In a population-based cohort study of 24307 premenopausal women, pregnancy-associated cancers had increased overall 5-year mortality, although not all cancer sites presented the same risk. Mortality was increased in those with any pregnancy-associated cancer after adjusting for age at diagnosis, stage at diagnosis, cancer site, and days to first treatment. When examining pregnancy and postpartum cancer diagnoses separately, individuals diagnosed during pregnancy had elevated mortality, but this association was not consistent across all cancer sites.

One-year survival was generally comparable between all groups, whereas 5-year survival was reduced in individuals with any pregnancy-associated cancer. More research on each cancer site is required to provide robust evidence to guide counselling and clinical care of affected patients according to Dr. Amy Metcalfe of the Department of Obstetrics and Gynecology, University of Calgary in Calgary, Canada, and colleagues who published the findings on 6 April 2023 in the JAMA Oncology.

The authors wrote in the background that cancer is the second leading cause of death in premenopausal women, with pregnancy-associated cancer, defined as cancer diagnosed during pregnancy and 1-year postpartum, estimated to occur once in every 1000 pregnancies. Multiple studies across Australia, Europe, and North America have observed an increase in the incidence of pregnancy-associated cancer over time. This rise may be attributable, in part, to increases in delayed childbearing, improvements in diagnostic testing, awareness of genetic risk factors and subsequent screening, and expansion of population-based screening programmes.

Pregnancy-associated cancers are overwhelmingly diagnosed in the postpartum period, with only 25% diagnosed during pregnancy. This timing of diagnosis could be attributable to an increase in health care encounters in the antenatal and postnatal periods, or possibly due to physician hesitancy in performing potentially harmful diagnostic tests during pregnancy, resulting in delayed diagnosis.

Pregnancy-associated breast cancers have been well established as having an increased mortality compared with breast cancers diagnosed outside of pregnancy or postpartum. One possible explanation for poorer survival associated with pregnancy-associated breast cancer is that hormonal changes in pregnancy, delivery, and lactation enhance the growth of malignant cells, producing a more aggressive and advanced disease at time of diagnosis. Diagnostic delay may also play a role because common physiologic changes during pregnancy and postpartum may mimic early physical symptoms of cancer.

However, evidence on long-term health outcomes following diagnosis of cancer other than breast cancer is limited. The lack of research on cancers other than breast cancer precludes an evidence-based discussion between physicians and affected individuals about prognosis and treatment options.

The objectives of the current study were to determine long-term mortality in premenopausal women with pregnancy-associated cancers compared with premenopausal women diagnosed with cancer remote from pregnancy, overall and stratified by cancer type, and to assess 1- and 5-year survival rates of those affected.

This population-based retrospective cohort study included premenopausal women aged 18-50 years living in 3 Canadian provinces (Alberta, British Columbia, and Ontario) diagnosed with cancer between 1 January 2003 and 31 December 2016, with follow-up until 31 December 2017, or date of death. Data analysis occurred in 2021 and 2022. Participants were categorised as being diagnosed with cancer during pregnancy (from conception to delivery), during the postpartum period (up to 1 year after delivery), or during a time that was remote from pregnancy.

Outcomes were overall survival at 1 and 5 years and time from diagnosis to death due to any cause. Cox proportional hazard models were used to estimate mortality adjusted hazard ratios (aHRs) with 95% confidence intervals (CIs), adjusting for age at cancer diagnosis, cancer stage, cancer site, and days from diagnosis to first treatment. Meta-analysis was used to pool results across all 3 provinces.

During the study period there were 1014, 3074, and 20 219 participants diagnosed with cancer during pregnancy, postpartum, and periods remote from pregnancy, respectively. One-year survival was similar across the 3 groups, but 5-year survival was lower among those diagnosed with cancer during pregnancy or postpartum. Overall, there was a greater risk of death due to pregnancy-associated cancer among those diagnosed during pregnancy (aHR 1.79, 95% CI 1.51-2.13) and postpartum (aHR 1.49, 95% CI 1.33-1.67); however, these results varied across cancer sites.

Increased hazard of mortality was observed for breast (aHR 2.01, 95% CI 1.58-2.56), ovarian (aHR 2.60, 95% CI 1.12-6.03), and stomach (aHR 10.37, 95% CI 3.56-30.24) cancers diagnosed during pregnancy, and brain (aHR 2.75, 95% CI 1.28-5.90), breast cancers (aHR 1.61, 95% CI 1.32-1.95), and melanoma (aHR 1.84, 95% CI 1.02-3.30) diagnosed postpartum.

The authors commented that previous studies on this subject have been limited by small sample sizes and lack of detailed cancer diagnosis data, e.g. cancer stage. Although this study did have data on stage, they were not available for all cancer sites. Furthermore, because the study relied exclusively on administrative data, there are potential confounding factors that the researchers were unable to account for in this study including body mass index, race and ethnicity, oral contraceptive use, socioeconomic status, and other prognostic factors that may influence survival. Similarly, they were unable to evaluate other important long-term health outcomes such as cancer recurrence, which is not routinely captured in cancer registries, and has been shown to have low sensitivity when captured by hospital discharge data. Despite pooling results across 3 provinces, analyses of rarer cancer sites were likely underpowered to detect relevant associations.

This study was funded by an operating grant from the Canadian Institutes of Health Research, (CIHR), New Investigator Grants in Maternal, Reproductive, Child and Youth Health. The CIHR is the federal funding agency for health research in Canada.

Reference

Cairncross ZF, Shack L, Nelson G, et al. Long-term Mortality in Individuals Diagnosed With Cancer During Pregnancy or Postpartum. JAMA Oncology; Published online 6 April 2023. doi:10.1001/jamaoncol.2023.0339.

Source