AIDS and cancer registry linkage: measurement and enhancement of registry completeness. The National AIDS/Cancer Match Study Group

Human Papillomavirus Infection and Cervical Cancer in HIV+ Women

Human papillomavirus (HPV) is the first identified necessary cause of human cancers and is associated with nearly 100% of all cervical cancers. Compared to the general female populations, HIV+ women have higher prevalence and incidence of cervical HPV infections, higher risks of persistent HPV infections and subsequent cervical intraepithelial lesions, and a higher incidence of cervical cancer. Although the wide use of combined antiretroviral therapy (cART) has improved the immune function and the longevity of HIV+ women, the incidence of cervical cancer in HIV+ women has not declined. For HIV+ women who follow routine cervical cancer screenings, their incidence of cervical cancer is comparable to that in HIV-negative women. Thus, adherence to the recommended cervical cancer screening is still critical for HIV+ women to prevent cervical cancer. Prophylactic HPV vaccines may also benefit HIV+ women, but prospective studies are needed to determine the effectiveness of HPV vaccination on reducing cervical cancer incidence in HIV+ women.

Sociodemographic differences in linkage error: an examination of four large-scale datasets

BACKGROUND

Record linkage is an important tool for epidemiologists and health planners. Record linkage studies will generally contain some level of residual record linkage error, where individual records are either incorrectly marked as belonging to the same individual, or incorrectly marked as belonging to separate individuals. A key question is whether errors in linkage quality are distributed evenly throughout the population, or whether certain subgroups will exhibit higher rates of error. Previous investigations of this issue have typically compared linked and un-linked records, which can conflate bias caused by record linkage error, with bias caused by missing records (data capture errors).

METHODS

Four large administrative datasets were individually de-duplicated, with results compared to an available 'gold-standard' benchmark, allowing us to avoid methodological issues with comparing linked and un-linked records. Results were compared by gender, age, geographic remoteness (major cities, regional or remote) and socioeconomic status.

RESULTS

Results varied between datasets, and by sociodemographic characteristic. The most consistent findings were worse linkage quality for younger individuals (seen in all four datasets) and worse linkage quality for those living in remote areas (seen in three of four datasets). The linkage quality within sociodemographic categories varied between datasets, with the associations with linkage error reversed across different datasets due to quirks of the specific data collection mechanisms and data sharing practices.

CONCLUSIONS

These results suggest caution should be taken both when linking younger individuals and those in remote areas, and when analysing linked data from these subgroups. Further research is required to determine the ramifications of worse linkage quality in these subpopulations on research outcomes.

Changing pattern of lymphoma subgroups at a tertiary academic complex in a high-prevalence HIV setting: a South African perspective

BACKGROUND

HIV infection has been associated with an increased risk of non-Hodgkin lymphoma, particularly in the first world. Despite the high burden of HIV infection in sub-Saharan regions, published data on HIV and malignancies are sparse from these areas.

MATERIALS AND METHODS

We recently published data on lymphomas diagnosed from January 2004 to December 2006, at a single center in Johannesburg, to serve as a baseline for long-term comparison during the period of highly active antiretroviral therapy rollout. We report a retrospective analysis of the follow-up data collected from January 2007 to December 2009 at the Johannesburg academic hospital complex (Gauteng, South Africa).

RESULTS

There were 2225 new diagnoses of lymphoproliferative disorders made during 2007-2009 as compared with 1897 cases diagnosed during 2004-2006. A significant increase in both high-grade B-cell lymphomas and Hodgkin lymphoma was documented during 2007-2009. This was associated with a statistically significant increase in HIV prevalence in those tested (from 44.3% in 2004-2006 to 62.0% in 2007-2009). HIV-positive patients presented at a statistically significantly younger median age and showed a relative overrepresentation of females when compared with HIV-negative patients. HIV-positive patients were diagnosed at later stages of HIV infection when compared with patients in the first world.

CONCLUSIONS

The pattern of lymphoma subtypes and the demographics of the patients diagnosed have altered in association with significantly increased HIV prevalence. These changes have important public health implications. In particular, scale-up and earlier access to highly active antiretroviral therapy is essential with continued monitoring as access to therapy improves.

Data linkage: a powerful research tool with potential problems

BACKGROUND

Policy makers, clinicians and researchers are demonstrating increasing interest in using data linked from multiple sources to support measurement of clinical performance and patient health outcomes. However, the utility of data linkage may be compromised by sub-optimal or incomplete linkage, leading to systematic bias. In this study, we synthesize the evidence identifying participant or population characteristics that can influence the validity and completeness of data linkage and may be associated with systematic bias in reported outcomes.

METHODS

A narrative review, using structured search methods was undertaken. Key words "data linkage" and Mesh term "medical record linkage" were applied to Medline, EMBASE and CINAHL databases between 1991 and 2007. Abstract inclusion criteria were; the article attempted an empirical evaluation of methodological issues relating to data linkage and reported on patient characteristics, the study design included analysis of matched versus unmatched records, and the report was in English. Included articles were grouped thematically according to patient characteristics that were compared between matched and unmatched records.

RESULTS

The search identified 1810 articles of which 33 (1.8%) met inclusion criteria. There was marked heterogeneity in study methods and factors investigated. Characteristics that were unevenly distributed among matched and unmatched records were; age (72% of studies), sex (50% of studies), race (64% of studies), geographical/hospital site (93% of studies), socio-economic status (82% of studies) and health status (72% of studies).

CONCLUSION

A number of relevant patient or population factors may be associated with incomplete data linkage resulting in systematic bias in reported clinical outcomes. Readers should consider these factors in interpreting the reported results of data linkage studies.

AIDS-related Burkitt lymphoma in the United States: what do age and CD4 lymphocyte patterns tell us about etiology and/or biology?

Trimodal or bimodal age-specific incidence rates for Burkitt lymphoma (BL) were observed in the United States general population, but the role of immunosuppression could not be excluded. Incidence rates, rate ratios, and 95% confidence intervals for BL and other non-Hodgkin lymphoma (NHL), by age and CD4 lymphocyte count categories, were estimated using Poisson regression models using data from the United States HIV/AIDS Cancer Match study (1980-2005). BL incidence was 22 cases per 100 000 person-years and 586 for non-BL NHL. Adjusted BL incidence rate ratio among males was 1.6× that among females and among non-Hispanic blacks, 0.4× that among non-Hispanic whites, but unrelated to HIV-transmission category. Non-BL NHL incidence increased from childhood to adulthood; in contrast, 2 age-specific incidence peaks during the pediatric and adult/geriatric years were observed for BL. Non-BL NHL incidence rose steadily with decreasing CD4 lymphocyte counts; in contrast, BL incidence was lowest among people with ≤ 50 CD4 lymphocytes/μL versus those with ≥ 250 CD4 lymphocytes/μL (incidence rate ratio 0.3 [95% confidence interval = 0.2-0.6]). The bimodal peaks for BL, in contrast to non-BL NHL, suggest effects of noncumulative risk factors at different ages. Underascertainment or biological reasons may account for BL deficit at low CD4 lymphocyte counts.

Population-based linkage of AIDS and cancer registries: importance of linkage algorithm

BACKGROUND

This study evaluates linkage algorithms used in 1997 for Los Angeles County by the AIDS-Cancer Match Registry to estimate the risk of cancer among people with HIV/AIDS.

METHODS

In 2001, a linkage between the Los Angeles County cancer and AIDS registries using a five-pass algorithm was compared with the 1997 linkage results that used a two-pass algorithm.

RESULTS

The five-pass linkage detected all of the previously identified matches (6571) as well as another 431 (6.2%), which had been missed.

CONCLUSIONS

Record linkage methodologic decisions can markedly affect matching sensitivity and specificity. It is probable that the AIDS-Cancer Match Registry two-pass linkage has underestimated the risks of some cancers among people with AIDS.

How valid is using cancer registries' data to identify acquired immunodeficiency syndrome-related non-Hodgkin's lymphoma?

OBJECTIVE

We sought to determine the accuracy of cancer registry data regarding the human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS) status of patients with non-Hodgkin's lymphoma (NHL).

METHODS

We used the population-based San Diego/Orange County cancer registry to identify 392 patients with HIV-related NHL diagnosed 1994-1999. After matching for age, sex, race, period of NHL diagnosis, and hospital type, we were able to find 324 corresponding patients among the remaining 4,863 NHL patients diagnosed 1994-1999 (who did not have HIV infection according to cancer registry records). We sought to review these patients' charts at 41 hospitals with 15 separate institutional review boards to determine if the HIV serostatus from the cancer registry was correct. We performed a forward conditional multivariate logistic regression to determine characteristics associated with a false positive HIV status.

RESULTS

The false positive rate was 8% while the false negative rate was 3%. The positive predictive value was 93% while the negative predictive value was 97%. Compared to correctly identified patients, false positives were more likely to be > or =50 years old, female, and treated with chemotherapy and less likely to be single with high grade or extranodal disease.

CONCLUSION

Using cancer registry data to identify AIDS-related NHL is a valid research practice.

Changes in acquired immunodeficiency syndrome-related non-Hodgkin lymphoma in the era of highly active antiretroviral therapy: incidence, presentation, treatment, and survival

BACKGROUND

The authors sought to determine whether the availability of highly active antiretroviral therapy (HAART) coincided with changes in the epidemiology of acquired immunodeficiency syndrome (AIDS)-related non-Hodgkin lymphoma (NHL).

METHODS

Cancer registry data from 1988-2000 were linked with AIDS registry data from 1981 to July 2003 for San Diego County to identify 537 AIDS-NHL patients. By using the total number of patients with AIDS who were alive as of July 1 annually as the AIDS population denominator, the average annual incidence of NHL was estimated among patients with AIDS for the pre-HAART period (1988-1995) and post-HAART period (1996-2000). The chi-square test was used to compare proportions, and a Cox proportional hazards model was used to compare survival between the pre-HAART and post-HAART periods.

RESULTS

The incidence of NHL decreased from 29.6 per 1000 person-years pre-HAART to 6.5 per 1000 person-years post-HAART. The proportion of patients who had NHL of central nervous system (CNS) origin decreased from 28% pre-HAART to 17% post-HAART. Among patients with systemic NHL, 54% received chemotherapy pre-HAART, and 72% received chemotherapy post-HAART. The percentage of intermediate-grade NHL increased from 33% pre-HAART to 49% post-HAART, and the percentage of high-grade NHL decreased from 38% to 19%, respectively. A diagnosis of human immunodeficiency virus infection preceding the NHL diagnosis and Stage IV NHL were associated with worse survival, whereas a diagnosis of NHL in the post-HAART period and chemotherapy were associated with better survival. The median survival was 4 months pre-HAART and 9 months post-HAART.

CONCLUSIONS

Since the introduction of HAART, there has been a decrease in the incidence of systemic and CNS NHL among patients with AIDS. Among patients with systemic, AIDS-related NHL, there has been decreased high-grade histology, increased use of chemotherapy, and improved survival.

Population-Based Surveillance of HIV-Associated Cancers: Utility of Cancer Registry Data

Long-term cancer risks are uncertain in HIV-infected persons, particularly those using highly active antiretroviral therapy (HAART). Timely, population-based surveillance of HIV-associated malignancies in the United States has been challenging because of various data inadequacies. Cancer registries represent a resource for this surveillance, if uncertainties around accurate differentiation of HIV-associated and unassociated cancers can be resolved. To inform the utility of cancer registry data for classifying and monitoring HIV-associated cancers, the completeness and quality of cancer registry-available information about patient HIV status was assessed. For all 10,126 non-Hodgkin lymphomas (NHLs), 1497 Hodgkin lymphomas (HLs), and 895 anal cancers reported to the Greater San Francisco Bay Area registry during 1990-1998, 6 indicators of patient HIV status were retrieved from 2 cancer registry-available sources (cancer registry records, death records) and from linkage with the California AIDS registry. Cross-tabulations were used to examine the distributions of patients with evidence of positive HIV status by indicator and source. Together, 5 cancer registry-available HIV indicators identified 25% more presumed HIV-positive NHL patients and nearly 50% more HL and anal cancer patients than were detected by AIDS registry linkage. Eighty-three percent of NHL patients and at least half of HL and anal cancer patients were identified by multiple sources of HIV indicators, and most individual indicators agreed acceptably with others. However, optimal strategies for classifying HIV-associated patients differed by cancer site. At least in this region, cancer registry data represent a useful resource for monitoring HIV-associated lymphomas and anal cancer and may offer benefits over linkage-based means in the age of HAART.

Characteristics and survival of patients with non-Hodgkin's lymphoma with and without acquired immunodeficiency syndrome

Our objective was to determine the characteristics and survival of patients with non-Hodgkin's lymphoma (NHL) with and without acquired immunodeficiency syndrome (AIDS). A cancer registry and AIDS registry linkage for San Diego County was performed in October 1998 as part of a national multicentre study. We performed Kaplan-Meier analysis to compare survival in NHL patients with and without AIDS, after matching for age, sex, and race/ethnicity. We performed logistic regression to determine which patient and tumour characteristics were significantly associated with 1-year survival. Of the 4361 cases of NHL, 324 (7%) had AIDS and 4037 (93%) were not known to have AIDS. Patients with AIDS were more likely to have extranodal, high-grade, and disseminated NHL diagnosed by non-histologic means and were less likely to have received chemotherapy. Patients with AIDS and NHL who survived at least 1 year had less advanced disease stage and received chemotherapy. The median survival in patients with AIDS was 4 months (95% confidence interval (CI): 4-5) and 95 months (95% CI: 58-157) in patients without AIDS (P<0.001). Although these patients with AIDS-related NHL were unlikely to survive, the highly active antiretroviral agents currently used may improve outcomes in future patients.

Acquired immune deficiency syndrome and the increase in non-Hodgkin lymphoma incidence in New Jersey from 1979 to 1996

BACKGROUND

In New Jersey, the age-adjusted incidence rate of non-Hodgkin lymphoma (NHL) increased more than a third from 1979 to 1996, the largest increase among the major cancers.

METHODS

Data from a linkage of New Jersey's population-based cancer and acquired immune deficiency syndrome (AIDS) registries were used to obtain two sets of annual age specific incidence rates and estimated average annual percentage changes in the incidence rates, for each of five adult age groups within each gender, from Poisson regression models that 1) included all the NHL cases and 2) excluded the cases of AIDS-NHL.

RESULTS

During 1979-1996, of the NHL cases aged 15 years and older reported to the cancer registry, 687 (6%) of the 11,725 male cases and 139 (1%) of the 10,785 female cases were AIDS-NHL. The highest percentages of AIDS-NHL were in the younger age groups--15-29, 30-39, and 40-49 years. Among both men and women, average annual percentage increases in NHL occurred overall (3.1 and 3.0, respectively), and in each age group, ranging from 1.6 and 1.9, respectively, in the 50-59 years age group to 6.6 and 4.2, respectively, in the 30-39 years age group (P <<0.01). Excluding AIDS-NHL, the estimated average annual percentage increases in NHL were greatest in the 30-39 and the 60 years and older age groups among men, and these two age groups plus the 15-29 years age group among women, ranging between 2.4 and 2.9 (P < 0.05).

CONCLUSIONS

As elsewhere, factors in addition to AIDS are involved in the increasing incidence of NHL in New Jersey. Because diagnostic and classification changes probably do not explain the entire increase unrelated to AIDS, other risk factors are likely responsible. Public health interventions to reduce the incidence of NHL not related to AIDS are problematic until more is known about the causes of NHL.

Methodology used for "software for automated linkage in Italy" (SALI)

Linkage of epidemiological registries can provide cost-effective information on the associations between different diseases or exposures in the population under study and on completeness of surveillance system databases. We describe the program SALI (software for automated linkage in Italy) aimed at matching individual records from medium-sized registries (in the order of 100,000 records), where the desired outcome is to miss as few links as possible and, because of low link-likelihood (< 1%), a manual revision of matched pairs is feasible. SALI, developed in CA-Clipper language, uses registry files in dBase format. It requires only name, surname, and date of birth as key fields, and it allows for spelling errors in Italian or other Latin languages through a specific algorithm. Furthermore, a double-blind procedure ensures data confidentiality. The main linkage procedure is based on four stages, two automatic ones, and two where the operator can decide through specific windows whether to accept stage-selected matches. SALI takes into account possible errors in key fields thus reducing false negatives. It was used to solve the problem of linkage between AIDS and cancer registries in Italy. It can be used with every IBM-compatible computer system, assuring uniquely high portability.

AIDS oncology

Kaposi's sarcoma, non-Hodgkin's lymphoma, Hodgkin's disease, and squamous cell carcinoma are among the malignancies seen with increased frequency in patients infected with HIV. The outlook for patients with these malignancies has improved significantly with the utilization of highly active antiretroviral therapy (HAART) and more aggressive cytotoxic therapies. Novel biologic therapies with lesser side effects are currently being evaluated. This article reviews the current knowledge about HIV malignancies, their epidemiology, pathogenesis, clinical manifestations, and treatment.

HIV-related malignancies: community-based study using linkage of cancer registry and HIV registry data

For people immunosuppressed by human immunodeficiency virus (HIV), we expect an increase in cancer incidence similar to that documented in transplant patients. We examined the cancer spectrum in an HIV-infected cohort, specifically malignancies not currently associated with acquired immunodeficiency syndrome (AIDS), in relation to the general population. Cancer incidence data for residents of Harris County, Texas, diagnosed between 1975 and 1994, were linked to HIV/AIDS registry data by Soundex code and date of birth to identify malignancies in an HIV-infected cohort of 14,986 persons. Incidence of cancer in this cohort was compared to the general population by standardized incidence ratio (SIR) analysis. From the HIV-infected cohort, 2289 persons (15%) were identified as having one or more malignancies, with 97% occurring in males. The linkage alone identified 29.5% of the malignancies, of which only 28.7% were diagnosed in males. Adjusting for age, HIV-infected men and women had incidences of cancer that were 16.7 [95% confidence interval (CI) 16.1-17.3] and 2.9 (95% CI 2.3-3.7) times that expected for the general population of Harris County, Texas. Besides Kaposi's sarcoma, non-Hodgkin's lymphoma, cervix cancer and brain lymphoma, non-AIDS related malignancies of Hodgkin's lymphoma, non-melanotic skin cancer in males and colon cancer in females, exhibited significant SIRs of 5.6 (95% CI 3.6-8.4), 6.9 (95% CI 4.8-9.5) and 4.0 (95% CI 1.1-10.2). Increased incidences of lung, prostate and breast malignancies were not seen in this HIV cohort. Persons infected with HIV appear to be at increased risk for the non-AIDS related malignancies, Hodgkin's lymphoma, non-melanotic skin cancer in males and colon cancer in females.

Creating a population-based linked health database: a new resource for health services research

As the availability of both health utilization and outcome information becomes increasingly important to health care researchers and policy makers, the ability to link person-specific health data becomes a critical objective. The integration of population-based administrative health databases has been realized in British Columbia by constructing an historical file of all persons registered with the health care system, and by probabilistically linking various program files to this 'coordinating' file. The linkages have achieved a high rate of success in matching service events to person-specific registration records. This success has allowed research projects to be proposed which would otherwise not have been feasible, and has initiated the development of policies and procedures regarding research access to linked data. These policies and procedures include a framework for addressing the ethical issues surrounding data linkage. With continued attention to confidentiality issues, these linked data present a valuable resource for health services research and planning.

Association of non-acquired immunodeficiency syndrome-defining cancers with human immunodeficiency virus infection

Kaposi's sarcoma and non-Hodgkin's lymphoma were among the earliest recognized manifestations of the acquired immunodeficiency syndrome (AIDS) epidemic. Excluding these two tumors, the overall risk of all other cancers in human immunodeficiency virus (HIV)-infected individuals is similar to that of the general population. However, varying levels of evidence link several additional neoplasms to HIV infection. The evidence is strongest for an association with Hodgkin's disease, with lower relative and absolute risks than for non-Hodgkin's lymphoma. Anogenital intraepithelial neoplasia also appears to be HIV associated, but increases of invasive disease are still uncertain for both cervical and anal cancers. Various studies have suggested associations with testicular seminoma, multiple myeloma, oral cancer, and melanoma, but the data are inconsistent. Leiomyosarcoma and benign leiomyomas have increased in incidence in HIV-infected children but are unusual in HIV-infected adults. Conjunctival carcinoma is seen in HIV-infected individuals in sub-Saharan Africa but it is uncommon in Western countries. Most other cancers do not seem to have increased incidences in HIV infection. The etiologic mechanisms of HIV-related cancer likely differ among these diverse cancers and do not globally increase cancer risk.

Overview of the epidemiology of immunodeficiency-associated cancers

Immunodeficiency, be it congenital, therapeutic, or infectious in origin, increases the risk of certain, but not all, types of cancer. A common feature of these cancers is that specific infectious agents appear to be important in their etiology, not only in immunodeficient subjects but also in the general population. People with acquired immunodeficiency syndrome (AIDS) are at an increased risk of Kaposi's sarcoma, non-Hodgkin's lymphoma, Hodgkin's disease, squamous cell carcinoma of the conjunctiva, and childhood leiomyosarcoma. It is striking that most of these cancers have been associated with specific human herpesvirus (HHV) infections: HHV-8 with Kaposi's sarcoma and the closely related Epstein-Barr virus with non-Hodgkin's lymphoma, Hodgkin's disease, and possibly also with childhood leiomyosarcoma. Moreover, similar associations between these viruses and cancer have been found, albeit inconsistently, in people who are not immunosuppressed. Further research is needed to establish whether the risk of other cancers is also increased in people with AIDS, although, if so, the cancers are likely to be rare or to have comparatively small associated relative risks. Existing evidence suggests that there may be no marked increase in the risk of two common cancers that are known to be caused by infectious agents--hepatocellular carcinoma and invasive carcinoma of the uterine cervix. The apparent lack of an increase in invasive cervical cancer is unexpected and needs further investigation, especially since the prevalence of cervical infection with human papillomaviruses and of low-grade preneoplastic changes in the cervical epithelium is increased in women with AIDS. With the prospect of improved survival in people with AIDS, the effect of immunosuppression on cancer is likely to become an increasingly important issue.

Spectrum of AIDS-associated malignant disorders

BACKGROUND

To clarify which types of cancer result from AIDS, we compared the cancer experiences of people with AIDS with those of the general population by matching population-based cancer and AIDS registries in the USA and Puerto Rico.

METHODS

We used a probabilistic matching algorithm to compare names, birth dates, and, where available, social-security numbers of 98,336 people with AIDS and 1,125,098 people with cancer aged less than 70 years. We defined AIDS-related cancers as those with both significantly raised incidence post-AIDS and increasing prevalence from 5 years pre-AIDS to 2 years post-AIDS.

FINDINGS

Among people with AIDS, we found 7028 cases of Kaposi's sarcoma (KS), 1793 of non-Hodgkin lymphoma (NHL), and 712 other cases of histologically defined cancer. Incidence rates among people with AIDS were increased 310-fold for KS, 113-fold for NHL, and 1.9-fold (95% CI 1.5-2.3) for other cancers. Of 38 malignant disorders other than KS and NHL, only angiosarcoma (36.7-fold), Hodgkin's disease (7.6-fold), multiple myeloma (4.5-fold), brain cancer (3.5-fold), and seminoma (2.9-fold) were raised and increasing significantly (p<0.02) from the pre-AIDS to the post-AIDS period.

INTERPRETATION

Interpretation is complicated by screening and shared risk factors, such as sexual behaviour and cigarette smoking. However, our data indicate that AIDS leads to a significantly increased risk of Hodgkin's disease, multiple myeloma, brain cancer, and seminoma. Immunological failure to control herpes or other viral infections may contribute to these malignant diseases.

Linkage of AIDS and cancer registries in Italy

We report the first results of a comparison between the Italian Registry on AIDS (RAIDS) and 13 population-based cancer registries (about 8 million population in 1991) with respect to the notification of Kaposi's sarcoma and non-Hodgkin's lymphoma. Routine indicators of data quality and completeness have been found in both types of registry, consistent with the best international standards. A linkage process was carried out on about 339,000 cancer notifications and 3,134 AIDS notifications and was herein restricted to individuals under the age of 50. Out of 243 Kaposi's sarcomas at either type of registry, 90 (37%) were reported as such by both; 68% of individuals with Kaposi's sarcoma at cancer registries could be identified at the AIDS registry, including AIDS-defining illnesses other than Kaposi's sarcoma; 62% of individuals with Kaposi's sarcoma at RAIDS could be found at cancer registries. Of 2,104 non-Hodgkin's lymphomas at either type of registry, 55 were reported as such by both; 65% of individuals reported as having non-Hodgkin's lymphoma at the AIDS registry were found at cancer registries. Our present results indicate the scope for improving cancer assessment in individuals with HIV infection and AIDS and the potential of AIDS and cancer registries for such a purpose.

Acquired immunodeficiency syndrome-associated Kaposi's sarcoma. Biology and management

Kaposi's Sarcoma (KS), the most common AIDS-associated malignancy, occurs with increased frequency in all HIV transmission groups, but at a particularly high rate in homosexual men. Recent studies suggest that KS pathogenesis involves exposure to an infectious agent, altered expression and response to cytokines, and modulation of growth by HIV gene products. KS varies in its clinical presentation from a relatively indolent process to a widely disseminated, aggressive disease. A variety of local and systemic treatments provide effective, but usually temporary, disease palliation. Insights into KS pathogenesis suggest a number of targeted therapeutic approaches that may eventually lead to improved disease management and disease cure.

Kaposi's sarcoma and non-Hodgkin's lymphoma following the diagnosis of AIDS. Multistate AIDS/Cancer Match Study Group

We linked records of 83,434 AIDS cases reported to AIDS registries through 1990 to cancer registry records during times when overlap in registration existed. Of 8,496 Kaposi's-sarcoma (KS) cases meeting enrollment criteria, 1,045 occurred semesters 2 through 4 (6 through 23 months) after another AIDS-defining illness. KS risk in this period after AIDS declined steadily over the 1980s. Adjusting for age, gender, ethnic origin and calendar time period, we found the relative risk (RR) of KS to be 106,000 for homo/bisexual men with AIDS and 13,000 for other men with AIDS. Risk was highest for homo/bisexual men between 30 and 39 years old and among 20- to 29-year-old non-homo/bisexual men. The RR in black men was approximately half that reported in white men among homo-bisexual men and others. Between the second and fourth semester after AIDS, the ratio of RR of KS to that of non-AIDS-related cancers increased 1.5-fold. In a similar analysis, there were 335 non-Hodgkin's lymphoma (NHL) cases in semesters 2 through 4 after AIDS. The overall risk was elevated 283-fold in homo/bisexual men and the RR ratio increased 1.8-fold between semester 2 and 4 after AIDS. In summary, the risk of KS following another AIDS-defining illness is strikingly high, more in white men than in black men, and the risks of KS and, especially, NHL appear to increase with time from AIDS.

The epidemiology of AIDS--related neoplasms

The magnitude for and risk factors of the two most important AIDS neoplasm, Kaposi's sarcoma and non-Hodgkin's lymphoma, are reviewed in detail. The association between AIDS and other cancers is mostly speculative because surveillance biases tend to favor detecting associations that may be spurious. The overall relative risk of other cancers appears, however, to be only twofold above that in the general population, with associations being most convincing for anal (but not cervical) cancer and leiomyosarcoma and possible also for Hodgkin's disease, testicular cancer, and conjunctival cancers.

Problems in interpreting HIV sentinel seroprevalence studies

Estimating human immunodeficiency virus (HIV) prevalence from sentinel seroprevalence studies is difficult. We characterize these studies and show that most are investigations of incompletely defined (hypothetical) cohorts and are usually based on nonprobability samples. Prevalence in HIV sentinel serosurveys is also time-averaged and vulnerable to several time-dependent sources of bias (e.g., migration, deaths, and changes in incidence). Assumptions must be made that these time-dependent biases did not meaningfully affect the data, and this can be helped by reducing the period of investigation. Furthermore, we show that "reliability" can not be adequately measured by standard error, that "internal validity" is vulnerable to self-selection bias and laboratory problems, and that "generalizability" is limited. We propose that what is needed is a procedure (like formal metaanalysis methods) incorporating information from several separate HIV sentinel seroprevalence studies, in a manner that is reproducible and can take into consideration the differences between studies.