PD-1 Inhibitor Combined With Radiotherapy and GM-CSF (PRaG) in Patients With Metastatic Solid Tumors: An Open-Label Phase II Study

Abstract

Patients with metastatic cancer refractory to standard systemic therapies have a poor prognosis and few therapeutic options. Radiotherapy can shape the tumor microenvironment (TME) by inducing immunogenic cell death and promoting tumor recognition by natural killer cells and T lymphocytes. Granulocyte macrophage-colony stimulating factor (GM-CSF) was known to promote dendric cell maturation and function, and might also induce the macrophage polarization with anti-tumor capabilities. A phase II trial (ChiCTR1900026175) was conducted to assess the clinical efficacy and safety of radiotherapy, PD-1 inhibitor and GM-CSF (PRaG regimen). This trial was registered at http://www.chictr.org.cn/index.aspx. A PRaG cycle consisted of 3 fractions of 5 or 8 Gy delivered for one metastatic lesion from day 1, followed by 200 μg subcutaneous injection of GM-CSF once daily for 2 weeks, and intravenous infusion of PD-1 inhibitor once within one week after completion of radiotherapy. The PRaG regimen was repeated every 21 days for at least two cycles. Once the PRaG therapy was completed, the patient continued PD-1 inhibitor monotherapy until confirmed disease progression or unacceptable toxicity. The primary endpoint was objective response rate (ORR). A total of 54 patients were enrolled with a median follow-up time of 16.4 months. The ORR was 16.7%, and the disease control rate was 46.3% in intent-to-treat patients. Median progression-free survival was 4.0 months (95% confidence interval [CI], 3.3 to 4.8), and median overall survival was 10.5 months (95% CI, 8.7 to 12.2). Grade 3 treatment-related adverse events occurred in five patients (10.0%) and grade 4 in one patient (2.0%). Therefore, the PRaG regimen was well tolerated with acceptable toxicity and may represent a promising salvage treatment for patients with chemotherapy-refractory solid tumors. It is likely that PRaG acts via heating upthe TME with radiotherapy and GM-CSF, which was further boosted by PD-1 inhibitors.

Keywords: PD-1 inhibitor; chemotherapy refractory; granulocyte macrophage-colony stimulating factor; radiotherapy; tumor microenvironment.

Figure 1

Treatment schedule of the study.

Figure 2

Consort diagram.

Figure 3

Waterfall plots of maximum percent change in nonirradiated RECIST target lesions.

Figure 4

Kaplan–Meier curves of progression-free survival and overall survival.

Figure 5

Lymphocyte subset percentage changes after treatment from baseline between the three groups (CR+PR, SD, PD). The red boxplot represents percentage changes after one cycle of treatment from baseline. The green boxplot represents percentage changes after two treatment cycles from baseline. The blue boxplot represents percentage changes after three cycles of treatment from baseline. The differences in the proportion of changes after the first cycle of treatment, after the second cycle of treatment, and after the third treatment cycle was compared separately between the three groups (CR+PR, SD, PD). The one-way ANOVA was used for the homogeneity of consistent variance, and the rank-sum test was used for the homogeneity of inconsistent variance. None of the other lymphocyte subset percentage changes showed statistical differences (p > 0.05). (A) CD3⁺T cells percentage changes from baseline. (B) CD3⁺CD4⁺T cells percentage changes from baseline. (C) CD3⁺CD8⁺T cells percentage changes from baseline. (D) CD16⁺CD56⁺T cells percentage changes from baseline.