麻豆传媒

Expert Critique

FROM THE ASCO Reading Room

Brian S. Henick, MD Medical Oncology/Hematology Fellow Yale School of Medicine

Two dosing schedules were tested in a phase I study of DNX-2401 oncolytic virus intratumoral injection for recurrent malignant glioma. A 3+3 scheme was employed for 25 patients across eight dose-escalation cohorts, each given a single injection of stereotactic virus in Group A. Twelve patients enrolled to Group B underwent intratumoral injection 2 weeks prior to primary resection with intramural injection to the margins in four dose-escalation cohorts until viral replication had been achieved. Post-treatment specimens from Group B revealed comparable histopathologic patterns of viral-induced damage to animal models.

Drug-specific adverse events were minimal. Impressively, five patients from Group A at dose levels of 1x108 viral particles or higher experienced durable, disease-free responses lasting longer than 3 years, including three with complete responses (CRs) and two with initial tumor shrinkage followed by stable disease. All CRs occurred in glioblastoma multiforme, two of which recurred after DNX-2401 as gliosarcoma. Notably, the patient still alive 4.5 years after initial virotherapy was retreated upon recurrence. In two instances, new nodules on MRI turned out to be necrotic/inflammatory on biopsy. Two patients in Group B survived for 2 years after treatment. Statistically significant increases of CD4 cells were seen in post-treatment specimens when available, as was down-regulation of TIM-3 and increase in damage-associated molecular patterns. Virus was undetectable in specimens obtained beyond 1 month after treatment.

Taken together, this well-tolerated therapy achieved durable efficacy in a patient population with historical median survival of 6 months. The study was designed well to allow for detailed interrogation of resection specimens. The patterns of response on imaging and histopathology are compatible with induction of anti-tumor immune activity. Because the majority of patients treated did not experience durable benefit, understanding DNX-2401鈥檚 differential efficacy will be key to optimizing its role in glioma treatment paradigms and leveraging virotherapy in other malignancies.

Full Critique

A "smart bomb" oncolytic adenovirus can induce dramatic responses with long-term survival in recurrent high-grade gliomas, according to early results.

The common cold virus engineered to attack recurrent glioblastoma allowed 20% of patients to live for 3 years or longer, probably due to direct oncolytic effects of the virus followed by elicitation of an immune-mediated anti-glioma response.

The altered adenovirus, called Delta-24-RGD or DNX-2401, was injected one time directly into the tumors of 25 patients whose glioblastoma had recurred after surgery and other treatments -- a patient group that typically has a median survival of 6 months.

"We initially thought the oncovirus would kill tumor cells directly through oncolysis," Frederick Lang, MD, director of Clinical Research in the Department of Neurosurgery at the University of Texas MD Anderson Cancer Center in Houston, told ASCO Reading Room. "The idea that it would induce an immune response against the tumor is relatively new. This paper is one of the first clinical papers that provides evidence that this is possible, and sets the stage for further work to understand the mechanism by which this happens and then to exploit it."

The phase I, included 37 patients with recurrent malignant glioma; 25 patients received a single intratumoral injection of DNX-2401 into biopsy-confirmed recurrent tumor to evaluate safety and response across eight dose levels. In addition, a second group of 12 patients underwent intratumoral injection through a permanently implanted catheter, followed 14 days later by en bloc resection to acquire post-treatment specimens.

Of the 25 patients, 20% survived more than 3 years from treatment, and three achieved a complete response (CR). In the three CRs, imaging showed evidence of inflammation and immune activity 1 month after treatment, followed by a steady decline in tumor size until at least 95% of it vanished.

"Importantly, the complete responses were durable, lasting 3 and a half years," said Lang. "In addition, when we removed the tumor, we found the virus in the tumor and an increase in immune-infiltrating cells. Together, in the context of the responses, this shows we induced an immune response against the tumor."

Normally, glioblastomas do not attract the attention of the immune system, with virtually no penetration of tumors by T cells, he explained. Opening up the blood-brain barrier with the oncolytic virus improved pseudo-progression (therapy-mediated tumor enhancement), which is considered a radiographic hallmark of inflammation and provides more correlative evidence that an immune response was mediated, Lang said. "The oncolytic virus appears to induce an immune reaction by calling in the innate immune system by release of danger-associated molecular patterns. With that, macrophages come in and stimulate an adaptive response."

Dose escalation proceeded to the highest concentration of the virus that could be manufactured, with no dose-limiting side effects. Eighteen patients (72%) had some tumor reduction, and median overall survival was 9.5 months.

All three patients who had a CR had recurrences, two of them gliosarcomas, which ultimately proved fatal. All three patients lived for at least 4.8 years after treatment, and two patients had progression-free survival of 42.5 and 36.4 months, respectively.

Toxicities were minimal, with two patients having low-grade adverse events related to treatment, including grade 1 to 2 headache, nausea, confusion, vomiting, and pyrexia. No patients developed systemic side effects. The virus does not get into other organs in addition to the brain so there were no symptoms commonly seen with chemotherapy or radiation therapy, and the patients' quality of life was good, Lang said.

The virus was designed specifically to infect cancer cells, replicate inside those cells to kill them, and spread from cell to cell in a destructive wave throughout the tumor, said the study's senior author and co-inventor of the drug, Juan Fueyo, MD, also of of MD Anderson. In the long-term complete responders, he explained, the virus breaks the tumor's shield against the immune response by killing cells, creating multiple antigen targets for the immune system, and these tumors are then completely destroyed.

However, only 20% of patients responded, Lang noted. "The virus did not get into all patients, or activate their immune systems, but anti-immune molecules, such as programmed death-1 [PD-1] or programmed death-ligand 1, blocked the immune system. Another layer still needs to be overcome." A new multicenter clinical trial is combining the virus with the PD-1 inhibitor pembrolizumab, with 20 patients enrolled so far, and a goal of enrolling a total of 40 by the end of the year.

The team is now working to create next-generation viruses that contain other immune system molecules. "We could rev up the immune system with stimulatory molecules to increase antigen response and T-cell activation and overcome further immune inhibition," said Lang, noting that the can be redirected to tumor cells.

Different viruses may be more or less effective against different tumor types. "A panel of viruses could create a personalized approach and replicate better in different types of tumor, depending on the patient's immune status." Also being investigated are polio viruses, measles virus, and .

This research is opening up a new field within immunotherapy, Lang said: "We need to consider potentially unconventional cancer treatments. The tradition of giving a drug to treat a tumor is not the only way to treat cancer. Harnessing the immune system in different ways might have equal effect."