Technological advancements in molecular tumor profiling are starting to make a difference in pediatric oncology, where the search for new therapies is challenging and frustrating.
Rajen Mody, MBBS
The relative rarity of pediatric cancers has hampered the work of uncovering their causes and developing new therapies, but efforts to expand the use of genomic and biomarker testing are beginning to provide new options for children and their families, as well as more information to help tailor therapeutic plans.
In particular, investigators are looking to target markers that may occur in multiple tumor types in hopes of finding better treatments for recurring and refractory disease and for rare cancers that carry poor prognoses.
A prime example of this trend is larotrectinib (LOXO-101), a pan-TRK inhibitor with FDA priority review status for adult and pediatric patients with locally advanced or metastatic solid tumors with oncogenic fusions involving TRK proteins, which are encoded by NTRK genes. In a trial including pediatric patients with infantile fibrosarcoma, other soft issue sarcomas, and papillary thyroid cancer, larotrectinib induced an objective response rate (ORR) of 93% in those with TRK fusion—positive malignancies (n = 15).1 The median age of patients in the cohort was 4.5 years (range, 1 month-18 years).
“NTRK is a very good target. It is driving so much of a tumor’s growth that when you treat, it impacts the tumor’s clinical behavior. It’s not a very frequent target, but it’s clearly a very clean target,” said Rajen Mody, MBBS, a pediatric oncologist at the University of Michigan C.S. Mott Children’s Hospital in Ann Arbor.
Larotrectinib has excited interest not only because of its efficacy and the fact that it is one of the first agents developed as a “tumor-agnostic” drug, but also because its developer, Loxo Oncology, took the unusual step of including both adult and pediatric patients in the first trials. TRK fusions are rare, occurring in about 0.5% to 1% of cancers overall, and Loxo had to cast a wide net to populate its trials with patients who harbored the abnormality.
Larotrectinib’s development suggests a path that more new agents could take. It also points to the importance of expanding molecular marker testing of pediatric cancers and including both investigational and existing drugs in young patients in clinical trials, oncologists said.
The National Cancer Institute (NCI) and the Children’s Oncology Group (COG) have mounted a major effort to apply genomic principles to drug development for children and adolescents aged 1 to 21 years with the Pediatric MATCH trial. The study is testing 9 targeted therapies simultaneously in refractory solid tumors in approximately 200 cancer centers in the United States (Figure).2
“By definition, any pediatric tumor is an orphan disease. Now all of sudden people are realizing that with these molecularly targeted agents, there’s a very, very small subgroup of patients who are going to have the marker, and they want to find the small populations,” said Julia L. Glade Bender, MD, vice chair of pediatric clinical research at Memorial Sloan Kettering Cancer Center in New York, New York. “If you’re looking for the needles in the haystacks, you need more haystacks.”
Pediatric cancers make up less than 1% of diagnosed cancers. In children up to 14 years old, leukemias and tumors of the brain and central nervous system are most common, followed by neuroblastoma, Wilms tumor of the kidney, lymphomas, rhabdomyosarcoma, retinoblastoma, and bone cancers (osteosarcoma and Ewing sarcoma).3 More than 80% of children with cancer survive 5 years or more, but survival rates are poor for those with certain diseases such as diffuse intrinsic pontine glioma and sarcomas with metastatic disease.4
The different biology of tumors in children and adolescents creates both special challenges for research on precision oncology and possible opportunities, said Eugene Hwang, MD, a pediatric neuro-oncologist at Children’s National Medical Center in Washington, DC. “Mutations in children tend to be on the order of a magnitude fewer than they are in adults. In some ways that hurts us because we don’t actually have a lot of different mutations to go after,” he said. “In some ways maybe you could say that might help us because anything that’s there is more likely to be important than it is in an adult cancer that has so many mutations, where maybe you have too many targets to choose from and not all of them are important.”
If the pharmaceutical industry’s enthusiasm for targeted drugs boosts development of pediatric therapies, that would represent a change from past practice. Pediatric cancer drugs are less profitable than adult drugs, and pediatric-specific drug development has been uncommon. Young patients are often excluded from trials because of safety concerns or worries that a serious adverse effect (AE) in a child could torpedo a promising new drug, Hwang said. The absence of clinical trial data means the different tumor and AE profiles seen in children go unexplored, and child-friendly liquid formulations are often unavailable.
Although Hwang and others say they remain frustrated by the slow pace of drug development, regulatory changes and advances in testing technology are chipping away at the lag in research. Over the past 2 decades, federal laws and regulations have mandated pediatric studies for certain drugs and provided incentives for the development of new drugs for rare pediatric diseases.5 As genomic testing has become important in cancer treatment generally, pediatric oncologists have been quick to follow suit, initiating projects to integrate gene sequencing into clinical management and contributing to catalogs of genomic alterations.
Mody and his colleagues were early adopters, establishing a genetic testing program in 2012 for pediatric patients with refractory or relapsed cancer. An early study found that 46% of 91 patients aged 0 to 22 years had potentially actionable findings after their tumor tissue was sequenced.6 For 25% of these patients, clinicians were able to change their treatment or refer them to genetic counseling.6
In 2014, NewYork-Presbyterian Columbia University Medical Center began sequencing all patients with high-risk pediatric cancers and now uses foundation funding to provide the service to all qualifying New York—area children, said Glade Bender, who previously was medical director of the sequencing program.
Genetic analysis is becoming more common at large pediatric cancer centers, although Hwang noted that many patients are still not tested, especially at smaller centers where many children are treated. “That’s a problem,” he said. “It’s expensive to do genomic testing, only a few places can do it, and it takes time.” In addition, tissue samples are often not preserved or fully analyzed, depriving investigators of data needed to develop new therapies, although foundation-funded groups like the Children’s Brain Tumor Tissue Consortium are working to fill that gap, Hwang said.Chemotherapy, stem cell transplant, surgery, and radiation remain mainstays of pediatric oncology and hematology, but a small list of targeted therapies is available to treat cancers with mutations identified through genetic testing, particularly gene fusions, which Mody said drive 35% of pediatric cancers.
They include one of the first targeted therapies, imatinib (Gleevec), a tyrosine kinase inhibitor used to treat Philadelphia chromosome—positive chronic myeloid leukemia (Ph+ CML) and acute lymphoblastic leukemia (Ph+ ALL). Two successor drugs, dasatinib (Sprycel) and nilotinib (Tasigna), were also approved for pediatric Ph+ CML over the past year.
More targeted therapies could be approved for pediatric indications in the near future. Larotrectinib appears to be moving toward approval for treatment of locally advanced or metastatic solid tumors harboring a TRK fusion.
Mody said another promising drug is trametinib (Mekinist), a MEK inhibitor used to treat BRAF-mutant metastatic melanoma and non—small cell lung cancer in adults that has produced responses in some pediatric cancers.7,8 Active trametinib trials include studies of patients with relapsed or refractory juvenile myelomonocytic leukemia and of children and adolescents with solid tumors that have stopped responding to treatment or recurred, including Ewing sarcoma, rhabdomyosarcoma, neuroblastoma, and brain tumors.9,10
Other previously approved drugs that may become part of the pediatric toolkit include crizotinib (Xalkori) and ruxolitinib (Jakafi), Glade Bender said. Crizotinib produced ORRs of 83% and 90% among pediatric patients with relapsed/ refractory ALK-positive anaplastic large cell lymphoma treated with 165 or 280 mg/m2, respectively, and 86% across 3 doses for those with inflammatory myofibroblastic tumors, according to study findings. In all, 40 patients ranging in age from 3.7 to 20.8 years were treated during the trial.11 Crizotinib is being studied in at least 17 trials that include juvenile patients, according to ClinicalTrials.gov.
Ruxolitinib demonstrated a 45% ORR in findings from a small study (n = 11) of pediatric patients who underwent allogenic hematopoietic stem cell transplant and developed steroidrefractory acute graft-versus-host disease. The JAK inhibitor is in clinical trials for treatment of that condition and pediatric ALL.12
A very rare example of a pediatric-only approval is that of the immunotherapy drug dinutuximab (Unituxin), the first therapy specifically approved for neuroblastoma. It was developed by the NCI and COG and distributed by the NCI under a compassionate use exemption beginning in 2009. In 2015, the FDA approved the drug in combination with IL-2, granulocyte-macrophage colony-stimulating factor, and 13-cis-retinoic acid for pediatric patients with high-risk neuroblastoma who responded to prior induction therapy.
“Few pediatric-specific drugs have been making it to the market. You can give 1 or 2 examples, while 20 other drugs get approved for adult indications, so we need more examples like that,” Mody said.
Although oncologists often remark on the proliferation of cancer-signaling gene abnormalities and other biomarker targets for which they lack therapies, pediatricians bemoan the investigational drugs that exist in labs but remain out of their reach because the manufacturers declined to pursue pediatric indications.
“It is frustrating, because we don’t get to try the innovations that are coming out of the most recent discoveries, which is hard,” Hwang said.
Glade Bender described the winding path taken by ganitumab (AMG 479), an insulin-like growth factor 1 receptor antibody that produced some positive results in Ewing sarcoma several years ago. “It didn’t hit in all the big adult cancers and everyone shut down their programs. We never answered the Ewing question, and clearly there was something,” she said. Determined not to let a potentially important therapy go untested, she and other investigators affiliated with COG decided to “go for broke,” arranging to purchase the small amount of leftover drug in order to set up a phase III randomized trial, she said.13To speed advances in new medicines for pediatric cancer, COG and NCI emulated the NCI-MATCH basket trial and launched Pediatric MATCH for treatment of solid tumors in July 2017. The trial is testing 3 drugs approved for treatment of adult cancers—the PARP inhibitor olaparib (Lynparza), the CDK4/6 inhibitor palbociclib (Ibrance), and the BRAF kinase inhibitor vemurafenib (Zelboraf)— along with larotrectinib and 5 other investigational agents. Mody, a COG investigator, said the group selected additional agents but their manufacturers declined to participate for various reasons.
Unlike NCI-MATCH, Pediatric MATCH is so far testing only single-agent therapies, not combinations. Glade Bender and Hwang acknowledged the need for single-agent studies to provide clean safety and response data for each drug, but they said experimental monotherapy usually does not produce robust responses, particularly in children who have relapsed multiple times, and they look forward to advancing to combinations. Hwang called such studies a “necessary evil to get through” but also praised Pediatric MATCH for accelerating the process.
“A lot of the slowness in investigating these trials is from the operationalizing of a single trial. There’s so much that goes into writing, developing, getting approved, finding funding, and finally opening a clinical trial that all too often the science has moved very quickly and the rationale for opening that trial has disappeared or been mitigated significantly. The basket trial is a huge advance over serial phase I trials,” he said.
All the Pediatric MATCH drugs are also approved or being studied for adult diseases. Mody said that across oncology drug development, the FDA’s push for sponsors to conduct trials with children and adolescents is beginning to yield results. “Things are definitely improving. At least the testing is going on for a lot of these drugs for children 12 and older, if not for very young infants,” he said. “This is an incremental process. We are excited about discovery, and clearly the pace of discovery has been faster in the last 5 to 7 years compared with the previous 10 or 15 years.”
Hwang said he is looking forward to seeing greater application of approved checkpoint inhibitors such as pembrolizumab (Keytruda) and nivolumab (Opdivo), which are not regularly used in pediatric cancers but have been shown to be safe and potentially useful.14 He is leading a phase I trial of pembrolizumab to treat younger patients (aged 1 to 29 years) with recurrent, progressive, or refractory high-grade gliomas, diffuse intrinsic pontine gliomas, hypermutated brain tumors, ependymoma, or medulloblastoma.15
Mody said the approval last year of the chimeric antigen receptor (CAR) T-cell therapy tisagenlecleucel (Kymriah), for the treatment of patients up to age 25 with B-cell precursor acute lymphoblastic leukemia (B-ALL) that is refractory or in a second or later relapse, was another important advance for pediatric oncology. Studies are testing CAR T-cell therapies for children with neuroblastoma and other solid tumor diseases.16
In addition to improving cure rates and quality of life, the use of novel therapies for certain cancers may eventually reduce reliance on traditional chemotherapy, which can have long-term effects on fertility, kidney function, cognitive function, and other systems in patients who have survived pediatric cancer, oncologists said.
For diseases with high cure rates such as standard-risk pediatric B-ALL and non-Hodgkin lymphoma, some clinical groups are experimenting with lower-intensity regimens to see whether efficacy can be maintained while reducing the toxicity and discomfort of standard treatments. Glade Bender said dramatic responses from targeted therapies like larotrectinib and crizotinib suggest that those patients might do well with less chemotherapy. She noted that genomic testing can also help stratify patients and allow those with good prognoses to forgo more toxic therapies.
Yet Hwang also offered the counterexample of a trial he heard about that attempted to lower the chemotherapy dose for infants with a medulloblastoma with high cure rates but had to be quickly halted when the cure rate started to fall.
“People are trying to reduce doses, but unfortunately you don’t always have that option,” Mody said.