Background: Pregnancy-associated haematological malignancy is a rare event; therefore, data available to guide the treatment are scarce. We aimed to evaluate the incidence, overall survival, and maternal morbidity and mortality of women with pregnancy-associated haematological malignancies.
Methods: We conducted a nationwide observational cohort study using the French National Healthcare Data System (SNDS), a health-care administrative database covering up to 99% of the French population. We included all pregnancies in France ending between Jan 1, 2012, and Dec 31, 2022. Pregnancies with terminations or miscarriages managed on an outpatient basis, and women with a history of haematological malignancies before pregnancy were excluded. A Cox proportional hazards model was used to assess overall survival, defined as the date of haematological malignancy diagnosis to either death or the end of the study follow-up, in the haematological malignancy during pregnancy group (pregnancies with a diagnosis of haematological malignancy during pregnancy) compared with the haematological malignancy post-pregnancy group (pregnancies with a diagnosis of haematological malignancy in the year following pregnancy). Severe maternal morbidity was compared in the haematological malignancy during pregnancy group versus the reference group (pregnancies without a history of haematological malignancy or a diagnosis of pregnancy-associated haematological malignancy). Births were classified as very preterm (<32 weeks of pregnancy), preterm (32-36 weeks), and term (≥37 weeks) and compared in the haematological malignancy during pregnancy group versus the reference group. Inverse probability weighting (IPW) was used for confounder adjustment, using maternal age (categorised), comorbidities, socioeconomic status, and year of delivery (as a category).
Findings: Of 9 996 523 pregnancies in 5 995 235 women, 1366 pregnancy-associated haematological malignancies were identified: 413 during pregnancy (4·13 per 100 000 pregnancies) and 953 (9·53 per 100 000 pregnancies) within 12 months of the end of pregnancy (post-pregnancy). No significant differences in overall survival were observed between the haematological malignancy during and post-pregnancy groups across all types of haematological malignancy (IPW-adjusted hazard ratio 0·91 [95% CI 0·62-1·34], p=0·63), specifically for Hodgkin lymphoma (0·56 [0·07-4·53], p=0·59), aggressive B-cell non-Hodgkin lymphoma (0·52 [0·12-2·38], p=0·40), and acute leukaemia alone (0·84 [0·50-1·41], p=0·51). Severe maternal morbidity was more frequent in the haematological malignancy during pregnancy group than in the reference group (86 [26·2%] of 328 completed pregnancies vs 120 335 [1·5%] of 7 945 909 completed pregnancies; IPW-adjusted odds ratio 22·71 [95% CI 17·72-29·10], p<0·0001). We observed an increase in very preterm birth (32 [9·8%] vs 92 712 [1·2%]; IPW-adjusted odds ratio 11·90 [95% CI 7·91-17·91], p<0·0001) and preterm birth (116 [35·4%] vs 430 472 [5·4%]; 11·76 [9·34-14·81], p<0·0001) in the haematological malignancy during pregnancy group compared with the reference group.
