American Journal of Epidemiology

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (71) Request Permissions Disclaimer Google Scholar Articles by Gallagher, B. Articles by Fordyce, E. J. Search for Related Content PubMed PubMed Citation Articles by Gallagher, B. Articles by Fordyce, E. J. Social Bookmarking          What's this?

American Journal of Epidemiology Vol. 154, No. 6 : 544-556
Copyright © 2001 by The Johns Hopkins University School of Hygiene and Public Health

ORIGINAL CONTRIBUTIONS

Cancer Incidence in New York State Acquired Immunodeficiency Syndrome Patients

Brian Gallagher¹, Zhengyan Wang¹, Maria J. Schymura², Amy Kahn² and E. James Fordyce³

¹ New York State Department of Health, Bureau of HIV/AIDS Epidemiology, Albany, NY.
² New York State Department of Health, Bureau of Chronic Disease Epidemiology and Surveillance, Albany, NY.
³ New York City Department of Health, Office of AIDS Surveillance, New York, NY.

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
To identify cancers that occur at higher rates in acquired immunodeficiency syndrome (AIDS) patients, the cancer experience of New York State (NYS) AIDS patients aged 15–69 years who were diagnosed between 1981 and 1994 was compared with that of the NYS general population. Sex and HIV risk group-specific standardized incidence ratios (SIRs), post-AIDS relative risks, and trends of relative risks were calculated to determine cancer risk. Among non-AIDS-related cancers, elevated SIRs were found for Hodgkin's disease (male, 8.0; female, 6.4; heterosexually infected males, 31.3); cancer of the rectum, rectosigmoid, and anus (male, 3.3; female, 3.0); trachea, bronchus, and lung (male, 3.3; female, 7.5); and brain and central nervous system (male, 3.1; female, 3.4; heterosexually infected females, 23.8) cancers. Moreover, significant trends of increasing relative risks from the pre-AIDS to the post-AIDS period were found for cancers of the rectum, rectosigmoid, and anus; trachea, bronchus, and lung; skin; and connective tissues (all sites, p < 0.05) among males. For AIDS-related cancers in women, invasive cervical cancer had an overall SIR of 9.1 (95% confidence interval: 6.9, 10.8) and a post-AIDS relative risk of 6.5 (95% confidence interval: 4.1, 9.7). This population-based registry linkage analysis evaluates cancer risk in AIDS patients by sex and risk factors and adds evidence that HIV-associated immunosuppression increases the risks of specific types of cancer.

acquired immunodeficiency syndrome; HIV; lymphoma; AIDS-related; lung neoplasms; registries; risk factors

Abbreviations: AIDS, acquired immunodeficiency syndrome; CNS, central nervous system; HIV, human immunodeficiency virus; IDU, injection drug user; MSM, men who have sex with men; NHL, non-Hodgkin's lymphoma; NYS, New York State; RR, relative risk; SIR, standardized incidence ratio

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Cancer incidence in acquired immunodeficiency syndrome (AIDS) patients has been evaluated by population-based registry linkage analysis in the Unites States (1–3), Italy (4), and Australia (5). Cancer incidence has also been evaluated by using clinical trial cohorts and small case series. These studies have identified an increased incidence of specific types of malignant disease in patients with human immunodeficiency virus (HIV) infection. Kaposi's sarcoma, non-Hodgkin's lymphoma (NHL), and invasive cervical cancer have been designated as AIDS-related cancers (6). However, there is a lack of consistency in reported elevated risks for other HIV-associated malignancies (7–17). In addition, current analysis of linked data has focused on overall measures of association, not on stratifying by gender or known risk factors, leading to an incomplete assessment of cancer risk for different groups of AIDS patients. New York State (NYS) is the epicenter for the AIDS epidemic in the United States, with the largest number of AIDS cases of any state for both adults and children. NYS adult AIDS patients represent approximately 19 percent of all cases in the United States. An examination of data from NYS provides a unique opportunity to explore the role of immune system deficiency in cancer expression in which HIV prevalence is high. With the improved survival of patients with HIV infection because of better prevention and treatment of infectious complications and the development of more effective antiretroviral therapies, there may be an increase in malignant tumors and, in particular, those nondiagnostic of AIDS (13–17). It is difficult to predict the magnitude of change to cancer incidence in this population, and it will become increasingly important to know more about the risk of cancer in these patients.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Linkage procedure
To evaluate the risk of cancer among AIDS patients in NYS, we compared patients' cancer experience with that of the general population of NYS by matching the population-based NYS Cancer Registry with the combined NYS and New York City AIDS registries. A probabilistic algorithm was used to match name, birth date, and, when available, Social Security numbers between the AIDS and cancer registries. All cases of AIDS and cancer diagnosed between 1981 and 1994 were matched. Commercial matching software (18) was used by Oak Ridge National Laboratory staff to conduct the match as part of the National Cancer Institute's multistate AIDS/cancer match project.

Subjects and coding
Analysis of cancer incidence was restricted to AIDS and cancer cases diagnosed between 1981 and 1994 because these were the date ranges that were available. NYS residents aged 15–69 years were abstracted from the matched data set. Malignant disorders were grouped into types or sites of cancer by using the International Classification of Diseases, Ninth Revision codes (19). Although 57 separate sites or types of cancer were identified, analysis was limited to cancers with frequencies of four or more. Cancers were classified as AIDS defining if they were included in the 1993 AIDS case definition (6) or as non-AIDS defining if they were invasive malignancies not included in the AIDS case definition. Noninvasive neoplasms (International Classification of Diseases, Ninth Revision, codes 210–234) are also presented, although they are underreported to the cancer registry, and the results may be attenuated.

Data analysis
This study was limited to cancers diagnosed between 60 months before and 60 months after an AIDS diagnosis. This time interval was selected to make our results comparable with those of other published studies. The time interval before AIDS was also selected because there is evidence that the median duration of HIV infection before AIDS is at least 5 years (20–22). While cancers that occur in a patient prior to AIDS are, strictly speaking, HIV, not AIDS, related, we denote this period prior to an AIDS diagnosis as the pre-AIDS period. For the post-AIDS period, the median duration of survival post-AIDS diagnosis has been reported within the range of 3–26 (21) and 2–22 months (23). While these findings indicate that a short post-AIDS follow-up period may be sufficient to capture the majority of cancers that occur in AIDS patients after AIDS diagnosis, we have chosen to use an extended post-AIDS period to capture as many observations of cancer as possible in this time interval. Examining the NYS data, we found a small number of cancers occurring up to 5 years after AIDS diagnosis, and we chose to include these observations in the analysis. The calculation of person-time for the period at risk for cancer was defined as beginning in 1981 or 5 years prior to an AIDS diagnosis, whichever occurred later, and ending 5 years post-AIDS diagnosis, at the date of death, or on December 31, 1994, whichever occurred earliest. The observed number of cancers by type or site among people with AIDS was compared with the expected number of cases by calculating standardized incidence ratios (SIRs). Expected incident cases at the time of AIDS diagnosis or after AIDS diagnosis were calculated by multiplying NYS age-, sex-, region-, and race-specific cancer incidence rates from the NYS Cancer Registry by the corresponding person-years at risk for these time periods. However, calculation of the expected number of cancer cases preceding AIDS was complicated because HIV-infected subjects who develop cancer might die before progression to AIDS and therefore would never be recorded in the AIDS registry. In our study, the cohort that was matched to the cancer registry was diagnosed with AIDS; therefore, they survived until AIDS diagnosis. All of the cancer cases observed in this cohort diagnosed before AIDS survived to an AIDS diagnosis. Since the observed cancer cases survived until AIDS diagnosis, the expected cancer cases must be adjusted to reflect survival to AIDS diagnosis. Therefore, the joint probability of developing cancer and of surviving cancer was applied to the person-year distribution, and NYS cancer incidence rates and Surveillance, Epidemiology, and End Results differential survival rates for specific malignancies were used to calculate the expected number of cancer cases prior to AIDS diagnosis (24). All SIRs discussed in the results are adjusted. To calculate confidence intervals around the SIR, either the normal or the Poisson distribution was used, depending on the number of expected cases. If the number of expected cases was equal to or greater than 25, the normal distribution was used; otherwise, the Poisson distribution was used.

For determination of whether cancer risk increases with time, as the immune status of the HIV-infected patient declines, the time period between a cancer and AIDS diagnosis was divided into four intervals: early pre-AIDS diagnosis (-60 to -25 months), late pre-AIDS diagnosis (-24 to -7 months), at AIDS diagnosis (-6 to 3 months), and post-AIDS diagnosis (4–60 months) (figure 1). The trend in relative risk was analyzed over three time periods (the at AIDS time period was excluded) by using the Poisson trend statistic (25).

View larger version (8K): [in this window] [in a new window]   FIGURE 1. The period of 120 months during which cancer risk was assessed among people with acquired immunodeficiency syndrome (AIDS) in New York State, 1981–1994. Population-based AIDS and cancer registries were linked to identify people with AIDS who had ever had cancer reported to the cancer registry. The month of initial AIDS-related disease was defined as month +1 (). Cancer prevalence and incidence rates among people with AIDS were compared with those expected to have occurred among people of the same age, race, and sex distribution as people with AIDS.

 
In this study, a non-AIDS-related cancer is considered to be AIDS related when the overall SIR for the period from 60 months before to 60 months after AIDS diagnosis is statistically significantly elevated and there is a statistically significant increasing trend in relative risks from the pre-AIDS to the post-AIDS period.

To explore the association between risk factors and malignant disease, SIRs were calculated for groups defined by HIV exposure group (men who have sex with men (MSM) combined with MSM and intravenous drug user (IDU), heterosexual contact with HIV/AIDS patients, IDUs, patients infected via transfusion or transplantation, and unknown or other HIV exposure).

All statistical analyses were performed by using the Statistical Analysis System (26).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Overall, 12,698 instances of cancer among 122,993 people with AIDS were identified through linkage of the AIDS and cancer registries. There were 11,371 cancers with diagnosis dates within 60 months before and 60 months after an AIDS diagnosis. The characteristics of the study cohort are displayed in table 1. Of the 11,371 persons with cancer in this cohort, 82.0 percent were NYC residents, 18.0 percent were Upstate New York residents, 88.7 percent were men, and 11.3 percent were women. Among 10,083 men with AIDS, 84.6 percent were diagnosed with AIDS between ages 25–49 years, 77.9 percent were non-Black, 22.1 percent were Black, 76.6 percent were MSM or MSM combined with IDUs, and 17.7 percent were IDUs. Of the 1,288 women with AIDS, 86.1 percent were diagnosed with AIDS between ages 25–49 years, 55.2 percent were non-Black, 44.8 percent were Black, 53.8 percent were IDUs, and 31.5 percent were women who had heterosexual contact. There were a total of 567,254 person-years at risk for cancer—453,270 for men and 113,984 for women.

View this table: [in this window] [in a new window]   TABLE 1. Demographic characteristics and route of human immunodeficiency virus acquisition among patients aged 15–69 years in the acquired immunodeficiency syndrome/cancer matched cohort, New York State, 1981–1994

 
AIDS-related cancers and overall SIRs
Compared with incidence in the NYS general population, the incidence of Kaposi's sarcoma was increased 97.5-fold in men and 202.7-fold in women. The risk of NHL was increased 37.4-fold in men and 54.6-fold in women. Invasive cervical cancer was increased 9.1-fold (tables 2 and 3).

View this table: [in this window] [in a new window]   TABLE 2. Observed and expected cancer cases and standardized incidence ratios of specific types of cancer among men with acquired immunodeficiency syndrome aged 15–69 years in the acquired immunodeficiency/cancer matched cohort, New York State, 1981–1994

 

View this table: [in this window] [in a new window]   TABLE 3. Observed and expected cancer cases and standardized incidence ratios of specific types of cancer in women with acquired immunodeficiency syndrome aged 15–69 years in the acquired immunodeficiency/cancer matched cohort, New York State, 1981–1994

 
AIDS-related cancers and post-AIDS relative risks
The incidence of Kaposi's sarcoma was increased 86.3-fold in men and 266.2-fold in women. The risk of NHL was increased 48.6-fold in men and 96.3-fold in women. Invasive cervical cancer was increased 6.5-fold (table 4).

View this table: [in this window] [in a new window]   TABLE 4. Observed and expected cancer cases and relative risks of specific types of cancer after acquired immunodeficiency syndrome diagnosis in the acquired immunodeficiency syndrome/cancer matched cohort, New York State, 1981–1994

 
Non-AIDS-related cancers and overall SIRs
For all non-AIDS-related cancers combined, the SIR was 2.6 for men and 2.2 for women, and significantly elevated SIRs were found for several sites and types of cancer in both men and women (tables 2 and 3). Cancers with significant SIRs were tongue (male, 1.8; female 7.1); gum and other mouth (male, 1.9; female, 11.1); rectum, rectosigmoid, and anus (male, 3.3; female, 3.0); larynx (male, 1.9; female, 5.9); trachea, bronchus, and lung (male, 3.3; female, 7.5); skin, excluding Kaposi's sarcoma (male, 20.9; female, 7.5); brain and central nervous system (CNS) (male, 3.1; female, 3.4); Hodgkin's disease (male, 8.0; female, 6.4); and multiple myeloma (male, 2.7; female, 7.2). We also found significantly elevated SIRs for the pharynx (2.0), stomach (1.8), liver (5.1), testis (1.5), endocrine gland (3.8), and connective tissue (5.6) in men and for the esophagus (8.7) and leukemias (6.1) in women.

Non-AIDS-related cancers and post-AIDS relative risks
For all non-AIDS-related cancers combined, the relative risk was 3.1 for men and 2.9 for women, and significantly elevated relative risks were found for several sites and types of cancer in both men and women (table 4). Cancers with significant relative risks for men and women were liver (relative risk (RR) = 3.6 and 9.2, respectively); trachea, bronchus, and lung (RR = 3.0 and 7.1, respectively); skin, excluding Kaposi's sarcoma (RR = 44.6 and 12.0, respectively); brain and CNS (RR = 4.0 and 6.2, respectively); Hodgkin's disease (RR = 5.3 and 7.2, respectively); and multiple myeloma (RR = 3.5 and 9.2, respectively). We also found significantly elevated relative risks for the rectum, rectosigmoid, and anus (RR = 4.0) and connective tissue (RR = 9.9) in men and for gum and other mouth (RR = 13.5), esophagus (RR = 8.7), stomach (RR = 5.4), larynx (RR = 10.6), and leukemias (RR = 5.2) in women.

Non-AIDS-related cancers and relative risk trend
Significant increases in relative risk over time were found for several sites in men and women. For some of the sites with significant trends, the changes in relative risks over time were not linear. For men, cancers with significant trends were digestive system (p = 0.002); Hodgkin's disease (p = 0.0135); rectum, rectosigmoid, and anus (p = 0.0164); trachea, bronchus, and lung (p = 0.0046); brain and CNS (p = 0.0018); melanoma of the skin (p = 0.0478); skin, excluding Kaposi's sarcoma (p < 0.0000); leukemias (p = 0.0087); and connective tissue cancers (p < 0.0000) (table 5). The only site of non-AIDS-related cancer found to increase significantly over time in women was the digestive system (p = 0.0176) (table 6).

View this table: [in this window] [in a new window]   TABLE 5. Relative risks for selected types of cancer by time period among men with acquired immunodeficiency syndrome aged 15–69 years in the acquired immunodeficiency/cancer matched cohort, New York State, 1981–1994

 

View this table: [in this window] [in a new window]   TABLE 6. Relative risks for selected types of cancer by time period among women with acquired immunodeficiency syndrome aged 15–69 years in the acquired immunodeficiency/cancer matched cohort, New York State, 1981–1994

 
Noninvasive, non-AIDS-related cancers, overall SIRs, post-AIDS relative risks, and relative risk trend
Men had a significantly elevated SIR for other in situ neoplasms (SIR = 4.3) (table 2) and a post-AIDS relative risk of 4.2 (table 5). Women had an SIR of 5.2 (table 3) and a post-AIDS relative risk of 8.6 for cervix in situ (table 6). Significant trends in relative risks were found for benign neoplasms (p = 0.0050) (table 5) in men and for in situ cervical cancer (p < 0.0000) in women (table 6).

Non-AIDS-related cancers with SIRs lower than those of the general population
In men, prostate (SIR = 0.7) and bladder (SIR = 0.5) cancers both had SIRs below 1.0 (table 2). These sites had relative risks below 1.0 during all periods relative to AIDS diagnosis, including the AIDS period (table 5). For women, breast cancer had an SIR of 0.8 (95 percent confidence interval: 0.58, 1.04) (table 3). Relative risks for breast cancer in the early pre-AIDS and post-AIDS periods were below 1.0, with the post-AIDS period being especially low (RR = 0.2) (table 6).

Non-AIDS-related cancer risk by HIV exposure group
Cancers with significantly elevated overall SIRs were evaluated by HIV exposure group. All results are displayed in table 7. Results for groups with five or more observations are presented as follows.

View this table: [in this window] [in a new window]   TABLE 7. Selected cancer risks by human immunodeficiency virus exposure group among patients with acquired immunodeficiency syndrome in the acquired immunodeficiency syndrome/cancer matched cohort, New York State, 1981–1994

 
Anal. Men who acquired HIV infection through homosexual contact had an SIR of 5.8.

Hodgkin's disease. Male homosexual contact (SIR = 8.7) and intravenous drug use (SIR = 7.8) had comparable risks, while the SIR for heterosexual contact was 31.3. For female heterosexual contact, there was a higher standardized incidence ratio (SIR = 12.3) than for injection drug use (SIR = 9.4).

Trachea, bronchus, and lung. There was a higher SIR for male injection drug use (SIR = 4.6) than for male homosexual contact (SIR = 2.6). For female injection drug use, there was an elevated standardized incidence ratio (SIR = 9.8) compared with female heterosexual contact (SIR = 6.4).

Brain and CNS. In men, the SIR for homosexual contact was 4.7, which was higher than that for intravenous drug use (SIR = 2.7), while for women, the SIR for heterosexual contact was 23.8.

Connective tissue. Among men who acquired HIV infection through homosexual contact, the risk was increased 10.5-fold.

Oral cavity and pharynx. The SIRs for male and female intravenous drug use were 1.9 and 8.2, respectively.

Digestive system. For male intravenous drug use, there was an elevated SIR of 1.8, and the SIR for female heterosexual contact (SIR = 3.1) was higher than that for female intravenous drug user (SIR = 2.4).

Invasive cervix. The SIR for heterosexual contact was higher (SIR = 10.7) than that for intravenous drug use (SIR = 8.7).

In situ cervix. The SIRs for heterosexual contact (SIR = 5.4) and intravenous drug use (SIR = 5.3) were comparable.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In this study, we assessed the risk of cancer among people diagnosed with AIDS in NYS. The matched AIDS/cancer data for NYS is included as a part of the National Cancer Institute's multistate AIDS/cancer match project. While the National Cancer Institute has conducted some analysis of the national data (27, 28), in this paper, we describe the results of analysis of the NYS subset of the data, including analysis by risk group. As expected, Kaposi's sarcoma and non-Hodgkin's lymphoma represented the majority of cancers found in this cohort. There is strong evidence in this study that the immunosuppression associated with HIV infection increases the risks of Kaposi's sarcoma, non-Hodgkin's lymphoma, and invasive cervical cancer. This study has also found elevated SIRs, post-AIDS relative risks, and increasing trends of relative risks in people with AIDS for several types of cancer that are not AIDS defining. It is possible that our results for tumors of the brain, digestive system, and lungs could be misdiagnosed NHL, but an examination of the histology codes for these sites did not reveal miscoded lymphomas.

Hodgkin's disease
Hodgkin's disease is one of the most common non-AIDS-related cancers found in this cohort. The overall SIRs (male, 8.0; female, 6.4; both, 7.8) in our study are consistent with studies from San Francisco (male, 8.8) (9), the United States and Puerto Rico (both 8.8) (2), and Italy (male, 9.3; female, 7.7) (4), but are lower than those in the paper by Grulich et al. (5) (both 18.3). The post-AIDS relative risks (male, 5.3; female, 7.2; both, 5.4) are also similar to those in the paper by Goedert et al. (2) (both 7.6), but are much lower than those found by Franceschi et al. (4) (both 20.4) and Grulich et al. (5) (both 29.9). A significant, but nonlinear, increase in relative risk over time was found in male AIDS patients in this study (p = 0.0135). Goedert et al. (2) previously found a significant increase in relative risk over time for both sexes; therefore, our study contributes to the evidence that the risk of Hodgkin's disease increases as males with HIV disease advance in their immunodeficiency.

Rectal and anal cancer
There is strong evidence of a relation between HIV-related immunodeficiency, human papilloma virus infection, and development of anal cancer (29–34). Carcinomas of the anus and anal intraepithelial neoplasia occur at increased rates among homosexual men with HIV. The incidence among MSM was elevated in 1973–1979, which preceded the HIV epidemic (35). In our study, elevated overall SIRs for rectal and anal cancer were found (male, 3.3; female, 3.0), while post-AIDS relative risks were significant for men (RR = 4.0) but not for women. We also detected an elevated relative risk in male AIDS patients with declining immune status (p = 0.0164). Several other studies have also found an increased incidence of anal cancer in men but not in women (10, 36, 37).

Lung cancer
Lung cancer is the second most common cancer in both men and women in the general population. In our study, we found a significant overall SIR (male, 3.3; female, 7.5; both, 3.7), similar to a report by Grulich et al. (5) (both 3.8). Post-AIDS relative risks were also significantly elevated (male, 3.0; female, 7.1; both, 3.3), and this is also similar to the post-AIDS result from Grulich et al. (5) (both 3.9). A significant (p = 0.0046) increase in the trend in relative risk was also found in men.

This study found a significantly increased risk of lung cancer with advancing immunodeficiency. However, this result may be confounded, since it is widely known that lung cancer is related to tobacco use, and it is likely that HIV-infected patients smoke more cigarettes per day than do persons in the general population.

Brain and CNS cancer
In recent years, incidence of brain and CNS cancer has risen strikingly, and CNS lymphoma in people with AIDS is thought to represent much of the recent increase (38–40). In our study, we found elevated SIRs (male, 3.1; female, 3.4; both, 3.2), which is higher than the findings in the study by Goedert et al. (2) (both 2.0). Our analysis also revealed elevated post-AIDS relative risks (male, 4.0; female, 6.2; both, 4.3), which are higher than those in the study by Goedert et al. (5) (both 3.5) and a significant, but nonlinear, trend in relative risks over time for men (p = 0.0018). The trend in the paper by Goedert et al. was significant (both p = 0.006) and linear. Other studies that have examined the relation between brain and CNS cancer and HIV illness have had mixed results (41–43), and brain and CNS cancer has not been clearly linked to immunodeficiency. The etiology of CNS lymphoma among persons with and those without AIDS remains unknown, and infectious viral agents such as Epstein-Barr virus might contribute to lymphoma.

Skin cancer
For skin cancer, we found relatively high SIRs (male, 20.9; female, 7.5), even higher relative risks in the post-AIDS period (male, 44.6; female, 12.0), and an increased risk over time in men (p < 0.0001). The two most common forms of skin cancer, basal and squamous cell carcinomas, are not reportable to the registry. Consequently, most nonmelanomatous skin cancers reported to the registry are sarcomas, followed by adenocarcinomas. It is possible that Kaposi's sarcoma cases might have been misclassified as non-Kaposi's sarcoma skin cancers.

Connective tissue
Cancer of connective tissues occurred in statistically significant excess among men. The overall SIR was 5.6, the post-AIDS relative risk was 9.9, and the p value was less than 0.0001. Grulich et al. (5) have also found an elevated SIR of 9.2 for both sexes combined.

Noninvasive neoplasms
An increased risk for in situ cancers was found in this study (tables 2 and 3). We present findings from our data to allow comparison with other published data; however, there are serious potential biases relating to the interpretation of in situ cancer results. Reporting of in situ cancers is incomplete in the cancer registry. This is especially true for in situ cervical cancer. In fact, most state cancer registries have dropped in situ cervical cancer from their reportable list. Thus, rates for in situ cancers based on registry data would be artificially low, and the number of expected cases would also be low, so the SIRs would not be accurate and would tend to be elevated among a population with an increased likelihood of hospitalization.

Non-AIDS-related cancers with SIRs lower than the general population
While the focus of the analysis is ascertaining cancers with elevated SIRs, we found a few sites with SIRs below 1.0. In men, prostate (SIR = 0.7) and bladder (SIR = 0.5) cancers had SIRs below 1.0 (table 2). These sites had relative risks below 1.0 during all periods relative to AIDS diagnosis. In women, breast cancer relative risks during the early pre-AIDS and post-AIDS periods were below 1.0, with the post-AIDS period especially low (RR = 0.2) (table 6).

Risk factors
A person's risk of developing a specific cancer is affected by various factors, including age, sex, race, and exposure to environmental agents. As yet, there is only limited information about which characteristics and exposures of HIV-infected people promote the development of cancer. The association between risk factors for acquiring HIV infection and selected malignant disease in people with AIDS was examined in this study. Sex between men and intravenous drug use are the major behavioral risk factors for men in our study cohort. For six cancer types or sites (rectum, rectosigmoid, and anus; connective tissue; Hodgkin's disease; brain and CNS; oral cavity and pharynx; and trachea, bronchus, and lung), our study contains sufficient observations to detect meaningful differences between these two risk groups for men and six cancer types or sites for women (invasive cervix; in situ cervix; Hodgkin's disease; trachea, bronchus, and lung; oral cavity and pharynx; and digestive system).

Rectum, rectosigmoid, and anus
Our study confirms that the risk of anal cancer among homosexual men with HIV infection is higher than that in the general population.

Trachea, bronchus, and lung
Both male (SIR = 4.6) and female (SIR = 9.8) IDUs had increased SIRs for trachea, bronchus, and lung cancer compared with other risk groups. This finding is not surprising, since IDUs are known to be heavy smokers, which may confound the relation between HIV and trachea, bronchus, and lung cancer.

Brain and CNS
SIRs varied by route of HIV acquisition in men, MSM (SIR = 4.7), and IDUs (SIR = 2.7), although the confidence intervals overlap for these point estimates. While differences by risk group are noteworthy, it is likely that route of HIV infection is a proxy for other risks, such as coinfection with Epstein-Barr virus, hepatitis B or C, or human herpes virus 8. Women infected heterosexually had a relative risk of 23.8.

Connective tissue
In this study, risk of this cancer appeared almost exclusively among HIV-infected men with homosexual contact, and therefore, this may be a result of misdiagnosis of Kaposi's sarcoma as hemangiosarcoma.

Hodgkin's disease
The relative risk for the MSM risk group, 8.7, is comparable with that in the study by Reynolds et al. (9) (RR = 8.8). The relative risk for male IDUs, 7.8, is similar to the that of 8.1 for males and females in the study by Franceschi et al. (4). Heterosexually infected men had a greatly elevated relative risk (RR = 31.3), which was significantly higher than that for the intravenous drug use risk group.

Oral cavity and pharynx
SIRs were elevated for IDUs (1.9) but not for any other risk group.

Limitations
There are several potential biases in this study. First, it is possible that cancers with a higher incidence were detected because of intensive medical scrutiny among AIDS patients. Second, this cohort is the result of a match between two registries. If records are erroneously matched, the measures of association may be inaccurate. Third, in this study, Surveillance, Epidemiology, and End Results differential survival rates were applied to expected frequencies of cancer that occurred preceding AIDS. These adjusted relative risks for pre-AIDS expected cancers were then used to calculate the pre-AIDS relative risk and overall SIRs. However, this adjustment was approximate and could have resulted in overestimates of the early pre-AIDS prevalence and expected cancer cases.

Another potential limitation relates to the method we used to calculate post-AIDS relative risk. For types and sites of cancer that did not contribute a large number of observed cases in the 60-month post-AIDS period, the relative risks for this period will be attenuated due to the large number of expected cases that are derived with no or few corresponding observed cases.

This analysis was conducted on heterogeneous case definition data, with 1993 and 1987 AIDS definition cases analyzed together. Patients classified as 1987 definition cases are at a more advanced stage of HIV infection compared with 1993 definition cases, as indicated by significant differences in survival probabilities between the two groups (44, 45). A final potential limitation is that pulmonary and neurologic complications of HIV illness may have been misdiagnosed as tumors at these sites.

In a large, heterogeneous cohort of AIDS patients, we have confirmed that cancer occurs in excess for AIDS-related cancers, and for non-AIDS-related cancers, our results support previous findings for Hodgkin's disease, rectal and anal cancer, and brain and CNS system cancer. This analysis has contributed to the knowledge of the relation between AIDS and cancer by analyzing incidence by gender and risk group. Our analysis revealed an increasing relative risk of trachea, bronchus, and lung cancer in men over time and an increase in risk of cancer of connective tissues for MSM.

	ACKNOWLEDGMENTS

 
Supported in part by grant U62/CCU206207-08 from the Centers for Disease Control and Prevention.

	NOTES

 
Reprint requests to Brian Gallagher, New York State Department of Health, Bureau of HIV/AIDS Epidemiology, Room 729, Corning Tower, ESP, Albany, NY 12237 (email: bxg04{at}health.state.ny.us).

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Biggar RJ, Rosenberg PS, Cote T. Kaposi's sarcoma and non-Hodgkin's lymphoma following the diagnosis of AIDS. Multistate AIDS/Cancer Match Study Group. Int J Cancer 1996;68:754–8.[Web of Science][Medline]
2. Goedert JJ, Cote TR, Virgo P, et al. Spectrum of AIDS-associated malignant disorders. Lancet 1998;351:1833–9.[Web of Science][Medline]
3. Cote TR, O'Brien TR, Ward JW, et al. AIDS and cancer registry linkage: measurement and enhancement of registry completeness. The National AIDS/Cancer Match Study Group. Prev Med 1995;24:375–7.[Web of Science][Medline]
4. Franceschi S, Dal Maso L, Arniani S, et al. Risk of cancer other than Kaposi's sarcoma and non-Hodgkin's lymphoma in persons with AIDS in Italy. Br J Cancer 1998;78:966–70.[Web of Science][Medline]
5. Grulich AE, Wan X, Law MG, et al. Risk of cancer in people with AIDS. AIDS 1999;13:839–43.[Web of Science][Medline]
6. Centers for Disease Control. 1993 revised classification for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR Morbid Mortal Wkly Rep 1992;41/RR-17:1–19.
7. Feigal EG, Cote TR. Thoracic neoplasms associated with human immunodeficiency virus infection. In: Aisner J, ed. Comprehensive textbook of thoracic oncology. Baltimore, MD: Williams & Wilkins, 1996.
8. Hessol NA, Katz MH, Liu JY, et al. Increased incidence of Hodgkin's disease in homosexual men with HIV infection. Ann Intern Med 1992;117:309–11.
9. Reynolds P, Saunders LD, Layefsky ME, et al. The spectrum of acquired immunodeficiency syndrome (AIDS)-associated malignancies in San Francisco, 1980–1987. Am J Epidemiol 1993;137:19–30.[Abstract/Free Full Text]
10. Melbye M, Cote TR, Kessler L, et al. High incidence of anal cancer among AIDS patients. Lancet 1994;343:636–9.[Web of Science][Medline]
11. Goedert JJ, Cote TR. Conjunctival malignant diseases with AIDS in USA. (Letter). Lancet 1995;346:257–8.
12. Bavinck JN, De Boer A, Vermeer BJ, et al. Sunlight, keratotic skin lesions and skin cancer in renal transplant recipients. Br J Dermatol 1993;129:242–9.[Web of Science][Medline]
13. Masur H, Michelis MA, Greene JB, et al. An outbreak of community-acquired Pneumocystis carinii pneumonia: initial manifestation of cellular immune dysfunction. N Engl J Med 1981;305:1431–8.[Abstract]
14. Hymes KB, Cheung T, Greene JB, et al. Kaposi's sarcoma in homosexual men: a report of eight cases. Lancet 1981;2:598.[Web of Science][Medline]
15. Gallo RC, Salahuddin SZ, Popovic M, et al. Frequent detection and isolation of cytopathic retroviruses (HTLV-III) from patients with AIDS and at risk for AIDS. Science 1984;224:500–3.[Abstract/Free Full Text]
16. Ahmed T, Wormser GP, Stahl RE, et al. Malignant lymphomas in a population at risk for acquired immunodeficiency syndrome. Cancer 1987;60:719–23.[Web of Science][Medline]
17. Monfardini S, Vaccher E, Pizzocaro G, et al. Unusual malignant tumors in 49 patients with HIV infection. AIDS 1989;3:449–52.[Web of Science][Medline]
18. Jaro MA. Automatch generalized record linkage system, version 2.7. Silver Spring, MD: Matchware Technologies, 1994.
19. World Health Organization. International classification of diseases, injuries, and causes of death. Ninth Revision. Geneva, Switzerland: World Health Organization, 1979.
20. Time from HIV-1 seroconversion to AIDS and death before widespread use of highly-active antiretroviral therapy: a collaborative re-analysis. Collaborative Group on AIDS Incubation and HIV Survival including the CASCADE EU Concerted Action. Concerted Action on SeroConversion to AIDS and Death in Europe. Lancet 2000;355:1131–7.[Web of Science][Medline]
21. van Benthem BH, Veugelers PJ, Cornelisse PG, et al. Is AIDS a floating point between HIV seroconversion and death? Insights from the Tricontinental Seroconverter Study. AIDS 1998;12:1039–45.[Web of Science][Medline]
22. Mocroft AJ, Lundgren JD, d'Armino Monforte A, et al. Survival of AIDS patients according to type of AIDS-defining event. The AIDS in Europe Study Group. Int J Epidemiol 1997;26:400–7.[Abstract/Free Full Text]
23. Alioum A, Leroy V, Commenges D, et al. Effect of gender, age, transmission category, and antiretroviral therapy on the progression of human immunodeficiency virus infection using multistate Markov models. Epidemiology 1998;9:605–12.[Web of Science][Medline]
24. National Cancer Institute, Division of Cancer Control and Population Science. Surveillance Research Program, Cancer Statistics Branch. Surveillance, Epidemiology, and End Results (SEER) program public use CD-Rom (1973–1996). Atlanta, GA: National Cancer Institute, 1999.
25. Breslow NE, Day NE, eds. Statistical methods in cancer research. Vol 2. The design and analysis of cohort studies. Lyon, France: International Agency for Research on Cancer, 1987.
26. SAS Institute, Inc. Statistical analysis system, version 6.12. Cary, NC: SAS Institute, Inc, 1996.
27. Biggar RJ, Frisch M. Estimating the risk of cancer in children with AIDS. JAMA 2000;284:2593–4.[Free Full Text]
28. Frisch M, Biggar RJ, Goedert JJ. Human papillomavirus-associated cancers in patients with human immunodeficiency virus infection and acquired immunodeficiency syndrome. J Natl Cancer Inst 2000;92:1500–10.[Abstract/Free Full Text]
29. Carter PS, de Ruiter A, Whatrup C, et al. Human immunodeficiency virus infection and genital warts as risk factors for anal intraepithelial neoplasia in homosexual men. Br J Surg 1995;82:473–9.[Web of Science][Medline]
30. Frazer IH, Medley G, Crapper RM, et al. Association between anorectal dysplasia, human papillomavirus, and human immunodeficiency virus infection in homosexual men. Lancet 1986;2:657–61.[Web of Science][Medline]
31. Kiviat N, Rompalo A, Bowden R, et al. Anal human papillomavirus infection among immunodeficiency virus-seropositive and -seronegative men. J Infect Dis 1990;182:358–61.
32. Northfelt DW. Cervical and anal neoplasia and HPV infection in persons with HIV infection. Oncology (Huntingt) 1994;8:33–40.[Medline]
33. Palefsky JM. Anal human papillomavirus infection and anal cancer in HIV-positive individuals: an emerging problem. AIDS 1994;8:283–95.[Web of Science][Medline]
34. Palefsky JM, Holly EA, Gonzales J, et al. Natural history of anal cytologic abnormalities and papillomavirus infection among homosexual men with group IV HIV disease. J Acquir Immune Defic Syndr 1992;5:1258–65.
35. Rabkin CS. Association of non-acquired immunodeficiency syndrome-defining cancers with human immunodeficiency virus infection. J Natl Cancer Inst Monogr 1998;23:23–5.
36. Melbye M, Rabkin C, Frisch M, et al. Changing patterns of anal cancer incidence in the United States, 1940–1989. Am J Epidemiol 1994;139:772–80.[Abstract/Free Full Text]
37. Lobo DV, Chu P, Grekin RC, et al. Nonmelanoma skin cancers and infection with the human immunodeficiency virus. Arch Dermatol 1992;128:623–7.[Abstract/Free Full Text]
38. Miller BA, Ries LA, Hankey BF, et al. Cancer statistics review: 1973–1989. Bethesda, MD: National Cancer Institute, 1992. (DHHS publication no. (NIH) 92–2789).
39. Eby NL, Grufferman S, Flannelly CM, et al. Increasing incidence of primary brain lymphoma in the US. Cancer 1988;62:2461–5.[Web of Science][Medline]
40. Fine HA, Mayer RJ. Primary central nervous system lymphoma. Ann Intern Med 1993;119:1093–104.[Abstract/Free Full Text]
41. Mueller N, Hinkula J, Wahren B. Elevated antibody titers against cytomegalovirus among patients with testicular cancer. Int J Cancer 1988;41:399–403.[Web of Science][Medline]
42. Heinzer H, Dieckmann KP, Huland E. Virus-related serology and in-situ hybridization for the detection of virus DNA among patients with testicular cancer. Eur Urol 1993;24:271–6.[Web of Science][Medline]
43. Arthur RR, Grossman SA, Ronnett BM, et al. Lack of association of human polyomaviruses with human brain tumors. J Neurooncol 1994;20:55–8.[Medline]
44. Brettle RP, Gore SM, Bird AG, et al. Clinical and epidemiological implications of the Centers for Disease Control/World Health Organization reclassification of AIDS cases. AIDS 1993;7:531–9.[Web of Science][Medline]
45. Vella S, Chiesi A, Volpi A, et al. Differential survival of patients with AIDS according to the 1987 and 1993 CDC case definitions. JAMA 1994;271:1197–9.[Abstract/Free Full Text]

Received for publication July 18, 2000. Accepted for publication March 12, 2001.

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				E J Bini, B Green, and M A Poles Screening colonoscopy for the detection of neoplastic lesions in asymptomatic HIV-infected subjects Gut, August 1, 2009; 58(8): 1129 - 1134. [Abstract] [Full Text] [PDF]

				P. Patel, D. L. Hanson, P. S. Sullivan, R. M. Novak, A. C. Moorman, T. C. Tong, S. D. Holmberg, J. T. Brooks, and for the Adult and Adolescent Spectrum of Disease P Incidence of Types of Cancer among HIV-Infected Persons Compared with the General Population in the United States, 1992-2003 Ann Intern Med, May 20, 2008; 148(10): 728 - 736. [Abstract] [Full Text] [PDF]

				J.-S. Nam, M. Terabe, M.-J. Kang, H. Chae, N. Voong, Y.-a. Yang, A. Laurence, A. Michalowska, M. Mamura, S. Lonning, et al. Transforming Growth Factor {beta} Subverts the Immune System into Directly Promoting Tumor Growth through Interleukin-17 Cancer Res., May 15, 2008; 68(10): 3915 - 3923. [Abstract] [Full Text] [PDF]

				S. Sutcliffe, E. Giovannucci, C. A. Gaydos, R. P. Viscidi, F. J. Jenkins, J. M. Zenilman, L. P. Jacobson, A. M. De Marzo, W. C. Willett, and E. A. Platz Plasma Antibodies against Chlamydia trachomatis, Human Papillomavirus, and Human Herpesvirus Type 8 in Relation to Prostate Cancer: A Prospective Study Cancer Epidemiol. Biomarkers Prev., August 1, 2007; 16(8): 1573 - 1580. [Abstract] [Full Text] [PDF]

				N. A. Hessol, S. Pipkin, S. Schwarcz, R. D. Cress, P. Bacchetti, and S. Scheer The Impact of Highly Active Antiretroviral Therapy on Non-AIDS-Defining Cancers among Adults with AIDS Am. J. Epidemiol., May 15, 2007; 165(10): 1143 - 1153. [Abstract] [Full Text] [PDF]

				S. Philippe, C. Sarkis, M. Barkats, H. Mammeri, C. Ladroue, C. Petit, J. Mallet, and C. Serguera Lentiviral vectors with a defective integrase allow efficient and sustained transgene expression in vitro and in vivo PNAS, November 21, 2006; 103(47): 17684 - 17689. [Abstract] [Full Text] [PDF]

				J Cadranel, D Garfield, A Lavole, M Wislez, B Milleron, and C Mayaud Lung cancer in HIV infected patients: facts, questions and challenges. Thorax, November 1, 2006; 61(11): 1000 - 1008. [Abstract] [Full Text] [PDF]

				E. J. Bini, J. Park, and F. Francois Use of Flexible Sigmoidoscopy to Screen for Colorectal Cancer in HIV-Infected Patients 50 Years of Age and Older. Arch Intern Med, August 14, 2006; 166(15): 1626 - 1631. [Abstract] [Full Text] [PDF]

				G. T. Stathopoulos, Z. Zhu, M. B. Everhart, I. Kalomenidis, W. E. Lawson, S. Bilaceroglu, T. E. Peterson, D. Mitchell, F. E. Yull, R. W. Light, et al. Nuclear Factor-{kappa}B Affects Tumor Progression in a Mouse Model of Malignant Pleural Effusion Am. J. Respir. Cell Mol. Biol., February 1, 2006; 34(2): 142 - 150. [Abstract] [Full Text] [PDF]

				M. C. Cheung, L. Pantanowitz, and B. J. Dezube AIDS-Related Malignancies: Emerging Challenges in the Era of Highly Active Antiretroviral Therapy Oncologist, June 1, 2005; 10(6): 412 - 426. [Abstract] [Full Text] [PDF]

				M. Herida, M. Mary-Krause, R. Kaphan, J. Cadranel, I. Poizot-Martin, C. Rabaud, N. Plaisance, H. Tissot-Dupont, F. Boue, J.-M. Lang, et al. Incidence of Non-AIDS-Defining Cancers Before and During the Highly Active Antiretroviral Therapy Era in a Cohort of Human Immunodeficiency Virus-Infected Patients J. Clin. Oncol., September 15, 2003; 21(18): 3447 - 3453. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (71) Request Permissions Disclaimer Google Scholar Articles by Gallagher, B. Articles by Fordyce, E. J. Search for Related Content PubMed PubMed Citation Articles by Gallagher, B. Articles by Fordyce, E. J. Social Bookmarking          What's this?

Online ISSN 1476-6256 - Print ISSN 0002-9262

Copyright © 2009 Johns Hopkins Bloomberg School of Public Health

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics