AUTHORS
Alexander G Goglia, MD, PhD, Miriam Viazmenski2, Subyeta Chowdhury3, Ross L. Levine, Prof. Dr.4
AFFILIATIONS
1 Memorial Sloan Kettering Cancer Center, New York, New York, United States
2 Memorial Sloan Kettering Cancer Center, New York, New York, United States
3 Memorial Sloan Kettering Cancer Center, New York, New York, United States
4 Memorial Sloan Kettering Cancer Center, New York, New York, United States
ABSTRACT
Introduction: Clonal hematopoiesis (CH) is the premalignant clonal expansion of hematopoietic stem and progenitor cells harboring somatic mutations, and carries an increased risk of leukemia, cardiovascular disease, solid tumor progression, and all-cause mortality. While CH is largely an age-related phenomenon, exposure to cornerstone cancer treatments including chemotherapy and radiotherapy can select for and promote the expansion of a subset of CH cells carrying mutations in DNA damage response (DDR) genes, including TP53, PPM1D, ATM, and CHEK2. However, the molecular mechanisms underlying how exposure to these cancer therapies so potently promote clonal expansion remain unclear. Importantly, it also remains unclear whether we can rationally target therapy-related CH (t-CH) in high-risk patients to prevent long-term toxicity and the development of subsequent hematologic cancers.
Methods/Results: To answer these questions, we have developed genetically engineered mouse models of the four most common alleles seen in therapy-related CH (Trp53fl/fl, Ppm1dT476*, Atmfl/+, Chek2fl/+). Our models faithfully recapitulate human t-CH phenotypes, showing dose-responsive clonal expansion after exposure to ionizing radiation or genotoxic chemotherapy. These results are consistent in both ex vivo co-culture models and in vivo genetic mosaicism models (HSC-SCL-Cre-ERT; R26+/tm9[CAG-tdTomato]Hze/J). Age-related CH variants (Dnmt3a-/-, Tet2-/-, and Idh2R140Q) do not expand after genotoxic stress, while expansion of t-CH clones is detectable within 24 hours. We show that DDR mutant t-CH cells outcompete their non-mutant counterparts immediately after DNA damage both by avoiding cell death and by retaining their proliferative capacity, as t-CH populations retain more LK cells, undergo significantly less apoptosis, and repopulate more rapidly than wild-type cells.
In addition, while both DDR mutant and wild-type LK cells complete DNA repair within four days of exposure to DNA damaging cancer therapies, expanded t-CH clones exhibit long-term transcriptional rewiring, allowing them to retain a competitive fitness advantage long after DNA damage has been repaired. RNA-sequencing at eight days after irradiation shows that t-CH clones selectively activate and maintain stem cell identity (FDR <0.001), proliferation (FDR 0.04), and anti-apoptosis (FDR 0.035) gene expression programs.
Finally, using preclinical models, we identify allele-specific therapeutic vulnerabilities that prevent expansion of t-CH clones. In our ex vivo co-culture models, we see that Atr inhibitors (camonsertib, ceralasertib, berzosertib) exhibit synthetic lethality with Atm+/- and Chek2+/- but not Trp53-/- or Ppm1dT476* cells. Conversely, Prkdc (human DNA-PK) inhibitors (peposertib, AZD7648) are synthetic lethal with Trp53-/- and Ppm1dT476* but not Atm+/- or Chek2+/- cells. We extend these findings to our in vivo genetic mosaicism models, as Atr inhibition entirely abolishes the expansion of Atm+/- CH after irradiation, while Prkdc inhibition significantly blunts the expansion of Trp53-/- CH.
Conclusions: Radiation and chemotherapy remain cornerstones of modern oncology, and as long-term survivorship continues to improve across disease sites, it has become increasingly important to minimize the long-term impact of our treatment modalities. Our work provides a mechanistic framework for the emergence of CH after oncologic therapy and opens the door to the development of targeted clinical interventions to prevent and reverse t-CH.
DISCLOSURES
Alexander G Goglia, MD, PhD: No relevant conflicts of interest to declare.