People infected with human immunodeficiency virus (HIV) are at a higher risk of developing cancers such as non-Hodgkin’s lymphoma (NHL), Kaposi’s sarcoma, and cervical cancer due to immunosuppression compared to the general population, with a 77-fold increased risk of NHL (1, 2). Although antiretroviral therapy (ART) has reduced mortality among people living with HIV, cancer remains a significant cause of death. The attributable mortality rate of HIV-associated cancers was 386.9 per 100,000 population per year, with NHL causing 3.5% of deaths (3). Among patients with HIV-associated NHL, the three major subtypes, namely, Burkitt lymphoma, diffuse large B-cell lymphoma, and central nervous system lymphoma, exhibit a particularly high risk (4). The incidence of cutaneous T cell lymphoma (CTCL) is 10.2 cases per million people (5), which accounts for 3.4% of all NHL (6). There are very few studies on CTCL in HIV-infected patients.

CTCL is a clinically and biologically heterogeneous group of extranodal NHL characterized by skin involvement, with mycosis fungoides (MF) and sézary syndrome (SS) being the most common tumor subtypes. In the early stage of the onset, patients usually present with discrete skin lesions resembling eczema or extensive erythema. Patients with advanced disease may have fungating tumors or leukemia, eventually involving the lymph nodes and viscera (7). Since the early manifestations of MF/SS are similar to a benign, inflammatory or autoimmune skin disease, it is not easy to detect characteristic histologic changes even after multiple pathologic biopsies. Therefore, the diagnosis of MF/SS requires a comprehensive analysis combining clinical manifestations, pathological assessment and molecular studies of patients, which poses a great challenge to clinical diagnosis. So far, the epidemiological characteristics, clinical features, and long-term prognosis studies of CTCL in HIV-infected patients have rarely been reported. Previously, Wang et al. (8) showed that HIV-infected patients with CTCL had higher survival rates and lower overall risk of death compared with HIV-uninfected patients. However, the epidemiological characteristics of patients with HIV-associated CTCL and the prognostic factors associated with survival remain unclear.

Therefore, this study retrospectively analyzed the epidemiological features, demographic characteristics, and prognostic factors of HIV-associated CTCL based on patient data extracted from the Surveillance, Epidemiology and End Results (SEER) database, and established prognostic nomograms to guide clinicians to more accurately assess the patient’s overall survival (OS) and disease-specific survival (DSS) rates.

CTCL belongs to a relatively rare group of NHL, a heterogeneous group of lymphoproliferative diseases characterized by clonal proliferation of mature post-thymic T cells infiltrating the skin (10). At present, the detailed pathogenesis of CTCL remains largely unknown, and known risk factors include age, race, gender, and environmental, infectious, iatrogenic, and other exposures associated with the occurrence of CTCL. Among these, infectious factors including human T lymphotropic virus type-1 (HTLV-1), HTLV-2, Epstein-Barr virus (EBV), cytomegalovirus (CMV), human herpesvirus, merkel cell polyoma virus (MCV), and HIV are also considered to play an important role in pathogenesis (11). The global incidence of CTCL has increased 2 to 3 times in the last two decades, with 5 to 11 cases per 1,000,000 people per year (6, 12). The incidence of CTCL varies according to geographical location (11).

However, CTCL is uncommonly associated with HIV infection, and the incidence of HIV-associated CTCL is currently unclear (13). We calculated the incidence of HIV-associated CTCL by extracting patient information from the SEER database. To our knowledge, this is the largest study of HIV-associated CTCL to date.

Our study showed that the age-adjusted average incidence rate of HIV-associated CTCL per 100,000 persons per year for the period 2004-2017 was 0.071. The incidence of HIV-associated CTCL remained stable from 2004 to 2007, with an average annual incidence of 0.01/100,000. However, since 2007, the incidence rate of HIV-associated CTCL had increased sharply to 0.09 per 100,000 person-years by 2011. The incidence rate of HIV-associated CTCL per 100,000 person-years increased from 0.09 in 2011 to 0.17 in 2017.

CTCL can occur at any age, but more often the median age at diagnosis is 54 years (range, 10 - 89 years) with a male-to-female ratio of 1.6:1 (14), and our findings are similar in the HIV-associated CTCL population, with a median age at diagnosis of 59 years and a male-to-female ratio of 1.5:1. Our study showed that OS and DSS decreased with increasing age in HIV-associated CTCL patients. Similarly, another study showed that age >60 years predicted lower survival in CTCL patients (15). There is now growing evidence that the incidence of cancer increases with age, which may be related to various age-related changes such as immune system disorders and DNA damage (16, 17). In addition, the incidence of complications such as cardiovascular, renal, neurocognitive, and osteoporosis increases with age in HIV-infected patients (18). More comorbidities in patients may also have a direct adverse effect on the survival time of patients, which is the cause of the higher mortality rate in elderly patients (19).

In this study, female patients had longer OS than male patients, which may be related to the direct impact of estrogen on the proliferation of lymphoid cells or its effect on anti-tumor immune response (20). Treatment of NHL is dominated by combination chemotherapy regimens, and in female patients, the combination of rituximab can prolong overall survival. Better outcomes in terms of complete remission in certain lymphoid malignancies were associated with a significantly lower clearance of rituximab in female patients than in male patients (21). Marital status is also considered an independent prognostic factor for survival in various cancers, and it has been shown that marital status is a good predictor of prognosis in patients with HL, with widowed patients likely to have worse survival outcomes than other groups (22). Another study, consistent with our findings, showed that NHL patients who lacked partner or spousal support or had a high disease burden may severely impair health-related quality of life (23).

CTCL patients are at a significantly increased risk of developing a second primary malignancy, particularly lymphoma, and this risk increases over time (24). The vast majority of CTCL patients consists of MF subtypes (25, 26). Studies have shown that CTCL is more prevalent in Asian populations, with the MF subtype accounting for 62% of CTCL compared to 3% for the SS subtype (27). And in our study, MF subtypes accounted for 99.5% of HIV-associated CTCL patients. Patients with MF may have underlying immune dysregulation, including T-cell immunodeficiency resulting in reduced immune surveillance in the MF environment, which may be a factor in increasing the risk of malignancy (28), and the same immunodeficiency occurs in HIV-infected patients. Patients with MF have a significantly increased risk of being diagnosed with second primary malignancies such as NHL, HL, melanoma, lung cancer, female breast cancer, prostate cancer, colon cancer, and kidney cancer. OS was lower in MF patients with a second malignancy compared with patients without second malignancies (29). As shown in Figures 3 , 4 , similar results also appeared in the HIV-associated CTCL population.

In 2007, the International Society for Cutaneous Lymphoma (ISCL) and the European Organization for Research and Treatment of Cancer (EORTC) revised the classification and staging system of CTCL into a TNMB staging system specifically for MF/SS (30), and recommended its use in defining the burden of disease to guide physicians in their treatment strategies. A retrospective study by Agar et al. (14) validated the revised ISCL/EORTC MF/SS staging recommendations and confirmed that the revised T, N, M and B classifications were significantly associated with OS and DSS. As shown in Table 3 , the tumor staging system in this study included Summary stage, T N M stage, Stage, etc., and to our knowledge, this would be the first study to assess the prognosis of patients with HIV-associated CTCL using a tumor staging system. Numerous studies in the past have shown that the prognosis of patients with HIV-associated NHL in the post-antiretroviral therapy era is significantly better than before (31–34). The prognosis for HIV-positive NHL patients is now very close to or similar to that of HIV-negative NHL patients (35). Prognostic studies in patients with HIV-associated CTCL appear to have better results. Compared with the results of Agar et al. (14), the 5-year OS and DSS of HIV-associated CTCL patients were higher, which is consistent with the results of Wang et al. (8) showing that HIV infection status in CTCL patients is an independent protective factor.

Since the introduction of highly active antiretroviral therapy (HAART) in 1996, the life expectancy of HIV-infected patients has increased significantly, thereby increasing the likelihood of cancer in these patients (36). Studies have shown that despite HAART is used in HIV/AIDS patients in nearly 10 years, the survival rate of patients with HIV-related lymphoma is still poor (37). In this study, chemotherapy and radiotherapy were related with OS and DSS in patients with HIV-associated CTCL, but significantly decreased OS and DSS in patients, which was different from previously reported results. The prognosis of patients with HIV infection-associated tumors after chemotherapy and/or radiotherapy is related to their immune status, and lower CD4 counts in patients after treatment are associated with an increased risk of death (38). Another study showed that HIV patients with a CD4 count <200 cells/μL before treatment had an increased likelihood of toxicity after chemotherapy and radiation therapy, which may affect patient survival (39). Due to the lack of patient CD4 count information in this study, and the patient’s specific chemotherapy and radiotherapy regimens cannot be obtained from the SEER database, the results of this study may be biased. Mehta-Shah et al. recommended that patients with early CTCL disease and skin involvement use skin-directed therapies (such as local therapy, phototherapy, radiation therapy, total skin electron beam therapy) without major cumulative toxicity, while systemic therapy (brentuximab vedotin, romidepsin, bexarotene, pralatrexate, vorinostat, methotrexate, mogamulizumab, alemtuzumab, and pembrolizumab) be required for patients with advanced disease. Systemic treatments can be combined with skin directed treatments to provide maximum cumulative efficacy without cumulative toxicity (40). Recent studies have revealed that ectopic expression of paternally expressed gene 10 (PEG10) in large cell transformation (LCT) malignant T cells and suggested that PEG10 inhibition may serve as a promising therapeutic approach for patients with advanced CTCL (41). Nowadays, for patients with CTCL, allogeneic hematopoietic stem cell transplantation has been the only curative treatment (42). However, there is currently no large-sample randomized controlled trial to verify the specific treatment options for patients with HIV-associated CTCL, and further research on related treatment strategies is needed.

Nomogram converts scores for several independent variables into visual graphs, making the results of predictive models more readable and facilitating the assessment of individualized patient risk (43). The nomogram is increasingly being used for prognostic assessment of cancer patients (44). In this study, based on the results of univariate and multivariate COX regression analysis, we finally included “Age”, “Radiation”, “Chemotherapy”, “Summary stage”, “Sequence number” indicators to establish the nomogram of OS and DSS of patients. The C-index of the nomograms used to predict the OS and DSS of patients in the training cohort was 0.78 and 0.848, respectively, indicating that the model has good discrimination. The calibration plot also showed the consistency between the predicted and actual results of OS and DSS in two cohorts. All of these results suggested that the nomograms we constructed had good clinical predictive value.

Nevertheless, there exist some limitations in this study. Firstly, this study was conducted by retrospectively analyzing the information in the SEER database, thus making it impossible to guarantee the completeness and homogeneity of the data and inevitably influencing the choice of patients. Secondly, the SEER database only provides general information about the patients and the relevant treatment modalities, and does not provide information about the patients’ specific treatments such as chemotherapy and radiotherapy, which may lead to a certain degree of bias in the study results. Finally, although our study showed that the constructed model had positive predictive value, external validation in prospective clinical trials is still needed. Despite these limitations, we have determined the epidemiological characteristics, independent prognostic factors, and information on OS and DSS of patients with HIV-associated CTCL.

Our study shows that HIV-associated CTCL is very rare, and the incidence has been on the rise in the last decade or so. We determined the demographic, clinical characteristics and survival-related prognostic factors of patients with HIV-associated CTCL, and constructed and validated nomograms of OS, DSS of patients at different time points. The nomograms can help clinicians to better identify the individual risks of patients and accurately assess the prognosis of patients.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authors.

ZY, YS, and QH conceived the study idea, and performed interpretation, manuscript writing, and final approval. DG and FH performed the data analysis and collection, as well as language polishing. All authors reviewed and approved the final version of the manuscript.

This work was supported by Hubei Province Health and Family Planning Scientific Research Project (WJ2021M261 to YS) and the Natural Science Foundation of Hubei Province (2019CFB567 to YS).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.