In this study, we used a lung cancer model to demonstrate PD-1/PD-L1 inhibition of the anti-tumor immune response caused by 125I RPI. On day 6 after 125I RPI, PD-L1 expression was upregulated, but PD-1 expression was not. On day 12, PD-L1 was further upregulated; PD-1 expression was also upregulated. After 125I RPI combined with anti-PD-1 therapy, the immunosuppressive state was reversed, the infiltration of CD8+ T cells in the TME increased nearly two-fold, and the proportion of Tregs reduced three-fold, which significantly inhibited growth of the primary and secondary tumors. In addition, we showed that the sequence of 125I RPI and anti-PD-1 therapy affects the efficacy of the combined therapy, and delaying 125I RPI reduces the efficacy of combination therapy.
PD-1 is a surface marker of exhausted T cells. Our results showed that the proportion of CD8+ PD-1+ TILs did not increase on day six after 125I RPI, but increased significantly on day 12. The expression of PD-L1 on tumor cells also increased on day 12 after 125I RPI. This indicates that with increasing time after particle implantation and radiation dose accumulation, T cell function is gradually impaired, limiting the immune activation effect of 125I RPI. Our results show that 125I RPI combined with anti-PD-1 therapy can significantly inhibit the growth of the primary tumor and can also induce the abscopal effect, which can inhibit the growth of secondary tumors, whereas 125I RPI treatment alone cannot produce the abscopal effect. We showed that combination therapy can reduce immune suppression and activate systemic anti-tumor immune responses.
The conclusion achieved were further verified by the flow cytometry results. The proportion of CD8+ TILs in the TME increased significantly following the combined therapy, and the ability of CD8+ TILs to secrete IFN-γ and TNF-α was also significantly enhanced. This indicated an increase in infiltrated T cells in the TME and, therefore, enhanced activity and killing ability. Schaue et al. showed that a single dose of 15 Gy increased the expression of Tregs in mouse melanoma, but two fractions of 7.5 Gy reduced the expression of Tregs(23). In our study, the proportion of Tregs did not increase or decrease after 125I brachytherapy characterized by hyperfractionation, but decreased in combination with anti-PD-1 therapy. It is worth noting that CD4+ TILs did not increase after the combined treatment. It is possibly because that Tregs accounted for ~ 21% of CD4+ T cells in the PBS group, compared with ~ 7% in the 125I RPI+α-PD-1 group. The decrease in Tregs resulted in no change in the total number of CD4+ T cells, which masked the increase of effector CD4+ T cells.
Dovedi et al. found that administration of α-PD-L1 mAb seven days after completion of RT did not produce an effective anti-tumor immune response(5). Their results showed that the synergistic effect of RT and immunotherapy has a time limit, and concurrent therapy is better than sequential therapy. In our study, the α-PD-1 mAb administered three days after 125I RPI did not affect the efficacy of the combination therapy, possibly because the 125I RPI and anti-PD-1 therapy have a longer synergistic time window. However, administration of 125I RPI after dosing with three days of α-PD-1 mAb did reduce the efficacy. Our results are consistent with the study of Ahmed et al. where patients with non-small cell lung cancer (NSCLC) who received RT before or during the administration of anti-PD-1 had significantly longer survival than patients who received RT after anti-PD-1 therapy(21). This may be because radiation can promote the release of antigens, and this is conducive to the early initiation of anti-tumor immunity(24). In addition, α-PD-1 mAb is only moderately effective and cannot significantly inhibit tumor growth. When 125I particles are implanted, the tumor burden is too high, and 125I RPI is a low-dose-rate RT. Dosage accumulation takes time, and this causes the tumor volume to be significantly inhibited within the experimental endpoint. But, in humans, malignant tumors usually grow more slowly than the tumors in transplanted syngeneic tumor models. It may not affect the survival of patients if α-PD-1 mAb were given a few days before 125I RPI, but increasing the interval time may affect it, which requires more clinical trials to verify.
Clinical studies have found that single anti-PD-1/PD-L1 treatment does not have a curative effect in all cancer patients, and the objective response rate of NSCLC is < 20%(25). Studies have shown that anti-PD-1/PD-L1 therapy is beneficial in supporting sufficient numbers of T cells in the TME(26). Clinical studies have proven that RT is an independent predictor of good prognosis for anti-PD-1/PD-L1 therapy in the treatment of NSCLC(27). Here, we provide data on hyperfractionated brachytherapy combined with anti-PD-1/PD-L1 therapy. In terms of synergistic effects with immunotherapy, 125I RPI has unique advantages over EBRT. First, it is highly conformable; during EBRT normal tissues also receive radiation, resulting in a decrease in radiosensitive lymphatic immune cells in these normal tissues, whereas normal tissues do not receive any radiation during brachytherapy. Also, the radiation radius of 125I is only 1.7 cm, and the radiation intensity decreases as the distance increases, protecting normal tissues from radiation damage(28). Second, it has an ultra-long half-life; 125I can provide ~ 180 days of irradiation, which can continuously promote tumor immunity. Anti-PD-1/PD-L1 treatment mostly uses disease progression or unacceptable toxicity as the withdrawal criteria. The time window for the synergistic effect between 125I RPI and anti-PD-1/PD-L1 therapy is longer, which may result in longer anti-tumor immunity.
This study has some limitations. The effective radiation time of 125I RPI is 180 days. An increase in the cumulative dose may enhance the curative effect. However, due to the limited survival time of tumor-bearing mice, it was not possible to monitor the synergistic effect of 125I RPI and anti-PD-1 therapy over a longer time. We did not conduct any toxicity experiments, but no obvious toxicity was observed during our experiments. Our study found that 125I RPI combined with anti-PD-1 therapy inhibited the growth of secondary tumors, but because distant metastases in humans usually have additional mutations and different tumor microenvironments, there may be different responses to the combination therapy. Also, our study has not been verified in other tumor models. We believe that 125I RPI combined with anti-PD1 has broad applications in solid tumors, and especially in prostate cancer. Prostate cancer is less sensitive to anti-PD-1/PD-L1 treatment(29, 30), but clinical trials have shown that activated T cells increase in the peripheral blood of prostate cancer patients gradually and continuously after 125I RPI(31). 125I RPI may improve the effect of prostate cancer immunotherapy.