|Home | About | Journals | Submit | Contact Us | Français|
Non-AIDS Defining Cancer (NADC) has emerged as an important cause of morbidity and mortality in the HIV-infected population. Insight into the incidence, spectrum, risk factors, management, and outcome of these cancers has continued to emerge. The recent literature on this topic is reviewed.
Several recent studies have explored the shifting spectrum of NADC in both developed and underdeveloped regions of the world. Investigators have shown only a minor difference in the geographic spectrum of NADC. While several NADC have continued to occur at rates significantly higher than expected, a noticeable decline was observed in other cancers despite antiretroviral therapy. Factors other than HIV and immunosuppression proved to be important in the risk, treatment response, and outcome of these neoplasms. Studies dealing with the management of several NADC were published, including the role of highly active antiretroviral therapy (HAART).
An increased overall relative risk of developing NADC continues to be reported in the HIV-infected population worldwide. The development of NADC appears to be multifactorial. While control of HIV viremia has proven to be beneficial, the impact of HAART on NADC incidence rates and survival is not uniform. Further effort is needed to resolve the direct and indirect effects of HIV on NADC in order to guide effective prevention and treatment strategies of these malignancies.
An increased overall risk of developing Non-AIDS Defining Cancers (NADC) has been reported in the HIV-infected population [1-2]. The pending impact of this “sleeping giant” in people with HIV infection has been realized be several authors . Apart from a bona fide increased prevalence of NADC, there is greater screening, increased detection of more incidental cancers with imaging studies, augmented reporting, and improved survival in an ageing HIV-infected population. NADC include carcinomas, hematopoeitic malignancies and miscellanous neoplasms like melanoma, sarcomas other than Kaposi sarcoma, and germ cell tumors. The spectrum of NADC has expanded as the size of the HIV-positive population has increased [**4]. NADC are responsible for significant morbidity and mortality in the highly active antiretroviral therapy (HAART) era . In general, patients with NADC have been reported to have an overall poor outcome manifesting with adavanced cancer at presentation, rapid progression, frequent metastases, a high rate of relapse, and poor therapeutic response. Further insight into the incidence, spectrum, risk factors, management, and outcome of NADC has continued to grow. The recent literature pertaining to this topic is reviewed in this paper.
Several studies have documented an increased incidence of NADC [6-7]. Prior investigations showed that the incidence of NADC in the United States was at least twice that of the general population , with a rate of 980 new NADC diagnoses per 100,000 person-years in one study . A number of recent studies confirmed these findings. Investigators from the National Cancer Institute in the United States linked HIV/AIDS and cancer registries in three states where 57,350 HIV-infected persons were registered during 1991-2002 [*9]. Risk was shown to be elevated for Hodgkin lymphoma, lung cancer and hepatocellular carcinoma. In this study, NADC comprised 31% of cancers in 1991-1995 (pre-HAART) versus 58% in 1996-2002 (HAART era). Another study looking at data from 13 different cancer registries in the United States found the incidence of many NADC to be higher among HIV-infected persons than the general population from 1992 to 2003 [**4]. This particular study involved 54,780 HIV-infected persons, who contributed 157,819 person-years of follow-up (1992-2003). Specific NADC that were significantly higher in the HIV-infected population included anal, vaginal, liver, lung, oropharyngeal, colorectal, and renal cell carcinoma (RCC), as well as Hodgkin lymphoma, melanoma and leukemia. The incidence of prostate cancer in this study, however, was significantly lower among HIV-infected persons than the general population [**4]. A smaller study involving 2,566 patients in a single urban HIV clinic in the United States reported increased rates of several NADC . Moreover, researchers from Spain found similar high standard incidence ratios (SIRs) for Hodgkin lymphoma (SIR 31.1), liver cancer (SIR 29.4) and lung cancer (SIR 9.4) in men . Another recent study that examined linked population-based cancer and HIV/AIDS United States registry data for 268,950 men who developed AIDS in 1980 to 2003 demonstrated that seminoma risk was increased significantly with HIV/AIDS, whereas non-seminoma risk was not . Extratesticular germ cell tumor risk in this paper was also increased.
In developing countries where HAART is largely unavailable, the incidence of NADC appears to have likewise advanced with the escalating HIV epidemic. The incidence and spectrum of NADC has varied only slightly in these developing regions of the world . In particular, the incidence of HIV-related squamous cell carcinoma (SCC) of the conjunctiva is significantly increased in equatorial Africa . In a study exploring the spectrum of HIV related cancers at the largest tertiary referral cancer center in India, the proportional incidence ratio (observed number of site-specific cancer cases divided by the expected number) was shown to be increased for anal cancer, Hodgkin lymphoma, testicular cancer, colon cancer, as well as a few head and neck cancer sites . Also, a related case-control study from South Africa conducted between 1995-2004 reported significantly increased risks with HIV-associated Hodgkin lymphoma and anogenital cancer (excluding the cervix) .
Immunodeficiency (i.e. low CD4 count) has been shown by several investigators not to be the crucial facor for the development of NADC [17,18]. However, in a recent large meta-analysis conducted by researchers from Australia that involved seven studies involving people with HIV/AIDS (n=444,172) and five with transplant recipients, the increased risk of many NADC was similar to that of transplant recipients [**19]. This strongly suggests that immunodeficiency, rather than other risk factors for cancer, is perhaps responsible for the increased risk [**19]. Table 1 summarizes the published SIRs for several NADC from this meta-analysis. In a study evaluating the prevalence of neoplasia before and after HAART, researchers from Puerto Rico found a higher prevalence of NADC in the HAART era . These findings lead these authors to suggest that factors other than severe immunosuppression may be involved in the pathogenesis of NADC.
Indeed, several other key factors probably do play a role in NADC tumorigenesis, including the duration of HIV immunosuppression, longevity (age >40 years), lifestyle habits (e.g. sexual behavior, smoking, sun exposure), coinfection with oncogenic viruses, familial cancer history, and increased genomic instability (e.g. microsatellite alterations). A recent review article on the viral etiology of AIDS-associated malignancies, including those induced by HPV (penile, anal, oral, and conjunctival cancers), pointed out that the distribution of oncogenic HPV subtypes in Africa differs from that in the United States and Europe . Many HIV positive individuals have a history of cigarette smoking. Therefore, it is not surprsing that in the Swiss HIV Cohort Study, no lung cancers were observed in HIV-positive persons who were non-smokers . Interestingly, a recent study from the School of Public Health at Johns Hopkins University showed that their calculated risk associated with HIV-related lung cancer (hazard ratio, 3.6; 95% confidence interval, 1.6-7.9) was independent of smoking status . Furthermore, preexisting lung disease, particularly noninfectious diseases and asthma, in this cohort of injection drug users was found to be associated with an increased lung cancer risk.
Longer life expectancy brought about by the introduction of HAART has, unfortunately, been met with the occurrence of certain malignancies that develop over a long latency period. An analysis of data from adults (n =14,210) in the San Francisco AIDS surveillance registry (from 1990-2000), computer matched with the California Cancer Registry, showed that the impact of HAART on NADC (n = 482) incidence rates and survival was, in fact, not uniform [**23]. For example, the risk of anal cancer in this study increased after 1995 (relative hazard 2.9), whereas that for hepatocellular carcinoma was lower with HAART use (relative hazard = 0.32). Furthermore, lung cancer and Hodgkin lymphoma showed increased cancer survival time with HAART use, while anal cancer survival was slightly decreased.
In a recent review of the literature looking at five of the most common NADC (Hodgkin lymphoma, anal cancer, hepatocellular carcinoma, oral cancer, and lung cancer), in the pre- and post-HAART periods, the authors remained unclear whether earlier initiation (i.e. CD4+ cell count above 350 cells/mm3) of antiretroviral therapy was beneficial in preventing any of these cancers . Moreover, in the SMART study thta compared the rates of malignancies between patients on a CD4 T-cell-guided (interrupted) versus a continuous antiretroviral therapy strategy, the risk of AIDS-defining cancers increased with interrupted antiretrovirals while the rate for NADC was similar between the interrupted and continuous antiretroviral groups . Clealry, further studies on the relationship between HAART and NADC are required.
Hodgkin lymphoma, particularly those with unfavorable histopathology (mixed cellularity and lymphocyte depleted), is one of the most common NADC. A 3- to 18-fold increased risk of developing Hodgkin lymphoma has been reported with HIV infection [26-29]. Most cases in HIV-positive persons are EBV positive. While some studies have linked Hodgkin lymphoma to advancing HIV-related immunosuppression , others have shown a slightly higher risk in HIV-positive patients receiving HAART [18,31]. Prior to the introduction of HAART, patients with HIV-related Hodgkin lymphoma had a limited median survival of 1-2 years, despite chemotherapy . Early chemotherapy trials were plagued by severe toxicity and frequent deaths due to bacterial and/or opportunistic infection . The use of granulocyte colony-stimulating factor did little to improve this outcome .
Since then, several studies have reported the benefits of combining dose-intense chemoradiotherapy regimens and HAART. Treatment with bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone (BEACOPP) and HAART was shown to result in a complete response rate of 100%, event/disease free survival of 83% at 2 years, and overall survival of 83% at 2 years . Treatment with doxorubicin, vinblastine, mechlorethamine, etoposide, vincristine, bleomycin, prednisone, and involved-field radiation for initial bulky disease (Stanford V regimen) was shown to result in a complete response rate of 81%, event/disease free survival of 68% at 3 years, and overall survival of 51% at 3 years . In a recent study, investigators looked at an additional regimen involving doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) [*37]. In this latter study, the authors reported a complete response rate of 87%, event/disease free survival of 71% at 5 years, and overall survival of 76% at 5 years.
Several studies have shown a marked increase in the incidence of both anal intraepithelial neoplasia and invasive squamous cell carcinoma (SCC) in men and women who are HIV-infected [38-41]. One of the largest studies to date, however, did not find an increase in risk . Groups at risk reported by others include individuals who practice receptive anal intercourse, men who have sex with men (MSM), and those with an anal coinfection (e.g. HPV, syphilis, gonococcus, etc.). This finding was recently confirmed in a Multicenter AIDS Cohort Study from the United States that followed MSM (n=6,972) from 1984 until 2006 [**42]. In this cohort, investigators reported a significantly higher incidence rate of anal cancer in HIV-positive than HIV-negative men (incidence rate = 69 vs 14 per 100,000 person-years). Among the HIV-positive men, they also found that the incidence of anal cancer was higher in the HAART era than the pre-HAART era (incidence rate = 137 vs 30 per 100,000 person-years). In multivariate analysis restricted to the HAART era, anal cancer risk increased significantly with unprotected receptive anal sex partners in HIV-positive men. Moreover, the use of HAART in this study was found not to decrease anal cancer risk.
HIV-associated invasive anal carcinoma can be treated with standard combined chemotherapy and radiation [43-45]. Chemotherapy regimens include mitomycin (or cisplatin) and 5-fluorouracil (5-FU). The response to multimodality therapy in HIV-positive patients appears to be equivalent to non-HIV patients . Using a large database of United States veterans (n=1,184) diagnosed with anal SCC between 1998 and 2004, investigators reported an equivalent survival rate between HIV-positive and HIV-negative individuals in the HAART era when treated similarly [*47]. The 2-year observed survival rates were 77% and 75% among HIV-positive and HIV-negative individuals, respectively. Another recent study from Switzerland compared the outcome of 40 HIV-positive patients receiving HAART to 81 HIV-negative patients treated with radiotherapy or chemoradiotherapy for anal SCC [**48]. These researchers found that either form of therapy resulted in a complete response in 92% of HIV positive patients, compared to 96% in HIV negative cases. Moreover, there was very little difference in the five-year overall survival. However, five-year local disease control was 38% (95% CI, 5% to 71%) in HIV-positive and 87% (95% CI, 79% to 95%) in HIV-negative patients, compromising cancer-specific survival and sphincter preservation. Furthermore, HIV positive patients more frequenlty experienced acute skin and hematologic toxicity. Similar findings were recently reported in a study from St. Vincent's Cancer Care Center in New York , in which the outcome in 32 HIV positive patients treated with HAART and chemoradiotherapy for HIV-related anal SCC was comparable to that in patients without HIV. Overall survival was associated with high HIV viral load and low CD4 cell count. Significant toxicity (skin, diarrhea, hematologic) was also docuemnted with standard treatment regimens.
The incidence of lung cancer is increased several fold with HIV infection compared to age- and gender-matched populations [50-52]. This increasing incidence was recently reported in a single institution analysis at the University of Maryland in the United States . Thirty of the patients in this study had non-small cell lung cancer and four had small cell carcinoma. Of the 30 patients with non-small cell lung cancer, 27 had stage IIIb/IV disease. This is in keeping with findings from other studies that HIV-related lung cancer in the majority of cases is usually diagnosed with locally advanced or metastatic disease. The patients in this Maryland study received standard chemotherapy, and their overall survival was 5.2 months . Data on the efficacy and toxicity of chemotherapy and radiation therapy for HIV-related lung cancer is limited. Recently, however, researchers from Baltimore in the United States published their findings of a phase II trial of gemcitabine/carboplatin followed by paclitaxel that included only two patients with HIV infection and advanced non-small cell lung cancer . Although these patient numbers are small, this is the first trial to prospectively evaluate chemotherapy for patients with HIV disease. Unfortunately, the prognosis for these patients was extremely poor despite treatment.
The link between breast cancer and HIV remains controversial [55-56]. The preponderance of evidence suggests that breast cancer incidence is not increased in people with HIV infection. Of the 1,416 HIV-infected patients seen at Ramathibodi Hospital in Thailand over a 5-year period (1999-2003), investigators recently reported 42 patients with malignancies, giving a prevalence of 3% . In this study, breast cancer was the most common NADC (10%). In the last year there were a few cases reported of breast cancer in HIV infected patients [58,59]. HIV-infected women seem to be at increased risk of presenting with advanced breast cancer, perhaps related to inadequate screening mammography because of socioeconomic disadvantages . There have been very few series dealing with HIV-associated breast cancer. In one small series of HIV-infected patients who presented with early breast cancer (stage I and II), the 5-year survival rate (80%) was noted to be similar to that of an HIV-indeterminate (control) group of patients . Further studies are needed.
There is also an increased risk of developing hepatocellular carcinoma (HCC) with HIV infection [61,62]. A recent retrospective analysis conducted by the North American Liver Cancer in HIV Study Group from 1992 to 2005, involving 63 HIV-infected HCC patients, found their HIV positive patients to be younger than controls (52 vs. 64 years of age), more commonly had chronic hepatitis B or C (97% vs. 73%), were more frequently symptomatic (51% vs. 38%), and had a higher median alfa-fetoprotein level (227 vs. 51 ng/ml), but exhibited a similar mean Child-Turcotte-Pugh score and HCC staging score [**63]. Viral hepatitis coinfection has previously been shown to result in accelerated progression to HCC, with only a short interval between viral exposure and carcinoma development. Indeed, Brau and colleagues found that HCC developed faster in HIV/HCV-coinfected than in HCV-monoinfected patients (mean, 26 vs. 34 years after HCV infection) [**63]. The median survival in this study was similar for HIV positive and negative individuals with HCC (6.9 vs. 7.5 months). Independent factors that predicted survival in these study patients included symptomatic presentation, HCC therapy, AST/ALT ≥ 2.0, AFP ≥ 400 ng/mL, and platelets ≥ 100,000/mm3, but not HIV-serostatus [**63].
In HIV infected patients, the conjunctiva may develop intraepithelial dysplasia, carcinoma in situ, and/or invasive SCC. Risk factors include age >50 years, high solar ultraviolet radiation exposure, geography (sub-Saharan Africa), and possibly HPV infection . In sub-Saharan Africa, several publications have reported a dramatic increased incidence of conjunctival neoplasia in concert with the HIV epidemic [65,66]. A recent epidemiologic study published by the Division of Cancer Epidemiology and Genetics at the National Cancer Institute in the United States has reported a similar increased incidence of conjunctival SCC in North America [**67]. In this study, overall SIRs were elevated for SCC of the conjunctiva (SIR, 12.2, 95% CI 6.8-20.2), primary ocular lymphoma (21.7, 95% CI 15.1-30.2) and ocular KS (109, 95% CI 63.5-175). Risk for conjunctival SCC was shown to be elevated regardless of HIV acquisition category, CD4 lymphocyte count and time relative to AIDS-onset. Relative proportions of conjunctival SCC risk were noted to be highest with age ≥ 50 years, Hispanic ethnicity and residence in regions with high-solar ultraviolet radiation [**67].
Several cases of prostate carcinoma have previously been reported in HIV-positive men [68-70]. Studies reporting on the incidence of prostate cancer in the HIV positive population have been conflicting. Most large scale studies have found rates that are significantly lower than in the general population [6,19]. Risk factors for the development of HIV-related prostate cancer include increased age, black race, familial history of prostate cancer, androgen use, and possibly prostatitis. Although rare cases of aggressive HIV-related prostate cancer have been reported , the findings from a recent multicenter series found that in men with prostate cancer receiving HAART, their HIV status did not appear to influence their PSA levels, clinical presentation, tumor grade, stage, and management [**72]. Nor was their outcome from treated cancer affected by HIV status. The authors of this series recommended that patients with prostate cancer and well-controlled HIV viremia therefore be managed similarly to their HIV-negative counterparts [**72].
Currently, only a few cases of bladder cancer have been reported in association with HIV infection [73-75]. Nevertheless, a meta-analysis of several studies did not find an increased incidence of bladder cancer in the HIV population [**19]. Neither did recent data from the HIV/AIDS Cancer Match Study in the United States show any increased incidence . A link between HIV infection and renal cell carcinoma (RCC) has also been reported in a limited number of publications. In the HIV/AIDS Cancer Match Study, investigators identified 130 cases of RCC . However, RCC risk was not significantly elevated in this study. The authors of this paper also showed that RCC risk was unaffected by race, despite higher rates of HIV-associated nephropathy in individuals of African ancestry.
HIV infected patients can develp paraproteinemia, plasmacytomas and multiple myeloma . Recently, investigators looked at the protein electrophoresis and quantitative immunoglobulin levels in samples from 320 consecutive HIV-1-infected patients [*78]. They found that 8.1% of HIV positive individuals had subtle oligoclonal banding and that 4.4% had a low-concentration (<5% of total protein) monoclonal band. HIV-related paraproteinemia appears to be associated with younger HIV-1-infected patients, HIV viral load, a more robust immune system (i.e. higher CD4 cell count), and hepatitis B and/or C virus coinfection [*78]. These bands may be transient or persistent, and in a subset of patients may be associated with the subsequent development of myeloma, plasmacytoma, or lymphoma .
HIV does not appear to be a risk factor for molar pregnancy . Nevertheless, there have been a few case reports showing that choriocarcinoma has a dismal outcome in patients with AIDS [81-82]. Nevertheless, a recent case of a 34-year-old woman with AIDS and advanced metastatic choriocarcinoma was reported not to be fatal following appropriate chemotherapy .
Within the last year, information about the incidence and spectrum of NADC has continued to evolve, in both developed and underdeveloped regions of the world. While several NADC continue to occur at rates significantly higher than expected, no appreciable increase and even a decline was observed in other cancers. The development of NADC is probably multifactorial. Factors other than HIV and immunosuppression appear to be important in the risk, management, and outcome of these neoplasms. While control of HIV viremia is beneficial, the impact of HAART on NADC remains unclear. Recent studies have shown us that patients with NADC and well-controlled HIV viremia can be managed similarly to their HIV-negative counterparts. Further research in this emerging field is required, especially since therapy in this setting is often complicated by comorbid disease, poor performance status, additive cytotoxicity, potential drug-drug interaction, chemotherapy enhanced immunosuppression, a predilection for severe radiation adverse effects, and the fact that cancer patients with AIDS make poor surgical candidates. The need for assertive prevention strategies is also paramount. Therefore, it is encouraging to see recent publications on the subject of screening for NADC and their precursor lesions .