Prenatal Exposure to Chemotherapy Increases the Mutation Burden in Human Neonatal Hematopoietic Stem Cells

Abstract

This study demonstrates that environmental mutagenic exposure during pregnancy can increase somatic mutation accumulation in the fetus. Given that detrimental early life exposures can adversely affect health outcomes later in life, our study highlights the need for further research into the impact of environmentally induced genomic insults during the perinatal period. See related commentary by Furudate and Takahashi, p. 870.

Figure 1.

HSPCs from cord blood of newborns prenatally exposed to chemotherapy display an increased mutational load. A, Total somatic SBS counts in neonatal HSPCs from healthy pregnant women (n = 9), pregnant patients with cancer not treated with chemotherapy (n = 4), and chemotherapy-treated pregnant patients with cancer (n = 13). Every dot represents the SBS count identified in one clone. B, Total somatic SBS counts in neonatal HSPCs were plotted per patient. Every dot represents the SBS count identified in one clone. C, Total somatic SBS counts in neonatal HSPCs were plotted per treatment regimen. Every dot represents the SBS count identified in one clone. D, Association of the mutational burden, shown as the number of SBSs per clone, with the amount of chemotherapy administrations received, colored per treatment regimen. The increase of SBS count related with a higher amount of administrations was mostly driven by the two samples exposed to a high number of (EC+)TCpt administrations. ns, not significant. *:p<0.05, **:p<0.01, ***:p<0.001.

Figure 2.

Platinum-based compounds show a direct mutagenic effect in HSPCs from neonates prenatally exposed to chemotherapy. A, 96-trinucleotide SBS profile constructed from the SBSs identified in neonatal HSPCs from healthy pregnant women (n = 9), pregnant patients with cancer who were not treated with chemotherapy during pregnancy (n = 4), and chemotherapy-treated pregnant patients with cancer receiving different regimens [n = 6 treated with ABVD, n = 4 treated with EC/A(+T), and n = 3 treated with (EC+)TCpt]. B, Contribution of mutation signatures per treatment group. C, Contribution of the mutation signatures identified per subject, showing the absolute number of mutations explained by the different signatures. For HSPCs exposed to (EC+)TCpt, an evident increase of contribution of the platinum signature SBS31 was observed in relation with an increased number of administrations during pregnancy. D, Contribution of DBS signatures per treatment group. E, Cosine similarity of the DBS mutational profiles identified per treatment group with DBS5 (DBS associated with platinum-based therapy), showing similarity only with the (EC+)TCpt group.

Figure 3.

In vitro validation of the platinum-specific signature after direct exposure of healthy HSPCs. A, Setup of an in vitro assay to determine the mutagenic effects of chemotherapeutic agents in healthy cord blood HSPCs. B, Number of SBSs in the different chemotherapy-exposed and control HSPC from cord blood. C, 96-trinucleotide SBS profile constructed from the SBSs identified in healthy HSPCs in vitro exposed to 1 μmol/L cisplatin, 10 μmol/L maphosphamide, or 10 nmol/L doxorubicin. D, Signatures from the in vitro exposures and COSMIC with a cosine similarity >0.75 were merged and used for refitting of the mutations identified in the clinical setting, leading to five different signatures explaining the somatic mutations in the clinical samples. *:p<0.05, **:p<0.01, ***:p<0.001.