HIV infection increases the risk of certain malignancies [1–4]. Three cancers (Kaposi sarcoma, non-Hodgkin lymphoma [NHL], and invasive cervical cancer) are considered AIDS-defining illnesses  and are characterized as AIDS-defining cancers (ADCs). With the advent of effective antiretroviral therapy (ART), the incidence of ADCs has declined [6,7], yet incidence rates remain many times higher than those in the non-HIV-infected population .
More recently, HIV-infected patients have also been observed to be at higher risk for non-AIDS-defining cancers (NADCs) [6,7,9,10]. NADCs include but are not limited to cancers of the lung, liver, oral cavity, anus, and Hodgkin lymphoma. These NADCs are increasingly important causes of morbidity and mortality in HIV-infected patients [11–16]. Many cohorts to date have had a small percentage of African–Americans and have not focused on describing causes of death. To understand the impact of cancer on survival, it is essential to distinguish between cancer, AIDS-related opportunistic infections, or other causes of death in these patients.
Despite representing a small proportion (13.6%) of the general US population, African–Americans accounted for 50.3% of new HIV infections during 2005–2008 . In Baltimore city, 85% of HIV/AIDS patients are African–American, and there is a high percentage of late presenters to care (22.1%) . Historically, inferior cancer treatment outcomes have been described for both African–Americans [19–22] and HIV-infected patients . The impact of cancer on urban, predominantly African–American, HIV-infected patients in the era of HAART is understudied, and it is unclear whether cancer screening protocols are adequate for HIV-infected patients from underserved minorities.
The purpose of this study was to characterize the numbers and types of cancers, patient characteristics, mortality, and causes of death among HIV-infected patients in an urban, predominantly African–American population.
The Institute of Human Virology (IHV) provides comprehensive care to approximately 5000 HIV-infected patients per year at the University of Maryland Medical Center's HIV clinic (UMMC), the IHV Jacques Initiative, Maryland General Hospital (MGH), and Baltimore Veterans Affairs Medical Center. The outpatient UMMC clinic population is largely African–American (88%; 82% at Veterans Affairs) and predominantly male (56%; 95% at Veterans Affairs). The incidence was calculated for the two largest clinic sites (UMMC and Veterans Affairs), which together comprised 70% of all cancers. The university is a tertiary/quaternary care referral center for patients diagnosed with cancer from around the State of Maryland and surrounding areas. Consequently, cancer incidence could not be calculated because some inpatient cases receive primary HIV care outside our system.
We reviewed all HIV-infected patients diagnosed with invasive cancer from 1 January 2000 to 30 June 2010 at the university and its clinics. Cases were identified by searching electronic clinic records for diagnoses of ‘cancer,’ ‘neoplasm,’ ‘malignancy,’ ‘tumor,’ and ‘lymphoma.’ The University Clinical Data Repository was also searched using ICD9 codes for HIV or AIDS and for any neoplasm. At the Veterans Affairs, the clinical case registry of HIV-infected patients was searched by ICD9 codes for any neoplasm. Institutional Review Boards at each site approved the protocol.
Cancer diagnoses were confirmed by pathology or clinician reports and were categorized as ADCs or NADCs. ADCs included Kaposi sarcoma, invasive cervical cancer, and NHLs. NADCs were all other cancers. Individuals diagnosed with incident cancer as the presenting manifestation of HIV infection were included (i.e. individuals tested for HIV as a result of a new cancer diagnosis); individuals diagnosed with cancer prior to being diagnosed with HIV were excluded (N = 4). Cancer treatments included chemotherapy, radiation, or surgery. When more than one cancer was diagnosed in the same patient (N = 23), only the first cancer was analyzed further.
Demographic and clinical data were abstracted from medical charts. The HIV diagnosis date was defined as either the first positive HIV screening test or clinical documentation of known HIV positivity; if neither was available in the record, the date of the first HIV-specific test (CD4 or RNA viral load) was used. HIV viral load suppression was defined as any HIV viral load less than or equal to 400 copies/ml (assays used over time had varying detection limits from 48 to 400 copies/ml).
Participant status as of 30 June 2010 was recorded as alive, dead, or unknown. Survival was calculated from the date of cancer diagnosis to the conclusion of the follow-up period (30 June 2010). Dates and causes of death were recorded if known. Causes of death were obtained from death certificates when patients died in the hospital; causes of death for patients with progressive, metastatic, or untreatable cancer who died after hospital discharge to hospice were attributed to the underlying cancer. Social security death index records were checked to ascertain status for patients lost to follow-up (LTFU).
The primary endpoint was mortality after cancer diagnosis. Categorical variables were analyzed using Student's t-test with a P value of less than 0.05 deemed significant. Survival analysis was conducted using Kaplan–Meier estimates and Cox proportional hazards models for all-cause mortality. Survival curves were compared using log-rank tests. All statistical analyses including data management were carried out using Stata 11.0 (Stata Corporation, College Station, Texas, USA).
A total of 470 cases of cancer were identified among 447 patients (Table 1). The median age at the time of first cancer diagnosis was 50 years (0.2%, <20 years; 4.3%, 20–29 years; 12.8%, 30–39 years; 31.3%, 40–49 years; 37.1%, 50–59 years; 14.3%, >60 years). The majority of patients was male (79%) and African–American (85%). The most common risk factor for HIV acquisition was injection drug use (IDU) (39.4%), although heterosexual (30%) and homosexual (17%) contact were also common. The median time from HIV diagnosis to cancer diagnosis was 7 years (range 0–25 years). Thirty-nine patients (8.9%) were diagnosed with HIV within 30 days of their cancer diagnosis; this happened more frequently among patients with ADCs [25 (19.2%)] than among those with NADCs [14 (4.6%)] (P < 0.001).
IDUs were more likely to have NADCs than ADCs (P = 0.004) whereas MSM were more likely to have ADCs (P = 0.014). Smoking (both current and former) was significantly more frequent in patients with NADCs than ADCs (P = 0.011 for current and 0.017 for former smokers). Hepatitis C infection was more common in NADCs (48.7%) than in ADCs (31.7%) (P < 0.001).
The median CD4 cell count nadir prior to cancer diagnosis was 172 cells/μl (Table 1). At the time of cancer diagnosis, the median CD4 cell count was 268.5 cells/μl, and the median HIV RNA load was 740.5 copies/ml. Patients with ADCs were much more likely to have a CD4 less than 200 cells/μl and a viral load less than 100 000 copies/ml than those with NADCs (P < 0.001 for both). The majority of patients (55.9%) was taking ART at the time of cancer diagnosis; ART was started after the cancer diagnosis in 26.9%. However, a substantial minority (17.2%) never received ART. Patients with NADCs were more likely to be taking ART at the time of cancer diagnosis (P < 0.001), whereas patients with ADC were more likely to be initiated on ART after cancer diagnosis (P < 0.001).
The distribution of cancers and related HIV characteristics are displayed in Table 2. NADCs (69%) were more common than ADCs (31%). The median CD4 for NADCs was higher than that of ADCs (343 vs. 103 cells/μl, respectively). The median viral load for NADCs was lower than that of ADCs (400 vs. 42 163 copies/ml, respectively).
The cumulative cancer incidence increased over time at both the Veterans Affairs and UMMC clinics (combined in Fig. 1); both trends were significant over the 10-year time period (P < 0.05). Comparison of combined incidence for these two clinics in the first 5 years of the study period (2000–2004) to the second half of the period (2005–2009) showed a 15% increase in cancers (P = 0.72). NADCs outnumbered ADCs each year.
The median cancer stage at the time of diagnosis in 313 patients with documented staging was stage 3; 67% of patients were stage 3 or 4 (no stage information for Kaposi sarcoma and missing for others).
Overall, 76% of patients received some form of cancer treatment, 22% received no cancer treatment, and 2% were not clearly documented. Cancer treatments included chemotherapy (50%), radiation therapy (37%), and surgery (25%) (categories were not mutually exclusive). There were no overall differences in the percentage of patients with ADCs vs. NADCs who received any cancer treatment. However, surgery and radiation were more frequently utilized in NADCs than ADCs (P < 0.001 for each), whereas chemotherapy was more commonly used in ADCs (P < 0.001).
Death was associated with progression of the underlying cancer in 68% of cases with a known cause of death (Table 3). Of 30 deaths attributed to HIV/AIDS, 29 were caused by the ADC. Only 5% of total deaths were attributed to infection. Of the patients who died, those with NADCs were more likely to die from their cancer than those with ADCs (P = 0.001).
Among the 447 patients diagnosed with cancer, there were 226 deaths during 1096 person years of follow-up time. The overall mortality rate was 206 per 1000 person years. The cumulative mortality rate within 30 days, at 1 year, and at 2 years was 6.5, 32.2, and 41.4%, respectively. The 30-day mortality rate was highest for pancreatic cancer (42.9%), cancer of unknown primary (37.5%), Hodgkin lymphoma (22.2%), and liver cancer (21.7%). The two most common cancers (NHL and lung cancer) had high 2-year cumulative mortality of 47.1 and 73.4%. The highest 2-year cumulative mortality was found in pancreatic cancer (100%), cancer of unknown primary (87.5%), and liver cancer (82.6%). Mortality rates for multiple myeloma and prostate, thyroid, and nonmelanoma skin cancer were the lowest throughout the study period.
Mortality was analyzed using both Kaplan–Meier estimates and Cox proportional hazards models. Mortality was not significantly different whether patients had an ADC compared with an NADC (Fig. 2a). Kaplan–Meier analysis of ADC mortality with and without Kaposi sarcoma compared with NADC showed no statistical difference from the original analysis depicted in Fig. 2a (data not shown). By Kaplan–Meier estimates, patients who never received ART were at higher risk for mortality than those who were either on ART at the time of cancer diagnosis or started on ART after cancer diagnosis (Fig. 2b). When these data were analyzed for ADCs and NADCs separately, patients not on ART remained at higher risk for mortality (P = 0.012 and P = 0.002, respectively) (Kaplan–Meier graphs not shown). Similarly, patients with a suppressed viral load at the time of cancer diagnosis had lower mortality than those with uncontrolled viral replication (Fig. 2c).
In univariate analysis (Table 4), alcohol use, current IDU, HIV viral load less than 100 000 copies/ml, and both the Veterans Affairs and primary HIV care site outside the university were all significantly associated with mortality. Sex, race, age, smoking, HBV and HCV status, CD4, and type of cancer (ADC vs. NADC) were not associated with increased risk of mortality. Patients who ever had ART had significantly lower mortality than those who never received ART. In the multivariate model, an HIV RNA viral load less than 100 000 copies/ml remained significantly associated with mortality [adjusted hazard ratio, 2.22; 95% confidence interval (1.31–3.77)]. Patients receiving care at other/outside HIV care sites remained at higher risk for mortality [adjusted hazard ratio, 2.38; 95% confidence interval (1.52–3.71)]. Ever ART use, current IDU status, and Veterans Affairs site of HIV care were no longer significant in the multivariate analysis.
Eleven percent (50) of patients were LTFU before 30 June 2010. These patients were demographically similar (sex, race, and age) compared with the rest of the cohort; however, LTFU patients were more likely to have ADCs and to be followed at outside HIV clinics. Patients with NADCs and those in the Veterans Affairs system were less likely to be LTFU.
In this urban, mainly African–American, HIV-infected population from 2000 to mid-2010, there was a trend toward an increasing incidence of cancer. These cancers were more frequently NADCs than ADCs. However, ADCs, and especially NHL, continue to be regularly diagnosed, even in the late HAART era (after 2005). The study population was 85% African–American with a high percentage of IDU (45.6%), a demographic disproportionately affected by the HIV epidemic in the United States  and often suffering disparate cancer survival outcomes .
A significant proportion (8.9%) of the cohort was diagnosed with HIV at the time of cancer diagnosis (usually ADCs). Late presentation of HIV infection has been a persistent problem in the United States and is often associated with poor treatment outcomes . Concurrent diagnosis of HIV/AIDS and an ADC can be prevented, underlining the CDC's recommendation to expand HIV screening in the general population . With two-thirds of patients presenting with late-stage cancers, HIV providers may need to consider being more aggressive with cancer screening.
The causes of mortality in HIV-infected patients have transitioned from AIDS-related infections and complications of the pre-ART and early ART era of the 1980s and 1990s to primarily non-AIDS events during the HAART era of the late 1990s and 2000s. Malignancy is increasingly recognized as one of the most common causes of death among HIV-infected patients in the later HAART era [14,25]. In this study, mortality in patients dually diagnosed with HIV and cancer was high: 6.5, 32.2, and 41.4%, within 30 days, at 1 year, and at 2 years, respectively. The observation that 6.5% of patients died within 30 days of their initial cancer diagnosis likely relates to advanced cancer stage at presentation (67% at Stage 3 or 4). Late-stage cancer patients are often poor candidates for chemotherapy, referred directly to hospice, or may suffer potentially fatal treatment complications (e.g. tumor lysis syndrome).
The overall 2-year survival of 58.6% in this study was similar to that reported by Achenbach et al.  in patients with ART treatment drawn from a cohort from eight U.S. sites between 1996 and 2009. However, their population was mainly white (52%), and nonwhites had a statistically nonsignificant higher hazard ratio for death. The university's urban clinics are largely African–American and of lower socioeconomic status with higher rates of substance abuse and hepatitis C infection, a population that often suffers disparate cancer outcomes [20–22]; despite this fact and including some patients who never took ART or received cancer-specific therapy, 2-year survival was comparable to that of patients on ART with cancer around the country . Previous work has demonstrated that racial disparities in cancer outcomes may be mitigated when cancer treatment and medical care is comparable . It is possible that this cohort had similar outcomes as a result of high engagement in care and multidisciplinary coordination among HIV providers, oncologists, and others.
HIV-infected patients diagnosed with cancer are at high risk for death from a variety of causes, including progression of the underlying cancer, toxic effects related to cancer treatment, infections resulting from cancer treatment-related immunosuppression, as well as AIDS-related opportunistic infections. Accurate determination of the cause of death has important implications for both HIV and cancer management programs. Although a number of studies have documented high rates of mortality from cancer in HIV-infected patients [14,27] and high mortality rates in dually affected patients with HIV and cancer [7,15,28], actual causes of death in dually affected individuals are not well documented. Recently, Worm et al.  reported from the Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D) cohort that 88% of deaths in patients with NADC were caused by the underlying cancer. Simard et al.  have also shown high mortality risk attributable to a cancer diagnosis in patients with AIDS at a population level. In the present study, the majority of deaths with a known cause was associated with progression of the underlying cancer. Fewer than 5% of total deaths were attributed to infectious causes. Data were incomplete on a number of patients, which limit the generalizability of this observation. With a better understanding of the natural history of dually diagnosed patients, clinicians can focus on cancer treatment and maintaining ART as keys to prolonging survival.
The present study has several limitations. Some malignancies in our clinics may have been missed with retrospective identification of cases; however, by searching both clinic and hospital databases, we increased the probability of including nearly all cases. We excluded noninvasive cancers such as breast ductal carcinoma in situ, anal carcinoma in situ, and cervical intraepithelial neoplasia to strengthen the mortality analyses. As some patient outcomes were unknown, the significance of the Cox proportional hazards model results may be affected by missing data.
Although cases reported here are from a single institution, the outcomes may be representative of urban HIV-infected populations with high numbers of African–Americans and IDUs. Because the university is a referral center and some patients were followed in outside clinics, it was not possible to calculate overall cancer incidence and prevalence rates. Still, in the two main HIV clinics served, there was a significant trend of increasing overall cancer diagnoses over this time period. Factors contributing to an increasing incidence of cancer in our population may include the aging of the general HIV population, older age of the Veterans Affairs HIV population , high rates of behavioral cancer risk factors (e.g., smoking) in HIV-infected patients , and increased duration of HIV infection with longer periods of HIV viremia [32–34] and immune activation.
In conclusion the present study describes the cancers and subsequent outcomes in an urban, predominantly African–American, HIV-infected population. Outcomes were similar to those reported from a national cohort of predominantly white, ART-treated patients. However, the findings reinforce the CDC's recommendation to make HIV screening routine so that patients do not present late with undiagnosed HIV and an ADC. With two-thirds of patients presenting with Stage 3 and 4 disease, it raises the question of whether cancer screening practices should be the same for HIV-infected patients as for the general population. The data also suggest a survival benefit for patients on ART, which mandates that HIV providers and oncologists work together closely to co-manage these complex patients. With the aging of the HIV population, cancer screening, diagnosis, and treatment will assume increased significance in the years ahead.
D.J.R., L.E.F., C.A., C.D.P., R.R.R., B.L.G. contributed to study conception; D.J.R., L.E.F. to data acquisition; D.J.R., E.I.W.M., M.B.H. to data analysis and interpretation; D.J.R. to drafting the manuscript; and all contributed to revisions. Approval of the final manuscript: all.
D.R. received support from a Paul Calabresi Clinical Oncology Training Program K12 award. C.D.P. was supported by PHS grant CA142458 from the National Cancer Institute. D.R. had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Conflicts of interest
All authors declare no conflicts of interest.
1. Engels EA, Pfeiffer RM, Goedert JJ, Virgo P, McNeel TS, Scoppa SM, et al. Trends in cancer risk among people with AIDS in the United States 1980–2002
2. Engels EA, Biggar RJ, Hall HI, Cross H, Crutchfield A, Finch JL, et al. Cancer risk in people infected with human immunodeficiency virus in the United States
. Int J Cancer
3. Silverberg MJ, Chao C, Leyden WA, Xu L, Tang B, Horberg MA, et al. HIV infection and the risk of cancers with and without a known infectious cause
4. Goedert JJ, Cote TR, Virgo P, Scoppa SM, Kingma DW, Gail MH, et al. Spectrum of AIDS-associated malignant disorders
5. Centers for Disease Control and Prevention. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults
. MMWR Recomm Rep
6. Shiels MS, Pfeiffer RM, Gail MH, Hall HI, Li J, Chaturvedi AK, et al. Cancer burden in the HIV-infected population in the United States
. J Natl Cancer Inst
7. Long JL, Engels EA, Moore RD, Gebo KA. Incidence and outcomes of malignancy in the HAART era in an urban cohort of HIV-infected individuals
8. Shiels MS, Pfeiffer RM, Hall HI, Li J, Goedert JJ, Morton LM, et al. Proportions of Kaposi sarcoma, selected non-Hodgkin lymphomas, and cervical cancer in the United States occurring in persons with AIDS, 1980–2007
9. Frisch M, Biggar RJ, Engels EA, Goedert JJ. Association of cancer with AIDS-related immunosuppression in adults
10. Patel P, Hanson DL, Sullivan PS, Novak RM, Moorman AC, Tong TC, et al. Incidence of types of cancer among HIV-infected persons compared with the general population in the United States, 1992-2003
. Ann Intern Med
11. Biggar RJ, Engels EA, Ly S, Kahn A, Schymura MJ, Sackoff J, et al. Survival after cancer diagnosis in persons with AIDS
. J Acquir Immune Defic Syndr
12. Palella FJ Jr, Baker RK, Moorman AC, Chmiel JS, Wood KC, Brooks JT, et al. Mortality in the highly active antiretroviral therapy era: changing causes of death and disease in the HIV outpatient study
. J Acquir Immune Defic Syndr
13. Sackoff JE, Hanna DB, Pfeiffer MR, Torian LV. Causes of death among persons with AIDS in the era of highly active antiretroviral therapy: New York City
. Ann Intern Med
14. Simard EP, Engels EA. Cancer as a cause of death among people with AIDS in the United States
. Clin Infect Dis
15. Achenbach CJ, Cole SR, Kitahata MM, Casper C, Willig JH, Mugavero MJ, et al. Mortality after cancer diagnosis in HIV-infected individuals treated with antiretroviral therapy
16. Simard EP, Pfeiffer RM, Engels EA. Mortality due to cancer among people with AIDS: a novel approach using registry-linkage data and population attributable risk methods
17. Centers for Disease Control and Prevention (CDC). Disparities in diagnoses of HIV infection between blacks/African Americans and other racial/ethnic populations--37 states, 2005–2008
. MMWR Morb Mortal Wkly Rep
18. Maryland Department of Health and Mental Hygiene. Baltimore city HIV/AIDS epidemiological profile
. Baltimore: Maryland Department of Health and Mental Hygiene; 2011.
19. American Cancer Society. Cancer facts & figures 2011
. Atlanta: American Cancer Society; 2011.
20. Newman LA, Mason J, Cote D, Vin Y, Carolin K, Bouwman D, et al. African-American ethnicity, socioeconomic status, and breast cancer survival: a meta-analysis of 14 studies involving over 10 000 African-American and 40 000 White American patients with carcinoma of the breast
21. Schootman M, Jeffe DB, Lian M, Gillanders WE, Aft R. The role of poverty rate and racial distribution in the geographic clustering of breast cancer survival among older women: a geographic and multilevel analysis
. Am J Epidemiol
22. Schwartz K, Powell IJ, Underwood W 3rd, George J, Yee C, Banerjee M. Interplay of race, socioeconomic status, and treatment on survival of patients with prostate cancer
23. Waters L, Sabin CA. Late HIV presentation: epidemiology, clinical implications and management
. Expert Rev Anti Infect Ther
24. Branson BM, Handsfield HH, Lampe MA, Janssen RS, Taylor AW, Lyss SB, et al. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings
25. Bonnet F, Burty C, Lewden C, Costagliola D, May T, Bouteloup V, et al. Changes in cancer mortality among HIV-infected patients: the Mortalite 2005 Survey
. Clin Infect Dis
26. Bach PB, Schrag D, Brawley OW, Galaznik A, Yakren S, Begg CB. Survival of blacks and whites after a cancer diagnosis
27. Lewden C, Salmon D, Morlat P, Bevilacqua S, Jougla E, Bonnet F, et al. Causes of death among human immunodeficiency virus (HIV)-infected adults in the era of potent antiretroviral therapy: emerging role of hepatitis and cancers, persistent role of AIDS
. Int J Epidemiol
28. Monforte A, Abrams D, Pradier C, Weber R, Reiss P, Bonnet F, et al. HIV-induced immunodeficiency and mortality from AIDS-defining and non-AIDS-defining malignancies
29. Worm SW, Sabin C, Reiss P, Law M, Bonnet F, Fontas E, et al. Non-AIDS defining malignancies (NADM) in the D:A:D study: time trends and predictors of survival
. In: 13th European AIDS Conference
. Belgrade, Serbia; 2011.
30. Justice AC, Dombrowski E, Conigliaro J, Fultz SL, Gibson D, Madenwald T, et al. Veterans Aging Cohort Study (VACS): Overview and description
. Med Care
31. Saves M, Chene G, Ducimetiere P, Leport C, Le Moal G, Amouyel P, et al. Risk factors for coronary heart disease in patients treated for human immunodeficiency virus infection compared with the general population
. Clin Infect Dis
32. Bruyand M, Thiebaut R, Lawson-Ayayi S, Joly P, Sasco AJ, Mercie P, et al. Role of uncontrolled HIV RNA level and immunodeficiency in the occurrence of malignancy in HIV-infected patients during the combination antiretroviral therapy era: Agence Nationale de Recherche sur le Sida (ANRS) CO3 Aquitaine Cohort
. Clin Infect Dis
33. Zoufaly A, Stellbrink HJ, Heiden MA, Kollan C, Hoffmann C, van Lunzen J, et al. Cumulative HIV viremia during highly active antiretroviral therapy is a strong predictor of AIDS-related lymphoma
. J Infect Dis
34. Engels EA, Pfeiffer RM, Landgren O, Moore RD. Immunologic and virologic predictors of AIDS-related non-Hodgkin lymphoma in the highly active antiretroviral therapy era
. J Acquir Immune Defic Syndr