Diagnosing cancer during pregnancy poses an important ethical–medical dilemma in that therapeutic options for the pregnant woman and fetal welfare should be balanced. Breakthrough clinical studies from our group have proven the short-term fetal safety of chemotherapeutic treatment when administered in the second or third trimester of pregnancy. Since then, chemotherapy treatment during pregnancy became a common prescription. However, since chemotherapeutics are inherently mutagenic and can cross the placenta at rates depending on the compound class, chemotherapy during pregnancy could have detrimental effects to the unborn child’s DNA, an important research question that remains uninvestigated. To address this critical question, in SCAR we scrutinized the genome of cord blood stem cells for the presence of de novo somatic mutations. Cord blood, representing newborn DNA, was obtained at birth from pregnant cancer patients treated with chemotherapy. We included pregnant cancer patients diagnosed with breast cancer or Hodgkin lymphoma, being the two most frequent cancer types diagnosed during pregnancy and standardly treated with a combination of epirubicin, cyclophosphamide and carboplatin (further abbreviated as EC+Cpt) or adriamycin, bleomycin, vinblastine and dacarbicine (ABVD), respectively. As controls we included cord blood from healthy pregnant women and untreated pregnant cancer patients who were not receiving chemotherapy before delivery because of the late identification of the malignancy in pregnancy. Single cord blood stem cells were expanded into individual clones and subjected to whole genome deep sequencing. Strikingly, prenatal exposure to both EC+Cpt and ABVD was associated with a significant increase in the total number of de novo somatic mutations in cord blood stem cells compared to the control samples. Moreover, in the cord blood stem cells obtained from patients that were treated with EC+Cpt during pregnancy, a typical platinum-related signature (SBS31) was identified, pointing to a direct interference of carboplatin with the fetal DNA. Remarkably, in cord blood stem cells from patientsthat were prenatally treated with ABVD, two clock-like mutational signatures(SBS5 and SBS40) were found, which are known to increase with age in healthy cells suggesting that ABVD exposure is inducing general endogenous mutational processes in fetal cells. Lastly, when fitting the mutational load data to published age-related mutational loads in blood stem cells from healthy persons, we found that prenatal chemotherapy exposure to both EC+Cpt and ABVD was associated with a significant increased “mutational age” of up to 5 years in the newborns’ blood stem cell DNA. This unique study gives an unprecedented view on whether prenatal exposure to chemotherapeutic drugs can render to alterations imprinted on the DNA as genetic “scars”. These data, collected at one of the most early stages in life, could be exploit to estimate the potential enhanced risk for these children in developing diseases later in life. Given that the cohort of children, born to pregnant cancer patients, is subjected to a long-term clinical follow-up trajectory by our group, in which assessments of neurocognition, fertility issues and development of diseases are planned until adolescence and adulthood, these data obtained at the early life stages can be linked to long-term outcome data in this cohort and, in turn, can stimulate further large-scale epidemiological studies. Our pioneering work urges for further exploration of the precise correlation between prenatal chemotherapy exposure and genotoxic consequences to the neonatal genome. Furthermore, the results confirm that environmental mutagenic exposure during pregnancy can increase fetal somatic mutation accumulation. This provides new opportunities to investigate the extent of damage induced by other genotoxins to which pregnant women can be exposed.
