NEW YORK – Researchers at St. Jude's Children's Research Hospital have identified a genetically defined subtype of pediatric T-cell acute lymphoblastic leukemia with the potential to respond to PARP inhibitors.
As of now, children diagnosed with T-cell acute lymphoblastic leukemia (T-ALL) don't undergo much, if any, biomarker screening before starting treatment, which typically involves chemotherapy and radiation, said Charles Mullighan, deputy director and co-leader of the hematologic malignancies program at St. Jude's Comprehensive Cancer Center.
"T-ALL has very little risk stratification performed, in part because there haven't been studies that have consistently shown [there are] genomic or other biomarkers that are worth implementing in a diagnostic setting, and there are very few targeted treatments," said Mullighan, who is also one of the senior authors on the T-ALL subtype study, which, if other researchers validate, could change the treatment paradigm in this setting.
In the study, which was published this week in Cancer Discovery, Mullighan and coauthors described the subtype, which occurred frequently in patients with gamma-delta T-ALL under age 3 and was associated with "dismal" outcomes, according to the researchers.
In a cohort of 200 children with gamma-delta T-ALL, which researchers pooled from 13 clinical study groups, Mullighan and colleagues identified key genetic alterations associated with patient outcomes. Since these alterations resulted in both STAG2 inactivation and LMO2 activation, the researchers called this the STAG2/LMO2 subtype.
"It's quite a unique alteration that we've really not seen the likes of in other leukemias," Mullighan said. "It actually switches off one gene, STAG2, as well as activates another. It's very unusual and quite striking."
Before identifying the genomic alterations and their clinical implications, Mullighan and his team set off to validate the notion that gamma-delta T-ALL patients had poorer outcomes than those with non-gamma-delta T-ALL. As the name suggests, gamma-delta T-ALL patients express the gamma-delta T-cell receptor and comprise roughly 10 percent of T-ALL cases. Because gamma-delta T-ALL is so rare, Mullighan said it has been challenging to find enough patients for an in-depth molecular characterization of the disease.
"The limitation has been, in part, a numbers game," he said. The St. Jude team collaborated with researchers around the globe, pooling samples and clinical data across 13 centers to conduct the study.
Once the group had their hands on enough samples to study the characteristics and outcomes of the gamma-delta T-ALL cases, and found that these cases did indeed have worse outcomes, Mullighan and his team set out to sequence the cancers.
"We felt very strongly that we perform whole-genome sequencing," Mullighan said. "We knew it was likely to be important, because particularly in T-cell leukemia, a lot of the key drivers occur in noncoding regions of the genome, which can be missed by non-whole genome approaches, and even by RNA sequencing."
The researchers referenced a previous larger study of T-ALL to contextualize their findings. They also created xenograft models that they could use to improve their mechanistic understanding of gamma-delta T-ALL.
"There aren't many suitable models of gamma-delta T-ALL," Mullighan said. "Once we started to find some of the changes like the STAG2/LMO2 rearrangement, we could edit those in or out to look at changes."
Once Mullighan and colleagues identified the STAG2/LMO2 subtype and created the models, they began testing whether the cancer cells would be vulnerable to marketed and investigational therapeutic agents available at St. Jude. They found PARP inhibitors were effective in both cell lines and xenograft models.
"It made sense, given the role of STAG2 in double-stranded DNA break repair, that we saw activity with PARP inhibitors," Mullighan said. "The results were quite striking."
In their paper, Mullighan and colleagues assert that STAG2 inactivation might be the Achilles' heel of the STAG2/LMO2 subtype in that it increases stalled and collapsed replication forks, leading to double-stranded DNA breaks. PARP inhibitors then offer a way to block the pathways that repair these DNA breaks, which in turn overwhelms ALL cells with damage and kills them. The researchers found that Pfizer's Talzenna (talazoparib) was the most potent PARP inhibitor they tested in its ability to inhibit STAG2/LMO2 ALL cell proliferation in their models.
Importantly, Mullighan and his team reported that the STAG2/LMO2 ALL subtype doesn't result in synthetic lethality the way other PARP inhibitor-sensitive cells do because homologous recombination isn't affected. This suggests that there might be an opportunity to pursue combination treatment strategies to effectively treat STAG2/LMO2 ALL.
These findings need further validation, Mullighan cautioned, before his group can start testing PARP inhibitors in this subtype of gamma-delta T-ALL. The difficulty will be to replicate the genetic subtype and associated outcomes in another patient cohort. "The challenge is that many centers don't actually perform diagnostic analysis for expression of the gamma-delta receptor," he said.
As a result, the researchers have recommended that treatment centers incorporate phenotyping for the gamma-delta T-cell receptor in a young child that appears to have T-cell leukemia to see if the cancer is indeed the gamma-delta subtype.
"Then, the next question is, how do you address this therapeutically?" Mullighan said. "This is a rare subset of a rare disease, so this isn't going to be a subset where you can perform a large-scale, randomized Phase III trial to determine effectiveness." More realistically, he said, physician-scientists could seek permission to try a PARP inhibitor in a patient with gamma-delta T-ALL as an n-of-1 approach and document its efficacy.
"I'd anticipate that that's probably the first approach that will be used here," he said. "In the longer term, this is a big question and challenge for the rare pediatric leukemia treatment field: How do you incorporate these promising findings into large trials?"