detected cancers are low-grade and non-lethal). The current UK position (Prostate Cancer Risk Management Programme, PCRMP; separate from the NG12 symptomatic pathway): a man without symptoms who asks about PSA testing should receive balanced information about: (1) the benefits (detecting cancer early, potentially curative treatment), (2) the limitations (false positives causing unnecessary biopsies with complications, overdiagnosis of slow-growing cancers that would never have caused symptoms, PSA elevated by benign conditions), and (3) the uncertainties (whether PSA screening genuinely saves lives in the UK population). After this information, a man aged 50 or over who still wants a PSA test may have it; for any man, discuss his age, ethnicity, family history, BRCA2 status, comorbidity and life expectancy.

The suspected cancer pathway referral for prostate cancer is straightforward for most GPs but a PSA just above the age-specific threshold with a normal DRE is an area where clinical judgement and good documentation matter most — the probability of cancer varies with age, ethnicity, family history and prostate size. A hard prostate on DRE → suspected cancer pathway referral; other features (weight loss, bone pain) raise concern. NICE NG12 uses age-specific thresholds: for a PSA above the range for age, refer using a suspected cancer pathway referral for MRI-directed assessment, discussing preferences and comorbidities; repeat first only if infection or pre-test conditions (recent ejaculation or vigorous exercise, active UTI or prostatitis, instrumentation) may have raised it. The MRI prostate (mpMRI) is now the standard first investigation after referral for suspected prostate cancer in the UK — it stratifies patients by PIRADS score (1–5) and guides targeted biopsy, reducing detection of clinically insignificant low-grade cancer and reducing unnecessary biopsies. The GP's referral letter should include: PSA value and date, DRE findings, LUTS severity, family history, and medication history (5-alpha-reductase inhibitors such as finasteride lower PSA — state this in the referral).

RMI I (Jacobs et al., 1990) combines ultrasound features, menopausal status and CA125, and is used in secondary care after ultrasound to decide MDT referral (NICE CG122) — it is not the central primary-care pathway. In primary care, NICE NG12 (updated April 2026) sets the route: CA125 at or above the age-specific threshold → urgent direct-access USS → suspected cancer pathway referral if the USS suggests ovarian cancer; ascites or a pelvic/abdominal mass on examination → direct suspected cancer pathway referral. The critical point: once CA125 is raised, book the USS at once — do not let the steps drift.

The GTD referral to a designated centre is an RCOG mandate — gestational trophoblastic disease (hydatidiform moles, invasive moles, choriocarcinoma) is a rare but highly treatable condition that requires specialist management and registration with a GTD centre under the current regional arrangements; this centralised model gives high cure rates with appropriate treatment. hCG monitoring after molar pregnancy is managed by the GTD centres. GPs must refer any suspected molar pregnancy (irregular vaginal bleeding, a positive pregnancy test and suggestive ultrasound findings, or persistent hCG after a pregnancy event), after excluding ectopic pregnancy and assessing haemodynamic stability, to the GTD centre BEFORE surgical evacuation where possible, or immediately after, to register the patient in the national surveillance programme.

The AFP + USS liver HCC surveillance programme is recommended in NICE NG50 (cirrhosis, 2016) and NICE CG165 (hepatitis B, 2013), and by EASL (international), for people with cirrhosis and selected people with chronic hepatitis B — the rationale is that HCC detected early by surveillance is more often curable than HCC detected symptomatically. The surveillance interval is 6 months. There is no national AFP threshold for referral: a progressive AFP rise, new lesion or suspicious ultrasound should prompt urgent hepatology discussion, with imaging through the local liver pathway. GPs managing patients with known cirrhosis (from alcohol, NAFLD, hepatitis B/C) should ensure that 6-monthly ultrasound surveillance (± AFP per protocol) is happening — it is frequently omitted in patients not engaged with hepatology. Adding it to the chronic liver disease template in the practice clinical record system ensures regular checking.

The MRC OV05/EORTC 55955 trial finding for ovarian cancer CA125 monitoring is one of the most important counter-intuitive results in oncology — the trial showed that treating ovarian cancer recurrence based solely on CA125 rise (before clinical symptoms) did not improve overall survival, quality of life, or any patient outcome compared to waiting for clinical recurrence. This means that routine CA125 monitoring for ovarian cancer recurrence, while standard practice, should not automatically trigger chemotherapy at the first sign of CA125 rise — the decision to restart treatment should be based on symptoms, quality of life, and a full discussion between the oncologist and the patient. GPs monitoring CA125 post-ovarian cancer treatment should: record the result, communicate it to the oncology team, but not alarm patients unnecessarily about a modest rise in CA125 — the decision to treat is a specialist oncology discussion, not an automatic protocol.

The NHS bowel cancer screening programme with FIT (faecal immunochemical test) is one of the most cost-effective cancer screening interventions in medicine — the programme detects colorectal cancers at an earlier stage than they would otherwise be detected, reducing mortality. Participation is incomplete — GPs can significantly improve this by: actively encouraging eligible non-participants at every consultation, removing barriers (explaining the test is simple and does not require dietary restriction), and following up missed invitations. The screening threshold is far higher than the 10 µg Hb/g used for symptomatic patients (NICE NG12 / HTG690) — it is important to understand that a negative FIT does not exclude colorectal cancer in a symptomatic patient. Symptomatic patients with lower GI symptoms should be investigated clinically (quantitative FIT + clinical assessment) regardless of the result of their population screening FIT.

The medico-legal documentation principle for raised cancer markers is essential — a GP who identifies a raised CA125, PSA, AFP, or CEA and fails to act on it, or acts but does not document the action, is at significant medico-legal risk if a cancer is subsequently diagnosed at a late stage. The key documentation requirements: (1) record the abnormal result with date; (2) document the differential diagnosis considered and what prompted the test; (3) document the action taken (referral made, repeat test arranged, patient counselled); (4) document safety-netting given to the patient (what symptoms to return with, timeframe for review); and (5) document that the referral outcome was received and reviewed. A common error is making a suspected cancer pathway referral but not checking whether the patient was seen and the outcome received — the GP has a continuing duty of care to ensure the referral was acted upon and the result reviewed. Electronic clinical systems should flag outstanding referral outcomes for review at a defined interval (e.g., 4 weeks).