Discussion
In this retrospective study, we have evaluated the efficacy and the safety of the PD-1 inhibitor on locally advanced and advanced NSCLC patients with chronic viral infection. Results show that the PD-1 inhibitor-based therapy, especially combined with chemotherapy and bevacizumab, had moderate efficacy on this population. Notably, the incidence of elevated hepatic transaminase was higher in them than those without chronic viral infection and other toxicity profile was acceptable.
In recent past, immunotherapy, especially PD-1/PD-L1 immune checkpoint inhibitors, has brought new hope for lung cancer patients by activating the autoimmune system to achieve the effect of killing tumors. Successful anti-PD-1 / PD-L1 immunotherapy requires an adequate number of specific T cells in the tumor microenvironment. Similar to the tumor microenvironment, the chronic viral infection is a strong immunosuppression environment, leading to specific T cells exhausting4, in theory, it may influence the effect of ICI which has reversed the role of T cell failure at the same time and undermine the balance between the host immune system and virus control that has caused the risk of liver damage. In China the incidence of hepatitis B is higher. At present, the hepatitis B surface antigen carrying rate among people aged 1-59 is 7.2%. There are about 93 million chronic HBV infected people in China, among which more than 20 million active hepatitis B patients require proper treatment. The efficacy and safety of immunotherapy for these patients need to be clarified, but existing clinical trials tend to exclude patients with co-viral infection5.Thus there is not any consensus yet and appropriate strategy of ICIs in this population needs in depth assessment.
In the existing published articles about safety and efficacy of ICIs in patients with chronic viral infection and advanced-stage cancer, the types of cancer involved are not comprehensive, with liver cancer and melanoma accounting for the majority and NSCLC accounting for less than five percent 6. In one retrospective study of PD-1 inhibitors for NSCLC with special issues involving 32 HBV-infected patients, four of nine patients experienced severe AST/ALT elevation (grade 3 or higher) were HBV patients. And three patients developed viral reactivations or flares, despite receiving anti-HBV therapy prior to the immunotherapy. In our study, the incidence of elevated ALT/AST and increased virus-load was lower than in that study. Differences in ECOG performance status, prior lines of therapy, and agents combined with PD-1 antibody of the patients may contribute to the differences between the two studies7. Given the unavailability of such information and the limited number of patients in that study, the results of the two studies were not comparable.
Among 78 patients included in the study, the objective response rate (ORR) was 24.36% (19/78), and the disease control rate (DCR) was 69.23% (54/78). Comparing the phase III clinical trial Keynote-042 in the Chinese subgroup study8, the exclusion criteria of which included patients with a history of infectious diseases, the overall objective response rate of the immunotherapy group was similar to the results of this study (24.36% vs. 32.80%) (P=0.250), indicating that the history of infectious diseases does not affect the short-term efficacy. In terms of long-term efficacy, the median progression-free survival (PFS) reached 6.49 months (95% CI :3.71-9.27). Among them, the first-line treatment shows better efficacy than the second-line and later treatment (7.67 months vs. 5.57 months, P = 0.129). Previous studies have proved that the median progression-free survival of advanced NSCLC who received pembrolizumab as the first-line immunotherapy was about 8 months9,10, which is similar to our study, and that of advanced NSCLC who received nivolumab as the second-line immunotherapy was about 3 months11,12. In our study, the median PFS of second- and later-line treatment group seems improved than the previous studies. This phenomenon may be because the patients’ constitutions are different. In our study, 11.5% of patients were clinically diagnosed with stage IIIB lung cancer, which may have improved the results.
In our group analysis, the short-term efficacy of patients in the first-line treatment group was significantly better than that in the second-and multi-line treatment groups (48.00% vs. 13.20%) (P<0.05). The reason may be that patients in the first-line treatment group have not received other treatments, and there are more specific T cells in the tumor microenvironment than in second-line and multi-line patients, which promote the immune system to recover to a greater extent. The ECOG score of group 0 and group 1 was significantly higher than that of group 2 (88.23% vs.74.00% vs.18.18%) (P<0.05), indicating that patients with better physical conditions are more likely to benefit. The ECOG score of group 2 has poor efficacy, but no deaths related to adverse reactions of immunotherapy occurred, and the safety was fair. The 10 patients in the combined radiotherapy group were all in stable condition, which was better than the non-radiotherapy group, and the PFS was also significantly longer than the patients without combined radiotherapy (14.07 vs.4.62, P=0.027). Because local radiotherapy has a synergistic effect with immunotherapy by enhancing the uptake of antigen by APC, promoting DC activation and migration, and tumor-associated antigen cross-presentation13. We divided patients into four groups in accordance with the different agents that were used to combine with the PD-1 inhibitor, drawing the conclusion that patients who received the PD-1 inhibitor in combination with the chemotherapy and the bevacizumab therapy obtained the longest PFS (PD-1 antibody alone vs. PD-1 antibody combined with the chemotherapy vs. PD-1 antibody combined with the bevacizumab vs. PD-1 antibody combined with the chemotherapy and the bevacizumab groups: 1.44 vs.5.67 vs.1.67 vs.14.13, P=0.002), which is consistent with previous research.
The incidence of adverse events of any group among 78 patients was 73.07% (57/78), including 7 cases of grade 4 adverse reactions, 3 cases of bone marrow suppression, 2 cases of pneumonia, 1 case of superior vena cava obstruction, and 1 case of ketoacidosis. 3 patients who died due to respiratory failure after receiving immunotherapy for 2 months were considered to be affected by rapid tumor progression. The incidence of leukopenia in any grade of adverse reactions was the highest (57.69%), followed by anemia (25.64%), hepatic transaminase elevating (24.36%) and fatigue (21.79%). There is no big difference between the adverse reaction spectrum and the incidence of adverse reactions in patients with infectious diseases and those of patients without infectious diseases, but it is worth observing that the proportion of hepatic transaminase elevating was increased. In the clinical phase III trials of Keynote001, the proportion of alanine aminotransferase (ALT) elevating and aspartate aminotransferase (AST) elevating were respectively 2.2% and 3.0% in the treatment of pembrolizumab.14 Compared with patients without a history of infectious diseases, the patients in this study displayed a higher incidence of hepatotoxicity. Hepatic transaminase increased in 26.7% (16/60) of hepatitis B patients, and remained unchanged in 63.3% (38/60) patients, as is shown in Table 4. Notably, 5% of hepatitis B patients showed a decrease in hepatic transaminase. Of the 60 patients with hepatitis B, viral load remained unchanged in 53, viral load increased in 2, and viral load decreased in 5. All five patients with reduced viral load received antiviral therapy. Of the 14 patients who did not receive antiviral therapy, 3 patients had grade 3 or 4 adverse events, but in all three patients, the adverse events were reversed with steroids and ICIs were not discontinued, so the side effects were considered acceptable. In summary, there was no significant increase in the incidence of these adverse events in patients with a history of infectious diseases. Considering the retrospective nature of our study and small sample size, data from phase IV studies with relaxed inclusion criteria or from further prospective series are needed to shed more light on the safety and efficacy of ICIs in this challenging population.
In this retrospective analysis, the efficacy of PD-1/PD-L1 immune checkpoint inhibitors on locally advanced and advanced NSCLC patients with a history of infectious diseases was acceptable, safe, and the clinical outcome was not affected by the history of infectious diseases. Such patients can benefit from immunotherapy. However, considering that the incidence of hepatic transaminase elevating has increased, we recommend close monitoring for such patients in consultation with a hepatologist and to treat those with active viral hepatitis with antiviral therapy prior to the immunotherapy.