Novel Combination Explored in Pediatric Rhabdomyosarcoma

Dieter Zopf

Among the 13 compounds investigated in combination with regorafenib (Stivarga), irinotecan seemed to produce the most favorable response in subcutaneous xenograft models of pediatric rhabdomyosarcomas (RMS), according to findings presented at the 2018 ESMO Congress.

In the study, researchers included established and emerging agents for the treatment of pediatric patients with RMS. These drugs were tested in combination with regorafenib; the multikinase inhibitor is currently being explored in an ongoing phase I trial of pediatric cancers.

Subcutaneous xenografts were crafted from pediatric alveolar (RH30) and embryonal (RD) RMS tumor cell lines in vivo. Mice were treated with 13 compounds, including with irinotecan and vincristine. One mouse was examined in each group in a stepwise dose escalation, monitoring for tolerability and tumor growth.

After 4 weeks of treatment, time to regression was assessed. Results revealed that single-agent regorafenib at a dose of 10 mg/kg/d slowed the growth of both RMS xenografts. Complete responses (CRs) near the maximum-tolerated dose (MTD) were seen in combination with irinotecan in both models, paclitaxel in the RH30 model, and vincristine in the RD model. No drug-related deaths were reported.

Because irinotecan and vincristine demonstrate potent activity as single agents, further studies are needed to attribute the benefit in the study to the combination with regorafenib.

OncLive: Please provide an overview of the study you presented on regorafenib combinations in pediatric rhabdomyosarcomas.

In an interview with OncLive^® during the 2018 ESMO Congress, Dieter Zopf, principal scientist, director, Innovation Sourcing Oncology Bayer Pharmaceuticals, Germany, discussed the identification of effective drug combinations with regorafenib for the treatment of patients with pediatric rhabdomyosarcomas.Zopf: In Europe, there is a requirement by the regulatory authorities that you have to do a pediatric investigation plan (PIP). Since regorafenib has an approved indication in gastrointestinal stromal tumors (GISTs), which also occur in children, Bayer was requested to submit a PIP. Regorafenib is approved in GISTs, but GISTs don’t often occur in pediatric patients. A trial was already running in patients with GISTs, but they could not recruit their official number of patients in time to achieve the PIP.

We needed to negotiate with the European Medicines Agency (EMA) on how to come to an agreement or do this differently. The idea was to test regorafenib in solid malignant tumors in children. This would be a phase I, all-comer study that would investigate all sorts of tumors. Of course, all of the tolerability aspects, which are done in a phase I setting, would be included as well. This PIP constituted 2 different kinds of approaches; one was clinical and one was preclinical.

We were requested to test regorafenib in all sorts of models in these indications and see whether we could obtain antitumor activity in preclinical and clinical settings. We observed some activity in rhabdomyosarcoma. That's why we went back to the EMA and discussed how to proceed. We agreed that we would look into rhabdomyosarcoma. The next issue was connecting with experts and pediatric oncologists on how that can be done.

There was some reluctance to use regorafenib as monotherapy, based on the fact that our results were not dramatic. We didn't see complete remissions or partial remissions (PRs) in preclinical models. We did see some PRs, though not confirmed, in 1 patient in the clinical trial. That, with the knowledge about other receptor tyrosine kinase inhibitors—which mainly have an effect on antiangiogenesis—was not convincing enough for them to agree to a monotherapy trial.

Then, the question was how to get the next step together. We decided along with the Innovative Therapies for Children with Cancer (ITCC) Consortium. They have a large network of clinical institutions to recruit the necessary number of patients in order to do such a study.

After talking to these experts, they provided us with a collection of drugs, which were either standard drugs or were previously used in these kinds of pediatric tumors. We then designed a preclinical study in which we tested the combination of these drugs with regorafenib. This was a bit of a challenge because we had a lot of groups, so [we were tasked with] doing this in a reasonable manner. We decided to use only 1 mouse per group, which is challenging because you have to make sure the mouse survives the study before the study ends. In order to achieve that, we decided to do dose escalate with the combination.

We started with half of the MTD of chemotherapy. Then, we went up to the next dose, which was three-fourths of the MTD. If that was also tolerated, then we moved on to the MTD. That worked out quite nicely. We did not lose a single mouse. It turned out that, from the antitumor activity, that irinotecan was the best drug [to use in combination with regorafenib]. The question was, “How can we identify that the combination of irinotecan with regorafenib is adding some effect?”

If you looked at the growth curve [following treatment with irinotecan alone and irinotecan and regorafenib] the tumor shrinks and goes away. Then we stopped treatment. The plan was to treat for 4 weeks and then look at the time to progression. This took a while. The tumor stayed low or absent for 2 or 3 weeks. Then, we saw that the tumor started to regrow in the irinotecan group, whereas in the regorafenib plus irinotecan group there was no regrowth. This was after almost 80 days.

This made us believe that irinotecan is most likely the best combination partner with regorafenib. Regorafenib was also combined with irinotecan and vincristine in the mice. Though it is more complicated, it worked out as well. We then proposed to clinicians that it would be sufficient to combine regorafenib with irinotecan in the clinic.

As you know, there is physician experience and data from phase III trials showing that vincristine/irinotecan is more effective than monotherapies. That's what we're doing. Currently, we are in the process of doing an action analysis in order to better understand whether there is a molecular basis for the combination of regorafenib and irinotecan. There is some hypothesis around it. We’re also looking into the scheduling. It was tolerated in the mouse, but we also have made observations in the clinic that there is some toxicity.

Is there anything else you would like to mention?

The clinical trial design is such that we have 2 different arms; one arm is giving the irinotecan plus vincristine together with regorafenib, and the other arm is giving chemotherapy followed by regorafenib 1 week later. The latter seems to be better tolerated. We are keeping our fingers crossed that it works out nicely and we get some antitumor effects. We’re trying to get some ideas about scheduling, so that we can preclinically test a large number of different scheduling plans to support future clinical development.It is very interesting as a preclinical researcher to be part of these kind of investigations. Usually we are stuck with preclinical work; we’re rarely involved in the clinical aspects. It's very interesting to talk to the clinicians and get ideas and information about their observations when the drug is given to a patient. It's also important to have this flow of information towards the clinic, but also back. Reverse translations are also another important aspect. I'm trying to improve that.

Tsvetkova Y, Hoffmann J, Zopf D. Identifications of effective drug combinations with regorafenib (REG) for the treatment of pediatric rhabdomyosarcomas (RMS). Ann Oncol. 2018;29(suppl_8). doi: 10.1093/annonc/mdy299.