Advanced Melanoma: The Promise and Shortcomings of Drugs Injected Directly into Tumors

In the past 3 years, the treatment landscape for metastatic melanoma has changed dramatically. We saw the advent of drugs that inhibit mutant BRAF and activate MEK proteins (vemurafenib, dabrafenib, and trametinib) and drugs known as immune checkpoint inhibitors (ipilimumab, Keytruda, Opdivo, and others). These treatments are ‘systemic’; that is, they are taken by mouth or injected directly into the bloodstream and spread throughout the body. However, as I reported earlier this year, drugs that are injected directly into tumors—’intralesional drugs’—have recently gained some attention. Two of them were featured at the 2014 American Society of Clinical Oncology (ASCO) Annual Meeting. New data, and doubts, on these drugs have since emerged.

PV-10, also known as Rose Bengal, is an intralesional drug developed by a very small company called Provectus. It was recently tested in a clinical trial as a treatment for stage III and stage IV cutaneous (skin) melanoma. In the trial, PV-10 was injected up to four times into up to 20 tumors per patient over the course of 16 weeks. Final results of this phase II trial have been published.

The drug clearly has anticancer activity: of all the injected lesions (tumors), 51% responded by shrinking and 26% disappeared. This is impressive considering that all 80 patients in the trial had, on average, six previous treatments that all failed. The median duration of response was 4 months, and 8% of patients had no evidence of disease (N.E.D.) for one year. In 28 patients who had all of their tumors injected (meaning they had fewer than 20 tumors), the responses were higher: 50% had complete tumor disappearance, and the overall duration of response was 9.8 months. The limitations of the trial—no more than four injections into no more than 20 tumors per patient—may be responsible for the less-than-spectacular results overall.

Intralesional drugs like PV-10 have a major limitation: it might not be possible to inject all tumors due to inaccessibility, particularly metastases to visceral organs or to the brain. Therefore, PV-10 could be a ‘niche’ treatment for select patients, like those who have only cutaneous or subcutaneous lesions, or older patients who cannot receive systemic treatments because of their poor overall health status.

However, one potentially exciting finding that has emerged from the PV-10 studies is the so-called ‘bystander effect’: it turns out that some of the tumors that were not injected also shrank. The precise cause is unclear, but some research indicates that PV-10 has immune system-stimulating properties. In any case, if the bystander effect holds up in the upcoming phase III trial, the promise of PV-10 will be much improved.

Still, the general value and efficacy of PV-10 remains controversial. It may stop or shrink some tumors, but this does not mean that it will stop the progression of melanoma in a patient. Combining PV-10 with other treatments, like other immune system-boosting drugs, could be effective. But Provectus is small and may not be able to afford to expand its clinical projects.

The other intralesional drug featured at ASCO 2014 is known as talimogene laherparepvec, or T-VEC. Unlike PV-10, a chemical compound, T-VEC is an inactivated and reengineered version of a virus that causes cold sores (HSV-1). T-VEC is designed to infect tumor cells without harming normal cells. It is engineered to produce a protein called GM-CSF, which stimulates the immune system, hopefully to fight melanoma.

Since the ASCO meeting, a phase III clinical trial comparing T-VEC to GM-CSF (given directly, without T-VEC) has been completed. The results were mixed. Statistically, the goal of increased survival with T-VEC was not met, even though there was a trend towards it. Similar to the results of the PV-10 trial, long-lasting responses were observed, as well as the bystander effect (response in the tumors not injected with T-VEC). In short, T-VEC has the same limitations, and promises, as PV-10.

The T-VEC program was originally started by BioVex, a small company that was later purchased by the much larger company Amgen. Amgen now owns and develops T-VEC. With its fairly deep pockets, Amgen is able to push the development of T-VEC in a way that Provectus simply cannot afford to do with PV-10.

Amgen has already launched two clinical trials to explore treatments that combine T-VEC with other immunotherapies. In a small phase Ib trial (NCT01740297) that combined T-VEC with ipilimumab, tumors shrank or disappeared in 56% of patients with metastatic melanoma. The other T-VEC combination trial (NCT02263508) has just begun. It combines the recently approved immune checkpoint inhibitor Keytruda with T-VEC. Moreover, Amgen has now announced the submission of a Marketing Authorization Application to the European Medicines Agency, seeking approval for T-VEC for treatment of melanoma that is regionally or distantly metastatic.

PV-10 and T-VEC share very similar limitations in terms of the patient populations that could benefit from them. So far, the observed efficacy of the drugs does not appear to be dramatically different, and the side effects of T-VEC are worse. Nevertheless, if either drug makes it to the market, there is little doubt about which will be first.