This phase I trial was expanded to include another cohort of patients who were refractory to ipilimumab and, if they had the BRAF mutation, had previously been treated with a BRAF or MEK inhibitor, or both. They were randomly assigned to pembrolizumab 2 mg/kg or 10 mg/kg every three weeks. Just over a quarter of patients responded to treatment and 58−63% were still alive after a year (see Table).7 The efficacy of pembrolizumab in the phase I trial seemed to be independent of the dose.6,7
An analysis of 146 patients who received pembrolizumab 2 mg/kg found that response rates were better in those who had not previously been treated with ipilimumab compared with those who had (37% vs 26%). The median duration of progression-free survival was also longer (36 vs 22 weeks). At six months, overall survival was similar in ipilimumab-naïve and pre-treated patients (79% vs 83%). This analysis has not yet been published in full.
A randomised phase III trial compared pembrolizumab to ipilimumab. All enrolled patients had advanced melanoma but only 34% had been previously treated with systemic therapy. Pembrolizumab 10 mg/kg every two or three weeks improved progression-free and overall survival compared to ipilimumab. Response rates were also better with pembrolizumab (see Table).8
In the safety cohort of 411 patients, the most common treatment-related adverse events included arthralgia (14.8%), diarrhoea (14.8%), fatigue (30.2%), nausea (10%), pruritus (22.8%), cough (11.1%) and rash (19.8%). Albumin (36.7%), haemoglobin (51.6%) and lymphocytes (28.2%) went down with pembrolizumab. Decreased calcium (28.5%) and sodium (32.6%) concentrations were also observed. Liver function should be monitored as increases in alanine aminotransferase (23.6% of patients), alkaline phosphatase (22.6%) and aspartate aminotransferase (27.7%) were common.
Because of pembrolizumab’s mechanism of action, immune-mediated adverse reactions are a concern. In the safety cohort, these included pneumonitis (12 patients), colitis (4 patients), hepatitis (2 patients) and nephritis (3 patients). Immune-mediated endocrinopathies have also been reported including hypophysitis (2 patients), type 1 diabetes, hyperthyroidism (5 patients) and hypothyroidism (34 patients). Monitoring blood glucose and thyroid function at the start and during pembrolizumab therapy is recommended. Depending on severity of these events, pembrolizumab should be interrupted or stopped and patients should be treated with corticosteroids. Severe infusion-related reactions have occasionally been reported with pembrolizumab and this is a contraindication to further treatment.
More patients discontinued the 10 mg/kg dose than the 2 mg/kg dose because of an adverse event.7 The most common reasons for stopping pembrolizumab were pneumonitis, renal failure and pain.
Pembrolizumab is a category D drug in pregnancy. Although there are no data in pregnant women, blocking PD-1 in animals increases fetal loss. Contraception should be used during and for four months after treatment has finished.
The recommended dose of pembrolizumab is 2 mg/kg every three weeks. Around a quarter of patients with pre-treated metastatic melanoma responded to this dose. Response rates were better in those who had not previously been treated with ipilimumab. Autoimmune adverse reactions are a problem with this drug and regular patient monitoring is vital. Patients do not need to carry the BRAF mutation to be eligible for pembrolizumab.