The small phase 1 trial also identified some molecular signatures in responders that may inform future clinical trials.
The results, presented at the annual meeting of the American Association of Cancer Research, came from a single arm of the European Proof-of-Concept Therapeutic Stratification Trial of Molecular Anomalies in Relapsed or Refractory Tumors (ESMART) trial. This trial matches pediatric, adolescent, and young adult cancer patients with treatment regimens based on the molecular profile of their tumors.
In over 220 children to date, the trial has investigated 15 different treatment regimens, most of which are combination therapies.
In adults, PARP) inhibitors have been shown to be effective in tumors with deficiencies in homologous repair, which is a DNA repair mechanism, with notable successes in patients carrying the BRCA1 and BRCA2 mutations. But BRCA1 and BRCA2 mutations are rare in pediatric cancer, and there is a belief that there may be primary resistance to PARP inhibitors in pediatric tumors, according to Susanne Gatz, MD, PhD, who presented the research at the meeting.
Previous research identified alterations in pediatric tumors that are candidates for patient selection. “These tumors have alterations which could potentially cause this resistance effect [against PARP inhibitors] and [also cause] sensitivity to ataxia telangiectasia–mutated Rad3-related inhibitors. This is how this arm [of the ESMART trial] was born,” said Dr. Gatz.
The phase 1 portion of the study included 18 pediatric and young adult patients with relapsed or treatment-refractory tumors. There were eight sarcomas, five central nervous system tumors, four neuroblastomas, and one carcinoma. Each had mutations thought to lead to HR deficiency or replication stress. The study included three dose levels of twice-daily oral olaparib that was given continuously, and ceralasertib, which was given day 1-14 of each 28-day cycle.
Patients underwent a median of 3.5 cycles of treatment. There were dose-limiting adverse events of thrombocytopenia and neutropenia in five patients, two of which occurred at the dose that was recommended for phase 2.
There were some positive clinical signs, including one partial response in a pineoblastoma patient who received treatment for 11 cycles. A neuroblastoma patient had stable disease until cycle 9 of treatment, and then converted to a partial response and is currently in cycle 12. Two other patients remain in treatment at cycle 8 and one is in treatment at cycle 15. None of the patients who experienced clinical benefit had BRCA mutations.
An important goal of the study was to understand molecular signature that might predict response to the drug combination. Although no firm conclusions could be drawn, there were some interesting patterns. In particular, five of the six worst responders had TP53 mutations. “It is striking ... so we need to learn what TP53 in this setting means if it’s mutated, and if it could be a resistance factor,” said Dr. Gatz, an associate clinical professor in pediatric oncology at the Institute of Cancer and Genomic Sciences of the University of Birmingham, during her talk.
Although the study is too small and included too many tumor types to identify tumor-based patterns of response, it did provide some hints as to biomarkers that could inform future studies, according to Julia Glade Bender, MD, who served as a discussant following the presentation and is a pediatric oncologist at Memorial Sloan Kettering Cancer Center, New York.
“The pediatric frequency of the common DNA damage repair biomarkers that have been [identified in] the adult literature – that is to say, BRCA1 and 2 and [ataxia-telangiectasia mutation] – are exceedingly rare in pediatrics,” said Dr. Bender during the session while serving as a discussant. She highlighted the following findings: Loss of the 11q region on chromosome 11 is common among the patients and that region contains three genes involved in the DNA damage response, along with a gene involved in homologous recombination, telomere maintenance, and double strand break repair.
She added that 11q deletion is also found in up to 40% of neuroblastomas, and is associated with poor prognosis, and the patients have multiple segmental chromosomal abnormalities. “That begs the question [of] whether chromosomal instability is another biomarker for pediatric cancer,” said Dr. Bender.
“The research highlights the complexity of pediatric cancers, whose distinct biology could make them more vulnerable to ATR [kinase], [checkpoint kinase 1], and WEE1 pathway inhibition with a PARP inhibitor used to induce replication stress and be the sensitizer. The biomarker profiles are going to be complex, context-dependent, and likely to reflect a constellation of findings that would be signatures or algorithms, rather than single gene alterations. The post hoc iterative analysis of responders and nonresponders is going to be absolutely critical to understanding those biomarkers and the role of DNA damage response inhibitors in pediatrics. Given the rarity of these diagnoses, and then the molecular subclasses, I think collaboration across ages and geography is absolutely critical, and I really congratulate the ESMART consortium for doing just that in Europe,” said Dr. Bender.
The study is limited by its small sample size and the fact that it was not randomized.
The study received funding from French Institut National de Cancer, Imagine for Margo, Fondation ARC, AstraZeneca France, AstraZeneca Global R&D, AstraZeneca UK, Cancer Research UK, Fondation Gustave Roussy, and Little Princess Trust/Children’s Cancer and Leukaemia Group. Dr. Gatz has no relevant financial disclosures. Dr. Bender has done paid consulting for Jazz Pharmaceuticals and has done unpaid work for Bristol-Myers Squibb, Eisai, Springworks Therapeutics, Merck Sharp & Dohme, and Pfizer. She has received research support from Eli Lilly, Loxo-oncology, Eisai, Cellectar, Bayer, Amgen, and Jazz Pharmaceuticals.
From American Association for Cancer Research (AACR) Annual Meeting 2023: Abstract CT019. Presented Tuesday, April 18.