ORLANDO – In a platform trial, AstraZeneca's PARP inhibitor Lynparza (olaparib) plus the investigational ATR inhibitor ceralasertib showed activity in pediatric cancers with homologous recombination deficiency or DNA replication stress.
Researchers presented data from the Lynparza-ceralasertib arm of the ESMART trial, a European Phase I/II platform study, at the American Association for Cancer Research's annual meeting on Tuesday. The preliminary data showed the combination had anti-tumor activity in pediatric neuroblastoma and pineoblastoma. Furthermore, an analysis of responders and non-responders identified potential biomarkers of response or resistance to this combination in the pediatric setting, according to Susanne Gatz, a senior clinical lecturer in pediatric oncology at the University of Birmingham in the UK, who presented the ESMART data at AACR.
"BRCA1 and BRCA2, which are key biomarkers [for PARP inhibition] in adult cancer, do not apply to pediatric cancer," Gatz said, adding that pediatric patients typically don't respond to single-agent PARP inhibitor treatment and may be developing resistance through other genetic mutations.
Researchers hypothesized that combination approaches may work better in pediatric patients with primary resistance to PARP inhibitors. They considered combination strategies with ATR inhibitors not just because they are used in adult cancers to overcome acquired resistance to PARP inhibition, but also because pediatric patients tend to have "replication stress and quite high activation of this ATR pathway," Gatz said.
Among 18 evaluable patients treated with Lynparza-ceralasertib, two patients had confirmed responses, one with neuroblastoma and one with pineoblastoma; and nine patients had stable disease, including patients with neuroblastoma, Ewing sarcoma, osteosarcoma, and papillary spinal cord tumors.
The researchers found that neuroblastoma patients with the 11q deletion appeared to benefit from the combination treatment, which Gatz and colleagues will further assess as a biomarker of response going forward. Gatz noted that patients who had a response or stable disease also had tumors harboring mutations in ATRX, CHEK2, and FANCI. "There are heterozygous mutations which may build a signature to lead to responses," she said.
Among patients who progressed on Lynparza-ceralasertib, researchers flagged TP53 mutations as a possible resistance factor. And they continued to observe lack of response in pediatric patients with BRCA1/2 mutations. "If we look at [one patient with] a BRCA2 pathogenic variant with loss of heterozygosity, which normally would be indicative of response in adult patients, but this patient didn't benefit," Gatz said.
In the trial, Gatz and colleagues established a recommended dose for the combination for patients 12 years and older, which is the same dose adults would get. Gatz noted that her team will continue refining dosage for younger patients.
In a discussion of the ESMART data, Julia Bender, a pediatric oncologist at Memorial Sloan Kettering who was not involved with the study, highlighted that patients with a range of tumor mutations responded to the Lynparza-ceralasertib combination.
"This 11q loss is very interesting," she added. "On 11q are three genes involved with the DNA damage response, and one gene that's involved in homologous recombination, telomere maintenance, and double strand break repair." She noted that up to 40 percent of patients with neuroblastoma have 11q loss, and the biomarker is associated with very poor prognosis. "That begs the question whether chromosomal instability is another biomarker for pediatric cancer," Bender said.
Replication stress, which can be driven by alterations in several genes, including MYC, PAX3-FOXO1, EWSR1, and CCNE1, is another promising factor for selecting patients for treatment with the PARP-ATR inhibitor combination, Bender pointed out.
"Replication stress is what seems to be the most important [biomarker] in pediatric tumors," Bender said. "Pediatric translocation-driven tumors are sensitive to ATR monotherapy in in vivo models. Single-agent olaparib does little [in these models]; the ATR inhibitor actually causes growth suppression and a little regression, but the two together is synergistic."
In Bender's view, however, further study is needed to evaluate how the Lynparza-ceralasertib combination compares to conventional doses of chemotherapy in pediatric cancer patients. Additionally, more work is needed she said to establish where in the treatment pathway the combination should be used, after considering the potential benefits of giving it to patients in the adjuvant or maintenance settings.
"The biomarker profiles are going to be complex, content dependent, and likely to reflect a constellation of findings, such as signatures or algorithms [to select patients], rather than single-gene alterations," Bender said. "The post hoc analysis of responders and non-responders is going to be absolutely critical to understanding these biomarkers and the role of DNA damage response inhibitors in pediatrics."