Skip to main content
Premium Trial:

Request an Annual Quote

Roche Drug Shows Benefit in Pediatric Patients With Rare, Molecularly Defined Tumors


NEW YORK (GenomeWeb) – Pediatric cancer patients with a variety of tumors harboring NTRK, ROS1, and ALK alterations saw their tumors shrink after treatment with entrectinib, an investigational drug Roche is studying in a pan-cancer indication.

The Phase I/IB study, which was partly a dose-finding trial and a basket trial to explore the safety and efficacy of entrectinib across tumor types, enrolled 29 patients aged between 5 months and 20 years with tumors affecting the brain and spine, neuroblastoma, and other solid tumors. In this cohort, 12 patients had NTRK, ROS1, or ALK fusions, one patient had an ALK mutation, and 16 patients did not have these alterations.

"Entrectinib produced striking, rapid, and durable objective responses in all children with refractory central nervous system and solid tumors that actually harbored these fusions. [The drug] also produced a significant response in one ALK-mutated neuroblastoma patient," Giles Robinson, a pediatric neuro-oncologist at St. Jude Children's Research Hospital, said during a call hosted by the American Society of Clinical Oncology to preview data from studies ahead of its annual meeting later this month. "No responses were seen in tumors lacking aberrations in target kinases."

The molecular alterations harbored by these young patients are known to drive cancer, but they also occur infrequently. Robinson, who led the Roche-sponsored study, noted that while patients in this trial had very rare tumors, it is getting easier to identify individuals with these mutations because of increased access to next-generation sequencing.  

"So, we are noticing these fusions in tumors where we didn't expect to see them before," he said, adding though that even with increased screening, these molecularly defined tumor types are not going to be blockbuster indications.

NTRK fusions, for example, show up in less than 0.5 percent of solid tumors. However, because these alterations occur in a variety of tumor types, it makes NTRK fusions a good biomarker for tissue-agnostic drug development.

The US Food and Drug Administration last November approved Loxo Oncology and Bayer's larotrectinib (Vitrakvi) for adult and pediatric patients with advanced solid tumors with NTRK fusions who are out of treatment options. This was the second tissue-agnostic drug approval after the agency approved Merck's pembrolizumab (Keytruda) in 2017 for patients with microsatellite instability or mismatch repair deficiency.

The approval of larotrectinib really paved the way for other tissue agnostic drug approvals and is a "major milestone in precision oncology," said Vivek Subbiah, associate professor of investigational cancer therapeutics and an expert in treating pediatric cancers. "Entrectinib is following that model and adds more targets."

The FDA recently granted a priority review designation to Roche's new drug application for entrectinib as a treatment for adult and pediatric patients with NTRK fusion–positive locally advanced or metastatic solid tumors. The agency is slated to make a decision on that NDA by August 18.

Josina Reddy, senior group medical director at Roche subsidiary Genentech, said that for the NTRK fusion-positive solid tumor indication, the company has submitted data to the FDA, the European Medicines Agency, and the Japanese Health Authorities from the present trial, STARTRK-NG, along with results from an integrated analysis of the Phase II STARTRK2, Phase I STARTRK1, and PHASE I ALKA-372-001 trials.

"Inclusion of pediatric patients was always part of our clinical study plan for entrectinib," Reddy said. "Our key objective was to extend the benefit of entrectinib to as many patients with NTRK gene fusions as possible."

Subbiah highlighted that the responses to entrectinib seen in central nervous system tumors is particularly encouraging. What the STARTRK-NG study demonstrates, he added, is that drugmakers should consider pediatric populations very early in their drug development efforts and in particular should include these populations for tissue-agnostic drug development.

While Roche/Genentech would not comment on whether it would try to extend the tissue-agnostic indication for entrectinib beyond NTRK fusion-positive solid tumors to include the cancers with ALK and ROS1 fusions, such a strategy would allow the company to grow the market for an orphan drug. The FDA held a public workshop last year to discuss how the current orphan drug designation program may be used to encourage drugmakers to advance tissue-agnostic therapies and plans to put forth guidance on the topic.

The STARTRK-NG study, Robinson said, continues to enroll patients, but based on the findings so far will only recruit those with NTRK, ALK and ROS1 fusions. He highlighted that in this study patients had some very rare tumors, such as neuroblastoma, pediatric high-grade gliomas, and infantile fibrosarcoma, a cancer that forms on the limbs of young children and can be deforming.

Robinson put up the brain scans of high-grade glioma patients with NTRK and ROS1 fusions to show the dramatic tumor shrinkage they had in response to entrectinib. "I predict that all these [cancers] would be universally fatal if they didn't respond to this medicine," he said.

While patients generally tolerated the drug well, they did experience some side effects such as fatigue, elevated creatinine levels, loss of taste, and weight gain. Weight gain is not a common side effect of cancer medicines, but this became a worry for some of the patients on the trial.

In the STARTRK-NG study, a variety of molecular tests — such as next-generation sequencing, Sanger sequencing, and RT-PCR, as well as tests from NanoString and HTG Molecular Diagnostics — were performed at CLIA-certified labs to determine patients' NTRK or ROS1 status. Reddy noted that patients enrolled in the trial based on results from local molecular testing labs had to submit archival tumor tissue taken at diagnosis or upon disease relapse for testing at a central lab.

"Molecular testing, including next-generation sequencing, will play a critical role in identifying individuals most likely to benefit from entrectinib," Reddy said, adding that Genentech is working with Foundation Medicine to add NTRK and ROS1 genes fusions to the companion diagnostic indications for FoundationOne CDx. "If approved, this version [of the NGS panel test] will become the companion diagnostic test for entrectinib."

Another study researchers presented results from ahead of the ASCO annual meeting was the National Cancer Institute's MATCH trial in pediatrics, a basket trial in which patients with certain genetic alterations are receiving investigational drugs in a tissue-agnostic fashion. An interim analysis showed that 24 percent of 400 patients screened harbored alterations that made them eligible to receive at least one drug being studied in the trial. Prior to starting the study, researchers had estimated that 10 percent of young cancer patients would match to a drug arm.

Although increased access to sequencing and the availability of basket studies is improving the understanding of the molecular landscape of pediatric cancers, children have been challenging to treat with precision oncology approaches because they often lack targetable tumor markers. Researchers are hoping to address this problem, however, by looking at other features of children's cancers beyond looking for specific genetic alterations.

For example, researchers with the Treehouse Childhood Cancer Initiative are using RNA sequencing data to analyze entire genetic pathways involved in driving cancers and hoping that drugs effective against these pathways will benefit pediatric patients who don't harbor tumor mutations.

"The example I presented [from the STARTRK-NG study] is on the extreme and good side of things," Robinson said. "But … there are many tumors being sequenced that either do not have a targetable mutation or don't have any mutations in the disease." He said there needs to be stepped-up efforts to "figure things out" and give children with cancer more treatment options.

Another important consideration in pediatric cancer drug development is the method of administration. Noting that one of the side effects to entrectinib in the study was loss of taste, Subbiah said this may make it more challenging to administer the drug in kids. "We need to understand the biological basis of response to this drug to see if we can address that question about taste," he said.

If drugmakers are to include pediatric populations in early phases of drug development, they should also develop liquid formulations of their therapies to enable administration in children who can't swallow tablets. "We don't even think about this when we develop drugs, but it's very, very important, because otherwise we won't be able to administer these drugs to kids."

A Roche spokesperson said that the formulation used in the STARTRK-NG was a small tablet, but the firm is developing additional formulations.