Despite increased risk for lung cancer development in human immunodeficiency virus (HIV)-infected individuals,[1–3] this patient population has historically been excluded from lung cancer therapeutic clinical trials. One analysis performed in the preimmunotherapy era reviewed 462 lung cancer therapeutic clinical trial protocols and found that 26% explicitly excluded HIV-infected individuals while <1% of protocols explicitly included persons with HIV infection. Over the last several years, immune checkpoint inhibitors (ICIs) blocking the programmed death-1/programmed death ligand-1 (PD-1/PD-L1) axis have become an essential component of the treatment paradigm for both non-small-cell lung cancer (NSCLC) and small-cell lung cancer based on data from large phase III clinical trials;[5–13] however, these studies did not allow inclusion of HIV-infected individuals, due to safety concerns and the hypothesis that severe HIV/AIDS with low CD4+ T-cell counts may compromise the efficacy of these agents. Many uncertainties still exist regarding the use of ICIs for lung cancer in the HIV-infected patient population.
Among the biomarkers that have been evaluated to help determine which lung cancer patients will derive the most therapeutic benefit from anti-PD-1/anti-PD-L1 immunotherapy, PD-L1 protein expression by immunohistochemistry (IHC) has been the most extensively studied. There are limited data regarding PD-L1 expression in individuals diagnosed with concomitant lung cancer and HIV. One French study evaluated tumor PD-L1 expression using two antibodies among 34 HIV-infected and 54 HIV-undetermined patients with NSCLC; no difference in PD-L1 expression frequency (>5% considered positive) was found between the HIV-infected and HIV-uninfected groups (18.7% vs. 9.3% using E1 L3N clone; 10.0% vs. 5.6% using 5H1 clone). A case–control study evaluated PD-L1, PD-1, and B7-H3 expression among NSCLC patients with (n = 13) and without (n = 13) HIV infection (SP142 clone used for PD-L1 testing, staining >5% considered positive). PD-L1 expression on tumor cells was positive in 23% of cases and 46% of controls, while PD-L1 expression on tumor-infiltrating immune (TII) cells was positive in 31% of cases and 69% of controls. The study demonstrated no significant difference in PD-L1 percent expression on tumor cells (median 0% vs. 0%, P = 0.5) or PD-L1 percent expression on TII cells (median 0% vs. 10%, P = 0.2). A Japanese study evaluated the expression of PD-1/PD-L1 and immune cell infiltration within tumors from 15 HIV-infected and 29 HIV-uninfected NSCLC patients (E1 L3N clone used for PD-L1 testing). Tumor PD-L1 expression (IHC 2+ and 3+) was positive in 33.3% of HIV-infected and 27.6% of HIV-uninfected patients. In the HIV-infected cohort, high tumor tissue PD-L1 expression was associated with worse overall survival compared to those with low PD-L1 expression (p = 0.0003). Using propensity-scored matching analysis which accounted for age, performance status, epidermal growth factor receptor mutation sites, cigarette smoking, recurrence, and advanced stage, high PD-L1 expression remained associated with shorter overall survival in the HIV-infected cohort (p = 0.001). Collectively evaluating the results of these limited studies, PD-L1 expression does not appear to significantly differ between HIV-infected and HIV-uninfected individuals, and the data appear to support inclusion of HIV-infected lung cancer patients in immunotherapy clinical trials. Other potential biomarkers of interest, including tumor mutational burden, have not been evaluated in this patient population.
In chronic viral infections such as HIV, persistent viral replication leads to progressive loss of T-cell functions of proliferation and cytokine secretion.[17–19] Known as “T-cell exhaustion,” this process results in ineffective immune response and inability to adequately clear the virus. PD-1 is upregulated on both HIV-specific CD4+ and CD8+ T-cells, and PD-1 expression correlates with HIV-specific cytotoxic T-cell dysfunction.[17–19] Studies of simian immunodeficiency virus (SIV) in rhesus macaques, which is thought to closely model HIV disease in humans, have demonstrated that blockade of PD-1 and PD-L1 interaction in SIV results in reversal of functional exhaustion of T-cells [Figure 1]. Some investigators hypothesize that PD-1 or PD-L1 inhibition could theoretically improve outcomes from both HIV and malignancy standpoint.
Published data evaluating the use of immunotherapy agents in HIV-infected lung cancer patients have mainly been limited to case reports or small case series [Table 1].[20–28] Collectively, HIV-infected NSCLC patients treated with ICIs targeting the PD-1/PD-L1 axis do not exhibit significant differences from HIV-uninfected patients in terms of both activity (overall response rate [ORR] 27.5% and disease control rate 55.2%) and safety (immune-related adverse events [irAEs] 35% and grade 3/4 irAEs 9.7%). No cases have been reported to date evaluating chemotherapy–immunotherapy combinations or dual-immune checkpoint blockade in the HIV-infected NSCLC population.
Data from a recent systematic review analyzing 73 HIV-infected patients with various advanced solid tumors (including 25 NSCLC patients) treated with ICIs either as single agent or in combination showed similar results, with no new safety signals noted in this population and relatively good efficacy (objective response rate of 30% for NSCLC). Of the 34 included patients with known paired pre- and post-treatment HIV viral load (VL), HIV VL remained suppressed in 93% of the cases with pretreatment undetectable HIV VL. CD4+ T-cell counts was noted to increase in 14 of the 25 included patients with known paired pretreatment and posttreatment CD4+ T-cell counts; the mean (standard deviation) change in CD4+ T-cell count was 12.3 (28.5)/μL. These results suggest that ICI treatment does not negatively impact HIV VL or CD4+ T-cell counts in HIV-infected cancer patients, including those with NSCLC. Similar results were reported in phase I Cancer Immunotherapy Trials Network 12 study evaluating pembrolizumab in HIV-infected patients with different advanced cancers, including one NSCLC patient. All the participants were on antiretroviral therapy and none met the U.S. Department of Health and Human Services criteria for uncontrolled HIV. Pembrolizumab was well tolerated with most irAEs graded as mild/moderate (73.3%); 20% of the irAEs were grade 3. No statistically significant differences were noted in CD4 count in all participants (median increase of 19 cells/μL; P = 0.18) or in those with stable disease for ≥24 weeks (median increase 152 cells/μL; P = 0.13). HIV remained suppressed in all participants. One treatment-related death was reported in a Kaposi sarcoma (KS) patient from a diffuse KS herpesvirus (KSHV)-associated polyclonal B-cell lymphoproliferation. The patient had a previous history of elevated peripheral blood mononuclear cell-associated KSHV and KSHV-associated inflammatory cytokine syndrome. Therefore, in these patients, treatment with ICIs should be evaluated with caution. ICIs are particularly active in KS patients (ORR 67% with 1 complete response in a recent retrospective study), and upfront treatment with pembrolizumab is under evaluation in the phase II KAPKEY study.
In conclusion, as the number of indications for immune checkpoint blockade grows, the number of HIV-infected patients potentially treated with these agents is destined to rise. Current evidence, coming mostly from case reports and small case series, suggests that single-agent PD-1/PD-L1 inhibitors can be used safely in HIV-infected NSCLC patients with similar efficacy results observed in the overall population. Several ongoing clinical trials are evaluating ICIs in HIV-infected patients with different solid tumors (NCT03094286, NCT02408861) or NSCLC only (CHIVA-2/NCT03304093). The results of these studies are eagerly awaited and will provide further evidence on safety and efficacy of immunotherapy in this patient population.
Financial support and sponsorship
The authors disclosed no funding related to this article.
Conflicts of interest
The authors disclosed no conflicts of interest related to this article.