NEW YORK – Immunotherapeutic drugs have proven to be widely successful in treating a variety of cancers in adults. The development of drugs like pembrolizumab (Merck's Keytruda) and nivolumab (Bristol-Myers Squibb's Opdivo) has led to a push for the development of more histology-agnostic immunotherapeutic drugs, and some of these agents are even listed in the World Health Organization's List of Essential Medicines.
This new generation of therapeutics is also being tested in pediatric cancer patients. But they haven't all been as successful in kids as they are in adults. Checkpoint blockade therapy in particular hasn't made a lot of difference in the fight against pediatric cancer because the underlying biology of pediatric cancers is so different from cancers in adults.
Patients who typically respond to checkpoint blockade inhibitors tend to have an elevated mutation rate — melanomas due to UV damage, or lung cancer and bladder cancer due to smoking-related DNA damage, according to Elaine Mardis, co-executive director of the Institute for Genomic Medicine at Nationwide Children's Hospital. Pediatric patients likely don't respond to these treatments as well as adults do because childhood cancers have significantly lower mutation rates than adult cancers.
That doesn't mean, however, that researchers have given up on the idea of treating pediatric cancers with some form of immunotherapeutic agent, or even specifically with checkpoint blockade inhibitors.
The National Cancer Institute-Children's Oncology Group (NCI-COG) Pediatric Molecular Analysis for Therapy Choice (MATCH) study, which is ongoing, announced interim results in May. COG Study Chair and Baylor College of Medicine Pediatrics-Oncology Professor Will Parsons said the research team is developing protocols to add four additional drugs to the trial.
"There are several areas in which we've discussed or started to strategize evolution of matches. Obviously, we're looking to continue to add single targeted agents that we'd like to study in kids, as well as, conceivably, combinations of therapies," he said in May. "The other thing to consider is checkpoint inhibitors or other immune-based therapies. That's something we're actively considering and have been looking at, but there's not been a decision yet about doing that."
But for such therapies to be successfully developed for the treatment of children with cancer, researchers need to be smarter about the studies they design and which immunotherapies they choose to test, according to Crystal Mackall, a professor of pediatrics and medicine at Stanford University School of Medicine.
"In general, it's probably not the case that immunotherapies don't work in children. It's just that to take the type of immunotherapy that was developed in adults and apply it to children doesn't work," she said. "Immune checkpoint inhibition hasn't worked very well in children, with some with few exceptions. But when you look at CAR T cells, they've actually worked quite well in kids."
The reason for this, Mackall added, is that immune checkpoint inhibitors simply amplify or unleash a response that's already present in people with cancer by taking the brakes off the immune system and letting it fight the cancer. But that response depends on the presence of antigens in the body for the immune system to recognize.
"We don't know for sure what the whole landscape of antigens is, but we know that some of them are mutation-driven antigens. You have to have a lot of mutations in order for the immune system to find a tumor, and in pediatrics we have less mutation," she said. "CAR T cells are completely different. Here, you're hijacking the immune system and tricking it into believing that something is foreign by synthetic biology, by engineering the immune system to recognize a particular antigen. It doesn't rely on mutations."
Because pediatric cancers are largely disorders of development driven by epigenetic programming, that also adds to the problem of targeting them precisely with certain therapies. "If you've got too much of a gene turned on, it's not mutated, it's just got its enhancer open and the gene is over expressed or under expressed," Mackall said. "That's really hard to target because it's not a mutation. You've got just a cell that's poised to behave in a rogue manner."
The answer to this problem is not to test any and all immunotherapies on children, according to Mackall, but rather to be strategic in which drugs to test and how.
For example, Mackall said, combination treatments are proving to be more effective in pediatric cancers than single-agent therapy regimens. "You don't get anywhere without combinations too often. The CD19 CAR Ts were kind of an exception because that was giving one agent. But even now, we're starting to see antigen escape if you only go after one target," she said. "So, I think combinations are really important, just as they are in adult cancer. This is where we get back to the discussion about checkpoints — are we ready to throw them out? Did they fail? Well, they didn't work as single agents, but that doesn't mean that they aren't going to work in some kind of combination."
To that point, Mackall is planning to launch a trial to test the combination of nivolumab and ipilimumab (Bristol-Myers Squibb's Yervoy) in children with cancer. The participants will be stratified based on their tumor mutational burdens (TMB), and only patients with TMBs of 10 mutations per mega base and above will be enrolled, so-called hyper-mutant patients.
About 4 percent to 5 percent of sporadic pediatric cancers are considered hyper-mutant — having TMBs between 10 and 100 mutations per megabase — and these cancers have been associated with response to checkpoint blockade in adults, Mackall said. Whether the combination of checkpoint inhibitors for CTLA4 (ipilimumab) and PD-1 (nivolumab) in molecularly stratified hyper-mutant populations will show activity has yet to be seen, but that's the reason for the trial.
"The combination of nivolumab and ipilimumab has shown increased activity in melanoma, impressively increased activity, so I think the field is very anxious to understand whether that combination might have some activity in pediatric cancer," Mackall said.
She's also currently recruiting pediatric patients for a trial of CD19/CD22 CAR T cells in combination with chemotherapy, to see how well they work in treating children or young adults with CD19-positive B-acute lymphoblastic leukemia that has recurred or does not respond to treatment.
But even within the testing of combination regimens, researchers must be careful in prioritizing the therapies that are most likely to provide positive results. There are so many combinations that it would be impossible to test them all, especially when you start to combine immunotherapies with chemotherapy, Mackall said.
"One of the reasons, arguably, why childhood cancer is harder than adult cancer in order to make progress is that we have to be smarter. We've got to pick the winner. There aren't enough children to test things empirically," she added. "One of the challenges for people in our field is to know when a particular combination has enough signal, or traction, or promise, to warrant testing in children, whether we can infer something from the adult experience, telling us that this is a combination that may work in children."
For instance, Mackall noted, the melanoma space in adults isn't really a helpful guide for pediatric cancers because melanoma is a highly immunogenic and mutated cancer. On the other hand, triple-negative breast cancer is a tumor type with much lower immunogenicity, making it an adult cancer that could provide some insight into similarly immunogenic pediatric tumors.
"We have to be more selective and try to pick the winners by being shrewd rather than by brute force, which is sometimes what goes on in the adults," she added. "They just test everything and see what sticks."
Further, when it comes to developing new drugs for pediatric cancers, it's up to the pediatric oncology community to point pharmaceutical companies in the right direction, to point out when certain trials aren't really indicated, and to suggest combination therapies that seem to have promise in preclinical models, according to Mackall.
The RACE for Kids Act — which states that the US Food and Drug Administration may now require pediatric assessments when the molecular targets of drugs under FDA review are substantially relevant to children's cancers — "is going to change things a lot because historically, companies haven't thought about doing pediatric cancer trials, and so there really wasn't much prioritization [or] many options, she said. "Now there will be many more options, and it's going to be the job of the pediatric oncology community to prioritize the most promising agents and combinations. We don't want to be doing a lot of negative trials — our goal is to have our early-phase trials be positive."
Indeed, prioritizing targets in childhood cancer for testing and treatment is also a concern for Tim Triche, co-director for the Center of Personalized Medicine at Children's Hospital Los Angeles (CHLA). Triche and his team co-developed a targeted sequencing panel in partnership with Thermo Fisher Scientific called the Oncomine Childhood Cancer Research Assay, or OncoKids.
The assay was originally developed to assess the full coding regions of 44 cancer predisposition loci, tumor suppressor genes, and oncogenes; hotspots for mutations in 82 genes; amplification events in 24 genes; and 1,421 gene fusions that have been shown to be clinically relevant in a variety of childhood cancers. Since the first panel was developed, new features of importance have been identified and incorporated, and it now includes 203 unique genes and thousands of fusion drivers, according to Thermo Fisher.
But one issue the researchers faced when developing the panel was deciding which features to include. The most important diagnostic, prognostic, and treatment criteria were prioritized for inclusion, Triche noted, but the panel is not 100 percent comprehensive.
"For example, we had an agonizing discussion about what to do about Hodgkin's disease, and Hodgkin's is not on [the panel]. And the reason is Hodgkin's is a bizarre disease — it's 99 percent normal lymphocytes and 1 percent malignant, and it has a very good prognosis, overall," he said. "So, should you delegate a lot of the content of the panel to go find a 1 percent signal against a 99 percent normal background? Or do you say there are more important things to go after?"
The researchers also decided to omit BRCA1 and BRCA2 from the panel because they're inherited, and the panel was meant to represent somatic alteration. "We have been considering for a long time developing a comprehensive inherited panel, in which case BRCA1 and BRCA2, p53, and multiple others would be critically important content," Triche added. "But I think what is giving us pause for thought is that the pattern of inherited predisposition in childhood cancer is emerging to be somewhat different than adult cancer. So, on the inheritance side we need to give it a lot more study and more work."
He also noted that it would make more sense to prioritize drug targets for pediatric cancers that are more common, in order to benefit the largest group of patients possible.
"The continued development of a pantheon of targeted agents for the more common fusions, and kinases, and so forth I think is probably the better way to go," he added. "The really rare ones are heart rending when you come across one and there's no drug. But if you think about the total number of patients who will benefit, I think the more common ones are going to be the ones that will benefit more patients."
Indeed, a lot of research is being put into immunotherapies such as CAR T cells precisely because they seem to be a better option for pediatric cancer patients. The first FDA-approved CAR T therapy was Novartis' tisagenlecleucel (Kymriah), which is used to treat childhood acute lymphoblastic leukemia.
Mardis noted that Nationwide Children's and many other pediatric hospitals are actively working on developing CAR T cell therapies. "In the engineered cell therapy space, there's a lot of activity at our hospital around engineering natural killer cells, and also to have this unique and highly targeted impact towards cancer cells only," she said. "I think these have a lot more appeal in the pediatric setting because they're highly selective."
But even beyond CAR T, she added, some researchers are looking at using oncolytic viruses such as herpes and other modified viruses that have a tropism towards human cells, and can be used to display unique targets on the surface of the cell that would attract the immune system and signal for cell destruction.
Overall, however, researchers seem to agree that it's still too soon to give up on the possibilities for immune checkpoint inhibitors in pediatric cancer. Mardis pointed to research from a team at Toronto's Hospital for Sick Children, which has shown that some kids have either a germline pathogenic mutation in polymerase epsilon or in their brain tumors which elevates the level of mutations. This type of cancer may respond well to checkpoint inhibition.
"The other aspect that we've been looking at carefully is that when you have secondary cancers that either go through alkylating chemotherapy — which increases the mutation rate in the recurrent tumor — or through radiation therapy — which also gives sometimes not a secondary tumor, but a second primary tumor — there is an elevated level of immune infiltration," she noted. "So, we think by comparing the primary to the secondary, or the primary to the second primary, we may be able to build a case for a clinical trial for kids in the checkpoint blockade space."
Mackall is also hopeful that there's a way to use these drugs effectively in pediatric cancer. "They're incredibly powerful, revolutionary. And it's hard for me to believe that there may not be some way to use their activity to benefit these kids," she said. "It's just that we have to be smarter about either identifying the patients who may benefit and/or using them in combination."