Clinical overview
Cancer complicates roughly one in a thousand pregnancies, and the question the exam wants you to answer is bidirectional: what does pregnancy do to the neoplasm, and what does the neoplasm — and its treatment — do to the pregnancy. In South Africa this is not an academic curiosity. We are a young, high-fertility population superimposed on the world's largest HIV epidemic, and the commonest cancer diagnosed during pregnancy here is cervical cancer, precisely because HPV-driven disease is accelerated by immunosuppression and our screening coverage remains below the WHO 90-70-90 targets. Breast cancer, thyroid cancer, melanoma, lymphoma/leukaemia and ovarian neoplasms make up most of the remainder. A registrar will meet the abnormal smear in the antenatal clinic, the adnexal mass on the dating scan, and the persistently raised βhCG after a "miscarriage" that turns out to be a molar pregnancy.
The central teaching point is that pregnancy is rarely the cause of a worse cancer outcome — stage-for-stage, prognosis is usually equivalent to the non-pregnant patient — but it powerfully modifies detection, investigation and treatment. Physiological breast and pelvic changes mask early signs and delay diagnosis; ionising radiation and many cytotoxics threaten the fetus, especially in the first trimester; and the emotional and ethical weight of balancing two lives demands genuine multidisciplinary, patient-centred decision-making. The skill being tested is the ability to hold maternal oncological benefit and fetal safety in the same frame and arrive at a defensible plan.
Core knowledge
How pregnancy might influence tumour biology
Figure D15.1 — How pregnancy affects neoplasia: hormone-sensitive growth, immune tolerance, vascular/lymphatic boost, diagnostic masking, and placental/fetal metastasis (melanoma).
Several physiological features of pregnancy are theoretically pro-neoplastic, and you should be able to reason through them rather than memorise a verdict.
- Hormonal milieu. Oestrogen, progesterone and prolactin rise enormously. For hormone-receptor-positive tumours (a subset of breast cancer, low-grade endometrial-type lesions, some ovarian sex-cord tumours) this is a plausible growth stimulus, though pregnancy-associated breast cancers are more often receptor-negative and high-grade — partly a biological feature of younger women rather than a hormonal effect.
- Immune modulation. Pregnancy induces a tolerogenic state (shift toward regulatory T cells, altered Th1/Th2 balance) that protects the semi-allograft fetus but may also blunt anti-tumour surveillance. This matters most for immunogenic and virally-driven tumours.
- Angiogenesis and vascular permeability. Pregnancy is a pro-angiogenic, hyperdynamic state (rising VEGF, increased plasma volume and cardiac output), which could theoretically support tumour growth and dissemination.
- Mechanical and lymphatic changes. Increased pelvic and breast vascularity and lymphatic flow are invoked to explain occasional rapid progression, but the evidence that pregnancy genuinely accelerates metastasis is weak for most tumours.
Against this, large series and the dedicated ESGO/INCIP guideline for gynecological cancers during pregnancy (with the International Network on Cancer, Infertility and Pregnancy; first published 2019, updated 2025) conclude that stage-matched survival is not demonstrably worse for cancers diagnosed in pregnancy. The INCIP registry adds the finding that now anchors counselling: children exposed to chemotherapy after the first trimester show congenital malformation, growth, cardiac and cognitive outcomes no different from controls, and cognitive outcome tracks gestational age at birth, not cytotoxic exposure. The dominant clinically observable effect of pregnancy is delayed diagnosis, not accelerated biology.
Cervical neoplasia — the SA-relevant centrepiece
Figure D15.2 — Pregnancy-associated cancers and the cervical-cancer-in-pregnancy pathway (ESGO/ESTRO/ESP 2023): individualised by stage, gestational age and the woman's wishes.
Pregnancy is, for many women, their only sustained contact with the health system, making the antenatal visit a screening opportunity that the SASOG/BetterGyn 2024 guideline and NDoH cervical screening policy explicitly endorse. Use WHO 2020 / LAST terminology: LSIL (≈ CIN 1) and HSIL (≈ CIN 2–3), HPV-associated.
Key biological facts:
- HSIL does not progress to invasion during the pregnancy interval. Cytological "regression" postpartum is well described and partly reflects sampling and the transformation-zone eversion of pregnancy rather than true cure.
- Decidualisation of the cervical stroma can mimic dysplasia or invasion both cytologically and colposcopically — a classic pitfall. The transformation zone is everted and more accessible, which aids colposcopy but increases bleeding.
- HIV (high local prevalence) drives both higher HPV-HSIL incidence and faster progression; pregnant women living with HIV (WLWH) warrant a low threshold to evaluate. Under the May 2026 NDoH elimination framework, all women enter screening from age 25, and WLWH are then screened 5-yearly for life (HIV-negative women 10-yearly); pregnancy does not change the schedule, it is the contact at which to apply it.
For invasive disease, FIGO 2018 staging applies (IA microscopic by depth ≤5 mm; IB by size; nodal disease = IIIC). MRI without gadolinium is the preferred imaging in pregnancy for local extent.
Ovarian and adnexal neoplasms
Most adnexal masses found in pregnancy are benign and physiological or functional (corpus luteum, theca-lutein cysts), and the majority resolve by the second trimester. Among persistent masses, mature cystic teratoma and benign cystadenoma dominate. Malignancy is uncommon (roughly 1–6% of persistent masses); when present, germ-cell tumours and borderline epithelial tumours are over-represented relative to the older non-pregnant population. See Adnexal mass in pregnancy and Ultrasound malignancy signs. CA-125 and many tumour markers (AFP, βhCG, inhibin) are physiologically altered by pregnancy and must be interpreted with great caution.
Gestational trophoblastic disease and neoplasia
This is the one neoplastic process that arises from the pregnancy itself, and it is uniquely important. Complete hydatidiform mole is diploid androgenetic (46XX); partial mole is triploid. Gestational trophoblastic neoplasia (GTN) is defined by persistently rising or plateaued βhCG after evacuation, or by choriocarcinoma/PSTT/ETT. Choriocarcinoma is the malignant trophoblastic tumour that may follow any pregnancy event (mole, miscarriage, term delivery) and characteristically presents with haemorrhagic metastases (lung, brain, vagina). GTN is staged by the FIGO 2026 staging and scoring system, which replaced the 2000 edition: anatomical stages I to IV plus the FIGO-modified WHO prognostic score. The four stages are kept but restated. Stage I is now explicitly disease confined to the uterine corpus and cervix; stage II adds the broad ligaments to the adnexa and vagina; and nodal disease is staged for the first time, with lymph node metastasis below the diaphragm making it stage III and above the diaphragm stage IV. A node counts at a short-axis diameter of 10 mm or more, and the stage carries a notation of "r" for imaging or "p" for pathology (for example, stage III LN r). Surgical nodal staging is not recommended. GTN remains exquisitely chemosensitive (see GTD diagnosis and GTD pathology).