Discussion
The presented results show that PD-L1 score was higher in the PMF group vs the control group regardless of JAK2 mutation status, which turned out to be statistically significant. This helps build on the data reported by Prestipino et al that showed oncogenic JAK2 activity led to STAT phosphorylation which in turn enhanced PD-L1 promoter activity and PD-L1 protein expression in JAK2 mutant cells [1]. In addition, PD-L1 expression was higher on primary cells isolated from patients with JAK2-mutated MPNs as compared to healthy individuals and declines upon JAK2 inhibition [1]. Moreover, Lee et al were able to demonstrate that PD-L1 expression was significantly associated with overt myelofibrosis and JAK2 mutational status [9]. Moreover, in the previously mentioned study, there were 4 patients who were found to have a particularly high PD-L1 expression that also harbored the JAK2 mutation [9].
This supports further that PD-L1 may play a more important role than previously realized in MPNs and should perhaps be a future target in our current small armamentarium of viable drugs. To the best of our knowledge, there have only been 2 small phase II trials in which the investigators tested the utility of PD-L1 inhibition (pembrolizumab and nivolumab) in patients with PMF. Hobbs, et al conducted a phase 2, single arm study of pembrolizumab in patients with Dynamic International Prognostic Scoring System (DIPSS) intermediate-2 or greater, primary, or secondary, post essential thrombocythemia or post polycythemia vera MF who were ineligible for or previously treated with ruxolitinib [9]. This study had 10 patients, 5 with JAK2 mutation who were treated with pembrolizumab without objective clinical responses. However, an important takeaway from this data showed that flow cytometry, T-Cell receptor (TCR) sequence and proteomics demonstrated changes in the immune makeup of patients, suggesting improved T cell responses [10]. Although this study was terminated early as no objective clinical responses were seen, the latter changes mentioned suggest that perhaps PD-L1 inhibition is not enough to elicit a clinical response and combination therapy may be more effective.
Another study, in which Dalle et al investigated the efficacy and safety of single agent nivolumab in 8 adult patients with myelofibrosis, was also terminated early due to failure to meet predetermined efficacy endpoint (primary endpoint was objective response rate (ORR) defined as complete response (CR), partial response (PR) and clinical improvement (CI) after 8 doses) [11]. The median duration of enrolled patients on the study was 5.4 months with a median number of cycles of 3. Unfortunately, in this study, none of the patients responded to nivolumab therapy. These patients showed more advanced disease including intermediate 2 and high risk DIPSS score with 5 patients failing ruxolitinib and 7 with clonal evolution, i.e., progressive disease [11].
It is important to note that these two described studies had very small sample sizes, with most patients in the high risk DIPSS category, multiple previous lines of therapy, and complex mutational status or clonal evolution. However, they were able to characterize changes in the patient’s immune milieu after administration of PD-L1 blockade [10,11]. It’s noteworthy to the authors that both studies employed PD-L1 blocked only after patients had undergone multiple lines of therapy, in a relapsed or refractory setting, raising the question that perhaps this is not the right setting to use this line of therapy. In addition, it would be interesting to expand further on the hypothesis that perhaps PD-L1 blockade is not enough and combination therapy with ruxolitinib may be more effective in patients with PMF.