“There is increasing evidence that vaccines can be efficacious for the treatment of cancer,” said John Sampson, chair of the Department of Neurosurgery at Duke University in Durham, North Carolina.
“Our results indicate that vaccines targeted against tumour-specific mutations might be a successful way to leverage the immune system against cancer, and that even tumours in the brain might be able to be treated by immunotherapy.”
Their study has been published in the journal Clinical Cancer Research.
The epidermal growth factor receptor (EGFR) is an essential component in many signalling pathways that promote cell proliferation and survival.
“The EGFR splice variant EGFRvIII results in a constitutively activated form of the protein, thereby enhancing tumour cell growth,” explained Sampson.
“Roughly one-third of glioblastomas harbour EGFRvIII, and the presence of this splice variant is associated with reduced long-term survival in this patient population,” he said.
Rindopepimut, also known as CDX-110, is a vaccine that specifically targets EGFRvIII.
The EGFRvIII-specific peptide is conjugated to the immunogenic carrier protein keyhole limpet hemocyanin (KLH).
“While results from prior phase II clinical trials found that the addition of this cancer vaccine to standard treatment improved progression-free survival and overall survival in patients with newly diagnosed EGFRvIII-positive glioblastoma as compared with historical control datasets, a phase III study conducted in this patient population did not indicate that adding rindopepimut to standard chemotherapy improved overall survival,” Sampson explained.
Bevacizumab, a VEGF inhibitor, is standard treatment for patients with relapsed, EGFRvIII-positive glioblastoma.
The ReACT clinical trial was designed to study if the addition of rindopepimut to bevacizumab improved progression-free survival for these patients.
This double-blind, randomised study enrolled 73 bevacizumab-naïve patients with EGFRvIII-positive glioblastoma across 26 hospitals in the United States.
Of these patients, 36 were assigned to receive bevacizumab and rindopepimut, and 37 were assigned to receive bevacizumab and KLH as a control injection.
In the vaccine arm, patients received 500µg of rindopepimut admixed with 150µg of GM-CSF; in the control arm, patients received 100µg of KLH.
Vaccination occurred on days 1, 15, 29, and then once monthly.
All patients received bevacizumab (10mg/kg) every two weeks.
Treatment continued until intolerance, withdrawal of consent, or disease progression.
The primary endpoint was progression-free survival at six months (PFS6).
Compared with patients in the control arm, patients assigned to receive rindopepimut had increased PFS6 (16 percent versus 28 percent, respectively), yet the difference was not statistically significant.
However, the 24-month overall survival rate was significantly improved for patients who were assigned to receive rindopepimut compared with those assigned to receive bevacizumab alone (20 percent versus 3 percent, respectively).
Further, the difference in overall survival between the two arms was significant: compared with those in the control arm, those who were assigned to receive the vaccine had a 47 percent reduced risk of death.
Glioblastomas often result in cerebral oedema (swelling in the brain), which can be treated with corticosteroids.
Among patients treated for cerebral oedema at study entry, 33 percent in the rindopepimut arm were able to discontinue corticosteroids for at least six months, compared with none in the control arm.
“An active tumour generates substantial oedema and requires a lot of corticosteroids,” explained Sampson. “We feel that the reduction in corticosteroid use is a surrogate way of identifying patients that are responding to treatment.”
Limitations of the study include its small sample size and a lack of central pathology review.