Untangling the Effects of Chemotherapy and HIV on CD4 Counts— Implications for Immunotherapy in HIV and Cancer

Therapies for HIV infection and cancer have evolved exponentially during the past 20 plus years. The year 1996 marked the availability of 3-drug combination antiretroviral therapy (ART); in 1997, rituximab was approved as the first monoclonal antibody for the treatment of cancer. Despite advances in both fields during the ensuing decades, a question remains as to whether we are optimizing meaningful therapeutic advances for people living with HIV and cancer. Treatment of HIV-associated malignant neoplasms underwent a major paradigm shift with the introduction of ART. Before effective HIV therapy, many cancer treatment regimens in people living with HIV were dose reduced and largely palliative. However, availability of ART and associated immune reconstitution led to a series of studies that demonstrated that many people living with HIV were able to tolerate standard cancer therapies and, with the exception of those with very low CD4 counts (eg, <50–100 cells/μL), often had outcomes comparable to those of the general population. Despite improved cancer outcomes, many early ARTs had toxic effects that overlapped with cancer therapies. Furthermore, an important class of ART, protease inhibitors, has clinically significant CYP3A4-mediated drug-drug interactions that may increase the toxic effects of many common antineoplastic agents.

In this edition of JAMA Oncology, Calkins and colleagues¹ evaluated associations between cancer treatment and HIV viral load and CD4 counts in 196 people with cancer in the Johns Hopkins HIV Clinical Cohort spanning 1997 through 2016. The authors hypothesized that poor tolerance of concurrent chemotherapy and/or radiotherapy and ART could lead to increased HIV viremia during treatment. However, no detrimental effects of chemotherapy on HIV viral load were noted, and in fact, decreases in HIV viral load in the group of patients receiving chemotherapy and/or radiotherapy suggested higher rates of successful ART administration, perhaps through improved engagement with care. These data reflect the feasibility of treating HIV and cancer concurrently. Although pharmacokinetic considerations remain important for chemotherapy and targeted small molecules used to treat cancer in people living with HIV, at present, 33 antiretroviral medications are available, including many well-tolerated options that do not have significant interactions with common chemotherapeutic agents. This availability vastly simplifies cancer management in people living with HIV, including in low- and middle-income countries, where the addition of HIV integrase strand transfer inhibitors to first-line ART eliminates most significant drug-drug interactions. Controlled HIV viremia allows for CD4 immune reconstitution, decreased immune activation and senescence, improved control of coinfections, and improvements in survival in Kaposi sarcoma,² lymphoma,³ and other cancers. Treatment guidelines for a range of cancers in people living with HIV are based on the principle of concurrent ART.⁴

In the era of rapidly expanding options for immunotherapy for many cancers that are commonly associated with HIV, an improved understanding of the effect of cancer therapy on immune function in this population is required. To address this issue, Calkins et al¹ modeled the effects of chemotherapy and/or radiotherapy on longitudinal CD4 counts and evaluated the association of CD4 changes with mortality. Importantly, this study demonstrated increased mortality that was proportional to treatment-related declines in CD4 counts. The most common cancers in the cohort included lymphoma (33 [16.8%]), lung cancer (23 [11.7%]), anal cancer (14 [7.1%]), and breast cancer (12 [6.1%]), diseases usually treated with high-dose chemotherapy and/or radiotherapy. These treatments are known to have a detrimental effect on CD4 counts, particularly naive CD4 cells. Calkins et al¹ found that despite improved HIV control, chemotherapy and/or radiotherapy had a profound association with CD4 counts, which was greatest in those with relatively normal baseline counts, with a median decrease of 203 cells/μL in those with a baseline CD4 count of greater than 350 cells/μL but a decline of only 45 cells/μL for those with baseline CD4 count of 350 cells/μL or less. The increased mortality independently associated with treatment-related decreases in CD4 counts persisted after correcting for baseline CD4 count, HIV viral load, tumor type, and treatment era. In surviving patients, CD4 counts increased over time such that the relative immune reconstitution during ART appeared to overcome the immunosuppressive effects of chemotherapy and radiotherapy. For those with a higher baseline CD4 count, mean immune reconstitution to baseline required approximately 18 months.

This study illustrates several points relevant to the treatment of persons living with HIV and cancer. First, interpretation of CD4 counts in this population must account for the immunosuppressive effects of cancer and its therapy, because not all CD4 lymphocytopenia is due to HIV. As eligibility criteria for cancer clinical trials is evolving to be more inclusive of persons living with HIV, this study demonstrates that absolute lymphocyte depression may commonly be associated with prior cancer care and not only underlying HIV and calls into question optimal CD4 eligibility criteria for clinical trials in diseases in which prior therapy is known to cause CD4 lymphocytopenia in persons living with HIV and the general population alike.⁵ Second, increased mortality with treatment-related decreases in CD4 count suggest that introduction of immune-based therapies earlier in the course of cancer management may limit additional immunosuppression and be particularly beneficial for HIV-associated cancers.

This study has several limitations. First, the cancers and treatment modalities included were heterogeneous. Many of the findings are owing to radiotherapy and/or chemotherapy with known lymphocytotoxic effects and cannot be generalized to all cancer therapies. Second, longitudinal ART use was not evaluated; and with 72 patients (36.7%) without viral suppression but only 23 (11.7%) HIV treatment naive at baseline, recency of ART initiation and ongoing ART adherence may have accounted for unmeasured differences in CD4 trajectories. In addition, the type of ART was not specified, and decreases in CD4 counts may have been potentiated by drug-drug interactions if protease inhibitors were used. Last, cause of death is not described; if due to infectious complications, this would further emphasize a role for active monitoring of immune function and CD4-based infection-related prophylaxis in addition to ART.

The findings by Calkins et al¹ are important for the field of HIV-associated cancer. Clinical research is increasingly focused on appropriate use of checkpoint inhibitors and other immunotherapeutic agents in people with HIV and cancer, including several recent and ongoing studies of monoclonal antibodies targeting anti–PD-1/PD-L1 (programmed cell death 1/programmed cell death ligand 1) or anti-CTLA4 in this population. Untreated HIV infection includes a rapid period of immune dysregulation followed by a more variable gradual decline in CD4 counts over many years⁶ associated with perturbed T-cell repertoires, lasting abnormalities in mucosal immunity, and upregulation of immune checkpoint molecules. One historical question in the field of immuno-oncology is whether persons living with HIV have sufficient immune reserves to benefit from such immunotherapy as checkpoint blockade. In a recent study,⁷ safety and activity of anti–PD-1 therapy in patients with HIV and advanced cancer was demonstrated in participants with CD4 counts as low as 100 cells/μL and suppressed HIV while receiving ART, raising the likelihood that as with the general population, immunotherapy will play an important role in the treatment of many cancers in persons living with HIV. The study by Calkins et al¹ highlights the need to limit immunosuppressive therapies. Improved methods to evaluate immune function in people with HIV and cancer are also needed. For example, augmenting current CD4 counts with percentage of CD4, CD4:CD8 ratios, historic CD4 nadirs, and more detailed T-cell immunophenotyping may be useful for research and clinical practice. Future research should focus on further optimization of ART initiation and reducing ART interruption during cancer therapy and build on the success story of ART through development of additional highly effective immunotherapeutic approaches for a range of HIV-associated cancers in the United States and sub-Saharan Africa, which has the largest burden of HIV-associated cancers.