HIV infection and malignancy: Epidemiology and pathogenesis

INTRODUCTION — Individuals living with HIV have an increased propensity to develop malignancy [1,2]. The occurrence of an extremely high number of cases of Kaposi sarcoma (KS) was noted early in the AIDS epidemic and many of them had an unusually aggressive clinical course. KS was therefore included as an AIDS-defining illness in early case definitions from the Centers for Disease Control and Prevention (CDC). Non-Hodgkin lymphoma (NHL) and invasive cervical carcinoma were subsequently added as AIDS-defining conditions.

The spectrum of neoplasia in patients living with HIV has changed in areas where the use of potent antiretroviral therapy (ART) is widespread. The incidence of KS and NHL has decreased markedly, but there has been a relative increase in tumor types that collectively are referred to as non-AIDS-defining cancers (NADCs) compared with the general population. NADCs now are a major factor contributing to mortality in people living with HIV.

This topic will review the epidemiology and pathogenesis of malignancy in people living with HIV. General management considerations for these malignancies, as well as epidemiologic features and management considerations for specific tumor types are discussed separately. (See "HIV infection and malignancy: Management considerations".)

EPIDEMIOLOGY — The era of potent antiretroviral therapy (ART) began in 1996 with the availability of the protease inhibitors. In most patients, ART causes both an immunologic response manifested by normalization CD4 lymphocyte counts and a virologic response with nearly complete suppression of HIV viral replication. Both immunologic and virologic responses are important in achieving at least partial immune restoration, thus decreasing the incidence of opportunistic infections, reducing the risk of developing non-Hodgkin lymphoma (NHL) or Kaposi sarcoma (KS), and prolonging survival [3-6]. (See "Selecting antiretroviral regimens for treatment-naïve persons with HIV-1: General approach".)

As an example of the effect of ART on cancer incidence, in a cohort study of over 42,000 HIV-positive veterans and 100,000 uninfected matched controls, cancer incidence was highest for unsuppressed HIV-positive individuals compared with uninfected controls (relative risk [RR] 2.4), lower for those with early suppression (RR 2.0), and lowest among those with long-term suppression (RR 1.52), with stronger trends for AIDS-defining versus non-AIDS-defining malignancies [7].

AIDS-defining malignancies — The overall impact of the introduction of ART is illustrated by the Swiss HIV Cohort Study, which analyzed the incidence of AIDS-defining cancers and non-AIDS-defining cancers (NADCs) in 9429 patients with 54,715 years of patient follow-up [8]. The incidence of cancers was broken down into three periods: 1985 to 1996 (pre-ART), 1997 to 2001 (early ART), and 2002 to 2006 (late ART).

In this study, the incidence of AIDS-defining cancers compared with the population without HIV fell progressively (standardized incidence ratios [SIRs] 136, 28, and 15, respectively). Furthermore, the AIDS-defining cancers comprised a progressively smaller percentage of the cancer burden in the HIV population. Among the patients with AIDS or HIV infection, AIDS-defining cancers accounted for 467 of 529 malignancies (88 percent) in the pre-ART period, 89 of 189 (47 percent) in the early ART period, and 46 of 140 (33 percent) in the late ART period. Similar trends have been reported in other large series [9-13] and are anticipated to continue [14].

The impact of immunosuppression on the incidence of AIDS-defining versus non-AIDS-defining malignancies since the introduction of potent ART is illustrated by the following examples:

●In the HIV/AIDS Cancer Match Study, with the HIV Optimization and Prevention Economics model, it was projected that the proportion of patients living with HIV in the United States aged ≥65 years or older would increase from 8.5 percent in 2010 to 21 percent in 2030 [14]. Age-specific rates were projected to decrease through 2030 for malignancies including KS, NHL, and cervical cancer, while rates of prostate cancer were projected to increase. The cancer burden was projected to shift, with prostate and lung cancer expected to emerge as the most common types by 2030.

●In the Swiss HIV Cohort study, KS had a SIR of 246 compared with the population without HIV and comprised 51 percent of all cancers and 58 percent of the AIDS-defining cancers in the pre-ART period (1985 to 1996) [8]. By the late-ART period (2002 to 2006), the incidence of KS had fallen dramatically, and was approximately 23 times higher than the general population. In that period, KS comprised 30 percent of AIDS-defining malignancies and 10 percent of the total cancer burden in the study population living with HIV.

Similarly, the incidence of NHL relative to the population without HIV fell progressive through these same periods (SIRs 103, 27, and 16, respectively) [8]. Thus, NHL comprised 36, 28, and 23 percent of the cancer burden in this cohort.

Non-AIDS-defining cancers — The incidence of NADCs is elevated compared with the population without HIV (table 1) [8,15-18].

For example, a prospective cohort study in the National Cancer Database (NCDB) of approximately 14,500 HIV patients and over six million adults without HIV found higher odds of incident, advanced-stage NADCs in those with HIV, particularly for melanoma and cancers of the oral cavity, liver, female breast, prostate, and thyroid [15]. This study was conducted in the post-ART era, with a decade of follow-up (2004 to 2014).

Some, but not all, evidence also suggests that the incidence of NADCs rose after the introduction of ART, although data are conflicting. This may be due to the fact that patients living with HIV are living longer on ART and, thus, are more likely to develop malignancies. The rising incidence may also be related to HIV-related immunosuppression, but there are no clear data that ART is oncogenic. As examples:

●In a prospective cohort study of 11,112 patients followed in a single European HIV registry over a 25-year period, there were no significant increases in the incidence of non-AIDS-defining malignancies prior to the introduction of potent ART (1983 to 1995), but there was a significant increase in such malignancies after the introduction of ART (SIRs 2.05 and 2.49 for the periods 1996 to 2001 and 2002 to 2007, respectively) [9].

●In contrast, in the Swiss HIV Cohort Study, the incidence of NADCs was elevated compared with the general population without HIV but did not change significantly during any time period relative to the introduction of ART (SIRs 2.3, 2.7, and 2.2, respectively) [8].

The relative incidence of different tumor types, compared with individuals without HIV, varies in different studies [16,17,19]. For example, in an observational study with 3.1 million person-years of follow-up, people living with HIV had lower rates of invasive breast (SIR 0.63, 95% CI 0.58-0.68), prostate (SIR 0.48, 95% CI 0.46-0.51), proximal colon (SIR 0.67, 95% CI 0.59-0.75), distal colon (SIR 0.51, 95% CI 0.43-0.59), and rectal cancers (SIR 0.69, 95% CI 0.61-0.77) compared with the general population [19]. Many of these tumor types also have an increased incidence in immunosuppressed individuals following solid organ transplantation, although there are substantial differences between these populations for some tumor types. The results of a 2007 meta-analysis are summarized in the table (table 1) [16]. (See "Malignancy after solid organ transplantation".)

The increased incidence of NADCs has contributed to a changing pattern in the causes of death in patients living with HIV. Before the introduction of potent ART, malignancies accounted for fewer than 10 percent of all deaths among patients living with HIV [20,21]. By contrast, in the era of potent ART, malignancies account for a greater percentage of deaths among patients living with HIV. As an example, death was due to malignancy in 28 percent of 64,000 patients living with HIV in a French study with follow-up through the year 2000 [22].

Furthermore, patients with HIV and cancer are at a higher risk for overall mortality than patients without HIV with the same cancers. In an observational study conducted in the United States, overall mortality was increased among those with HIV and concurrent diagnoses of melanoma (hazard ratio [HR] 1.65, 95% CI 1.27-2.14) and cancers of the thyroid (HR 2.93, 95% CI 2.08-4.13), oral cavity (HR 1.66, 95% CI 1.44-1.92), cervix (HR 1.85, 95% CI 1.51-2.27), and bladder (HR 1.66, 95% CI 1.45-1.9) [15].

Fortunately, data suggest that cancer-specific mortality in HIV-positive patients is declining over time. As an example, in one observational study, mortality attributed to cancer decreased between the time periods of 2001 to 2005 (484 per 100,000 person-years) and 2011 to 2015 (314 per 100,000 person-years) [23].

Changes in the incidence of specific malignancies are discussed separately. (See "HIV infection and malignancy: Management considerations".)

HIV infection diagnosed during childhood — People diagnosed with AIDS during childhood remain at elevated risk for malignancy into adulthood, despite the management of their HIV infection with ART [24].

In a study of 5850 individuals diagnosed with AIDS at ages 0 to 14 years, the risk of any cancer decreased with the introduction of ART, but remained elevated compared with the general population (SIR 40 versus 17 in the 1980 to 1995 versus 1996 to 2007 periods) [25]. This increased risk was due almost exclusively to KS and NHL (SIRs 1694 versus 1146 for KS and 338 versus 116 for NHL, respectively, during these same time periods). The only non-AIDS-defining malignancy for which there was a significant increase compared with the general population was leiomyosarcoma. (See "HIV infection and malignancy: Management considerations", section on 'Sarcoma'.)

Secondary cancers — Cancer survivors living with HIV are at elevated risk of a second (or subsequent) primary cancer (SPC), as are cancer survivors without HIV. Therefore, cancer surveillance and screening are critical to patients living with HIV and a previous cancer diagnosis. Further details on the risk of SPCs in cancer survivors without HIV are discussed separately. (See "Overview of cancer survivorship care for primary care and oncology providers", section on 'Risk of subsequent primary cancer'.)

For cancer survivors living with HIV, the rates of being diagnosed with an SPC has ranged from 9 to 13 percent [26-28]. As an example, in the French Dat’AIDS study, the most common SPCs in males were non-Hodgkin lymphoma (NHL; 23 percent), skin cancers (10 percent) and Kaposi sarcoma (8 percent). In females, the most common SPCs were breast cancer (16 percent), skin cancers (9 percent) and NHL (8 percent) [27].

PATHOGENESIS — Multiple factors may contribute to the increased incidence of malignancy in patients living with HIV. These include immunosuppression, direct effects of the HIV virus itself, coinfection with other oncogenic viruses, environmental factors, and possibly the use of antiretroviral drugs.

Immunosuppression — The development of neoplasia in patients living with HIV is similar to that observed in solid organ transplant recipients who receive chronic immunosuppressive agents, as well as in patients with profound cell-mediated immune deficiencies (table 1). (See "Malignancy after solid organ transplantation".)

Untreated HIV infection is characterized by progressive immunologic deterioration. This immunologic decline, as reflected by the CD4 cell count, closely correlated with the increased incidence Kaposi sarcoma (KS) and NHL prior to the introduction of potent antiretroviral therapy (ART) [29]. In contrast, there does not appear to be the same correlation between immune compromise and the incidence of other malignancies. (See 'Epidemiology' above.)

Among people living with HIV, certain cancers are diagnosed at younger ages, possibly reflecting accelerated cancer progression or exposure to risk factors. In one study using data from the Surveillance, Epidemiology, and End Results program, younger ages at diagnosis were observed for patients living with HIV relative to the general population for lung and anal cancer and for myeloma (difference in medians of four years) as well as cancers of the oral cavity/pharynx and kidney (difference in medians of two years) [30].

Role of the HIV virus — The pathogenesis of HIV-associated neoplasia is complex. HIV infection may have a direct effect on a variety of cellular processes that contribute to the development of cancer [13]. Postulated mechanisms include the activation of proto-oncogenes, alterations in cell cycle regulation, inhibition of tumor suppressor genes, or other genetic alterations that lead to oncogenesis.

Coinfection with oncogenic organisms — Patients living with HIV are at increased risk of coinfection with other viruses that are known to cause cancer [13]. Furthermore, the natural history of these viral infections may be accelerated in patients with HIV.

HHV-8 infection — The epidemiology of KS suggested a link between the development of disease and a transmissible agent. In 1994, a novel gamma herpes virus was identified in biopsies from KS [9]. This virus was subsequently named human herpes virus 8 (HHV-8) or Kaposi sarcoma-associated herpes virus (KSHV).

HHV-8 has also been associated with the multicentric form of Castleman disease, primary effusion lymphoma, and plasmablastic lymphoma. These entities are found primarily in patients with HIV infection [31]. (See "HHV-8/KSHV-associated multicentric Castleman disease" and "Primary effusion lymphoma" and "HIV-related lymphomas: Epidemiology, risk factors, and pathobiology", section on 'HHV-8'.).

The role of HHV-8 in tumorigenesis, its epidemiology, and its transmission are discussed separately. (See "Human herpesvirus-8 infection".)

HPV infection — The relationship of human papillomavirus (HPV) associated cervical dysplasia and neoplasia to HIV infection may be due to lifestyle risks for acquiring both viruses, as well as to attributes of and host responses to HPV. Patients living with HIV may be unable to clear oncogenic strains of HPV due to T cell deficiency [32]. There is a significant relationship between the risk of persistent HPV infection and the degree of immunosuppression. (See "HIV and women" and "Virology of human papillomavirus infections and the link to cancer" and "Preinvasive and invasive cervical neoplasia in patients with HIV infection".)

The relationship between HPV infection and cervical neoplasia was illustrated in one series of 309,365 American patients with HIV infection or AIDS, in whom the relative risk for in situ HPV associated neoplasia (ie, squamous intraepithelial lesions [SIL]) increased over time, suggesting that the gradual loss of control over HPV-infected keratinocytes was related to advancing immunosuppression [33]. However, the incidence of invasive squamous cell cancers did not increase with advancing immunosuppression in HIV, which suggests that late-stage cancer invasion was not influenced by immune status.

In addition to its relationship to cervical neoplasia, HPV infection is associated with cancers of the head and neck (especially the oropharynx and base of tongue), eye [34], and anogenital (anus, penis, vagina, vulva) regions, which have increased in incidence since the introduction of ART. However, both HPV-related and HPV-unrelated head and neck cancer have increased during the ART era [35].  

Further discussion of HPV-related head and neck cancers as well as anal squamous intraepithelial lesions is found elsewhere. (See "Epidemiology, staging, and clinical presentation of human papillomavirus associated head and neck cancer" and "Anal squamous intraepithelial lesions: Epidemiology, clinical presentation, diagnosis, screening, prevention, and treatment".)

EBV infection — Hodgkin lymphoma is not an AIDS-defining malignancy, but its incidence has markedly increased in people living with HIV since the introduction of ART. Epstein-Barr virus (EBV) infection has an established etiologic role in some cases of Hodgkin lymphoma, and EBV is more frequently implicated in Hodgkin lymphoma in individuals living with HIV than in people without HIV. (See "Pathogenesis of Hodgkin lymphoma", section on 'Epstein-Barr virus'.)

The clinical manifestations and potential treatment of Hodgkin lymphoma in this setting are discussed separately. (See "HIV infection and malignancy: Management considerations", section on 'Hodgkin lymphoma'.)

EBV may also contribute to the development of monoclonal gammopathy in some patients living with HIV [36,37]. (See "HIV infection and malignancy: Management considerations", section on 'Plasma cell disorders'.)

HIV is also associated with EBV-related smooth muscle tumors (SMTs) [38]. HIV-associated smooth muscle tumors are discussed in more detail elsewhere. (See "Clinical manifestations and treatment of Epstein-Barr virus infection", section on 'Malignancies and HIV infection'.)

HBV and HCV infection — Infections with hepatitis B virus (HBV) or hepatitis C virus (HCV) are a major cause of hepatocellular carcinoma. Coinfection with HIV and either HBV or HCV can result in more rapid progression of liver disease, resulting in cirrhosis and hepatocellular carcinoma. (See "Epidemiology and risk factors for hepatocellular carcinoma" and "Epidemiology, clinical manifestations, and diagnosis of hepatitis B in patients living with HIV".)

Merkel cell polyomavirus — Merkel cell polyomavirus has a causative role in Merkel cell carcinoma, which is much more frequent in people living with HIV. (See "Pathogenesis, clinical features, and diagnosis of Merkel cell (neuroendocrine) carcinoma", section on 'Merkel cell polyomavirus'.)

There are also some data suggesting an association between Merkel cell polyomavirus infection and non-small cell lung cancer, though evidence is mixed [39-42]. Research to further evaluate is ongoing.

Hymenolepis nana — A single case report describes a patient with untreated HIV infection, presenting with signs and symptoms of metastatic cancer, whose biopsies revealed nests of undifferentiated cells of nonhuman origin with overt features of malignancy [43]. These cells were later identified by immunohistochemistry and sequencing as Hymenolepis nana. Further investigation is required to determine whether Hymenolepis nana infection may be a more widespread etiology of malignancy in the immunocompromised host.

Other environmental oncogenic stimuli — Infection with HIV may sensitize cells to the oncogenic effects of environmental stimuli such as tobacco [13]. As an example, there is a significant increase in the risk of lung cancer in individuals living with HIV. Although there is a higher incidence of tobacco use in this population, the risk of lung cancer remains approximately three-fold higher compared with the general population after adjusting for smoking status [44].

Antiretroviral drugs and timing of cancer development — Whether antiretroviral drugs contribute to the development of non-AIDS-defining cancers (NADCs) is unclear, and additional research is required to understand this issue [13].

One possible explanation for an observed association is that because ART extends the life expectancy of patients with HIV, it makes them more likely to develop certain cancers at some point in their lives. A cohort study including nearly 90,000 patients with HIV found that while the cumulative incidences of anal, liver, and colorectal cancer by age 75 years (approximating the lifetime risk) have increased over time, this increase was not explained by an equivalent increase in risk of developing these cancers at any given point in time [45]. Rather, an overall decrease in mortality also observed in these patients suggests that because they are living longer, they are more likely to develop these types of cancers.

However, other data suggest that certain malignancies such as KS or lymphoma may occur soon after the initiation of ART, arguing that the treatment itself rather than the increased life expectancy it confers may predispose towards these cancers. The timing of the development of malignancy relative to the initiation of ART was studied in a cohort of 11,485 patients [46]. In this cohort, a total of 457 cancers developed within 10 years after ART was begun. There was a much higher incidence of KS during the first six months after ART initiation, and this was attributed to the more severe immunosuppression present when treatment was started and possibly to unmasking of subclinical KS as part of the immune reconstitution inflammatory syndrome (IRIS). Similarly, a higher incidence of lymphomas was seen immediately after initiation of ART, followed by a decreasing incidence thereafter. For other malignancies, there was a gradually increasing incidence of cancer over time.

SUMMARY

●HIV infection and malignancy – Individuals living with HIV have an increased incidence of a wide range of malignancies, including both AIDS-defining malignancies (Kaposi's sarcoma [KS], cervical cancer, and non-Hodgkin lymphoma [NHL] including systemic NHL, central nervous system [CNS] lymphoma, primary effusion lymphoma, and Castleman disease), as well as many non-AIDS-defining cancers (NADCs). (See 'Epidemiology' above.)

●Pathogenesis – Factors contributing to pathogenesis may include direct effects of HIV, immunosuppression, coinfection with other oncogenic viruses, other environmental oncogenic stimuli, and possibly antiretroviral therapy (ART) itself. (See 'Pathogenesis' above.)

●Impact of antiretroviral therapy – The widespread use of highly active antiretroviral therapy (potent ART) has prevented severe immunosuppression for prolonged periods in most individuals living with HIV in the United States and Europe. This has been associated with substantial decreases in the incidence of KS and NHL, and a relatively better prognosis in response to treatment than in the pre-potent ART era. (See 'Epidemiology' above.)

●Non-AIDS-defining cancers – The frequency of NADCs is increased compared with the population without HIV, and cancer deaths from these NADCs account for an increasing fraction of the deaths in individuals living with HIV. (See 'Epidemiology' above.)