Division Gynecologic Oncology, Department of Obstetrics & Gynecology, Nelson R Mandela School of Medicine, Durban, South Africa
Address correspondence and reprint requests to: M. Moodley, MBChB, FCOG, MMed (O/G), 1 Laing Place, Escombe, Queensburgh, Durban 4093, South Africa. Email: firstname.lastname@example.org
Accepted for publication November 22, 2005
Although the human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS) is a global phenomenon, the bulk of the disease occurs in developing regions of the world such as sub-Saharan Africa(1). By previous estimates, up to 40% of patients with AIDS will develop some type of malignancy(2). Women who are infected with HIV have a higher rate of developing infections of human papillomavirus (HPV), especially of the high-risk types. Such women are at greater risk of developing squamous intraepithelial neoplasia and invasive cervical cancers(3). In areas ravaged by the HIV, the occurrence of female genital tract malignancies in association with the HIV has become a reality, creating new challenges for clinicians. The prevalence of HPV in HIV-infected women and in women not infected with HIV has been shown to be 65% and 29%, respectively(4). However, a more recent report documented the prevalence of HPV DNA in 98% of HIV-infected women(5). Women infected with HIV with severe immunosuppression are five times more likely than women not infected with HIV to have lower genital tract neoplasias(6).
Some studies have supported the concept that there is an increased prevalence of invasive cervical cancer among HIV-infected women(7–9), whereas other studies failed to establish this relationship(10,11). The prevalence of squamous intraepithelial lesions was reported to be 18.8% versus 5.3% among HIV-infected women compared with among women not infected with HIV(4). The risk of cervical cancer is also reported to be excessive relative risk (RR, 3) among persons with AIDS, with most of the excess risk resulting from sexually acquired HPV(9). A further study documented a prevalence of 7% of nearly 400 HIV-infected women with high-grade cervical intraepithelial neoplasia as compared to only 1% of 307 control subjects not infected with HIV(6). International HIV seroprevalence rates among women with cervical cancer vary from 1.6% in Hong Kong(12), 15% in Kenyan women(13) to 21% reported in our previous study from Durban, South Africa(14). In 2003, it was our clinical impression that there was a reduction in the number of women presenting with invasive cervical cancer compared to previous years. The aim of this study was therefore to compare demographic data of women with invasive cervical cancer who presented to our center in 1999 and 2003 and provide reasons for any changes in prevalence of cervical cancer.
Materials and methods
A retrospective study was performed from January to December 2003 among women with invasive cervical cancer managed by the combined gynecological oncologic clinic of Inkosi Albert Luthuli Central hospital, Durban, South Africa. This is a large tertiary hospital serving mainly the black indigent population of the province of KwaZulu-Natal. All data were obtained from the gynecological oncologic records. After the diagnosis was histologically confirmed, all patients tested for HIV infection had pretest and posttest counseling. Data collected included total number of patients, age, residential address, stage of disease, histologic type, and CD4 counts. Standard investigations performed included chest X-ray, ultrasound of the abdomen, complete blood count, and renal function.
These data were then analyzed and compared to similar data obtained from a previous study performed by the gynecology oncology unit in 1999 to establish if there was a difference in demographic data between the two time periods. The data were further analyzed in terms of the HIV/AIDS-related mortality figures from the Actuarial Society of South Africa (ASSA 2000) to establish if there was a trend of HIV mortality/HIV prevalence compared to the prevalence of cervical cancer for the two time periods. Statistical analysis for the comparison of various parameters was performed using the Fisher's test.
In 1999, there were a total of 672 patients with invasive cervical cancer of whom 138 tested positive for the HIV, giving a seroprevalence rate of 21%. In 2003, a total of 271 clinical records of women with cervical cancer were evaluated of whom 45 women were infected with HIV, giving a seroprevalence of 21.8%. The HIV seroprevalence between the 2 years was not statistically significant (P = 0.376). The corresponding HIV seroprevalence among antenatal attendees for 1999 and 2003 were 32.5% and 38.7%, respectively. The mean ages of HIV non-infected women compared to HIV-infected women for 1999 and 2003 were 55.8 and 39.8 years compared to 54 and 41 years, respectively. The differences in mean age of presentation between women not infected with HIV and HIV-infected women in 1999 compared to similar data in 2003 were 15 and 13 years, respectively. In both time periods, the majority of women not infected with HIV were in the 50–60 year age group compared to the HIV-infected women who were in the 30–40year age group. When the “non-black” women were compared as a single group, a significantly lower incidence of HIV infection and higher proportion of early-stage disease was found in this group relative to black women (P = 0.001). Residence was evaluated in terms of whether women resided in urban or rural areas. In 1999, there were 119 women not infected with HIV compared to 52 HIV-infected women residing in urban areas. Likewise in 2003, there were 92 HIV non-infected and 20 HIV-infected women residing in urban areas. The differences were statistically significant (P = 0.012). In contrast, there were 401 women not infected with HIV and 85 HIV-infected women in 1999 residing in rural areas compared to 69 women not infected with HIV and 23 HIV-infected women in 2003 from rural areas (P = 0.108). Most women not infected with HIV were from rural areas (P < 0.0001). The number of HIV-infected women residing in urban and rural areas declined in the 2003 period, in keeping with a general reduction in the total number of cancer cases seen.
A comparison of data for the two time periods is shown in Table 1.
Squamous cell carcinoma was the commonest histologic type for women not infected with HIV and HIV-infected women in both time periods (1999 and 2003: 89% versus 85%; P = 0.679). Majority of women in the 1999 period had poorly differentiated tumors, irrespective of HIV status, compared to moderately differentiated tumors in the 2003 period. The majority of women in the 1999 period (92%) presented with late-stage disease compared to 70% in the 2003 period. In the 1999 period, the proportion of women with early-stage disease varied from 12% to 6% in the HIV-infected group and group not infected with HIV, respectively (P = 0.018). A similar relationship was recorded in the 2003 period, where 13.6% of HIV-infected women and 5.6% of women not infected with HIV presented with early-stage disease (FIGO stage ≤IIA; P = 0.09). In both time periods, since majority of women had advanced-stage disease, CD4 counts were not routinely performed. However, there were 29 women in the 2003 period who were referred with CD4 counts, of whom three had early-stage disease. There were six women with CD4 counts below 200 cell/μL, all of whom had late-stage disease. Blood counts and renal function parameters for both time periods showed no significant differences.
Worldwide, invasive cervical cancer remains a major cause of morbidity and mortality among women, especially in developing countries. Cervical cancer screening and uptake thereof are either inadequate or nonexistent in such developing areas. This is in comparison to the developed countries especially Nordic areas where cervical cancer screening has reduced the prevalence and mortality of cervical cancer by as much as 80%(15). It is fairly clearly established that the HPV is causally linked to the pathogenesis of cervical cancer. High-risk HPV types (16 and 18) are present in almost 100% of cervical cancers and preinvasive lesions(16). The early genes (E6 and E7) of the HPV are capable of immortalization and oncogenic transformation and play a role in the initiation and oncogenic progression of tumors. Women infected with HIV are more likely to harbor multiple HPV genotypes. On a molecular basis, there is a close link between the HPV and the HIV(17). Although an increase in the prevalence of cervical cancer has been reported by some studies, this has not been firmly established. In the United States, it has been demonstrated that the incidence of invasive cervical cancer is uncommon in HIV-infected women participating in a regular screening program(18). In Kenyan women, the prevalence of HIV among cervical cancer patients was 15% compared to controls although women younger than 35 years were 2.6 times more likely to be HIV positive than controls of similar age (35% versus 17%)(13).
A comparison of the data for the two time periods in this study revealed similar findings with regards to HIV seroprevalence among women with cervical cancer, mean ages, histologic types, disease stage, and areas of residence. This reveals that the basic profile of women had not changed between the two time periods.
The prevalence of HIV infection established by way of screening antenatal attendees in South Africa has revealed that the province of KwaZulu-Natal has had the highest prevalence of HIV infection relative to other provinces over the past decade. A report from the province of Gauteng showed a HIV seroprevalence of 7.2% among women with cervical cancer in 1999, with a background HIV prevalence of 23.9%(19). In contrast, we previously reported a HIV seroprevalence of 21% among women with cervical cancer in 2001, for a background HIV prevalence of 32.5%(14). Our current finding of HIV seroprevalence of 21.8% among women with cervical cancer is in keeping with our previous report. While it might be expected that there should have been a greater number of HIV-infected women with invasive cervical cancers associated with a background rise in HIV prevalence, this was not evident from this study. Reasons for this are unclear except that there were an overall lower number of women in the study. Over the past 2 years, the oncology services had relocated to Inkosi Albert Luthuli Central Hospital, which is a new tertiary hospital for the province. However, all gynecology oncology patients in the province were referred to Inkosi Albert Luthuli Central Hospital. The reduction in number of women with invasive cervical cancer in our province is not a chance finding as there were 280 women with invasive cervical cancer who were referred to our center in 2004. Further, since there has been no change in the cervical cancer screening program for the province of KwaZulu-Natal, screening has had no impact on the reduction in the number of cervical cancer cases noted. In 1999 period, 14.5% and 4% of HIV-infected women and women not infected with HIV reported having Pap smears performed in their lifetime. This data were not recorded for the 2003 period, and therefore, no comparison can be made.
Despite the higher HIV prevalence for the province in 2003 (38.5%) compared to 1999 data (32.5%), there has been no corresponding increase in prevalence of HIV infection among women with invasive cervical cancer. Projections of HIV seroprevalence by the ASSA 2002 among antenatal attendees show fairly close correlation with actual data(20). The data from the ASSA 2000 clearly show a rise of HIV-infected women and therefore mortality with time(20). Statistics regarding total population, total number of HIV infections, and AIDS mortality are shown in Tables 2 and 3 and show a similar relationship. Our deduction from the available evidence is that with a rise in background HIV seroprevalence, there has been a dramatic rise in HIV-related mortality before women reach the age to manifest invasive cervical cancer. Although South Africa has embarked on an antiretroviral “rollout” program, which is available in most provinces, the availability and accessibility of antiretrovirals are still in its infancy stages. All women, irrespective of HIV status, from various areas of our province usually visit local clinics or district hospitals and are then referred to tertiary hospitals should the need arise. We are not aware of any different referral patterns among HIV-infected women and cannot therefore postulate on any differences in health-seeking behavior specifically among HIV-infected women. In our environment, most women with cervical cancer present with late-stage disease, as was described in our previous study(14). It therefore appears to be purely coincidental that in this study more women not infected with HIV presented with advanced-stage disease.
In African women, therefore, an increase in cervical cancer among HIV-infected women is not seen probably because of their rapid demise prior to developing cervical cancer, a relatively slow neoplastic process(6). Although the Centers for Disease Control expanded the case definition to include cervical cancer as an AIDS-defining illness in 1993, in the South African situation, over timeframes, there has been a high background HIV seroprevalence as well as high incidence of invasive cervical cancer cases(14), and therefore, it is most unlikely that cervical cancer among HIV-infected women reflect an AIDS-defining condition in an African setting. Paradoxically, as can be seen in our setting, the number of women presenting with invasive cervical cancer has drastically declined over time as the HIV epidemic spiraled out of control and supporting evidence indicates competing mortalities from HIV/AIDS-related illnesses.
In conclusion, this is the first report demonstrating an inverse relationship between the HIV/AIDS prevalence/mortality and the prevalence of invasive cervical cancer over two different time periods for the same area.
1 Ateka GK. HIV/AIDS. Global impact and human rights—a Southern African perspective. AIDS Scan
2 Smith C, Lilly S, Mann KP et al.
AIDS-related malignancies. Ann Med
3 Sun XW, Kuhn L, Ellerbrock TV et al.
Human papillomavirus virus infection in women with human immunodeficiency virus infection. Arch Pediatr Adolesc Med
4 Duerr A, Kieke B, Warren D et al.
Human papillomavirus associated cervical cytologic abnormalities among women with or at risk of infection with human immunodeficiency virus. Am J Obstet Gynecol
5 Levi JE, Kleter B, Quint WG et al.
High prevalence of human papillomavirus infections and high frequency of multiple HPV genotypes in human immunodeficiency virus-infected women in Brazil. J Clin Microbiol
6 Ferenczy A, Coutle F, Franco E, Hankins C. Human papillomavirus and HIV coinfection and the risk of neoplasias of the lower genital tract: a review of recent developments. CMAJ
7 Royansky N, Anteby SO. Gynecologic neoplasia in the patient with HIV infection. Obstet Gynecol Surv
8 Franceschi S, Dal Maso L, Pezzotti P et al.
Incidence of AIDS-defining cancers after AIDS diagnosis among people with AIDS in Italy, 1986–1998. J Acquir Immune Defic Syndr
9 Goedert JJ. The epidemiology of acquired immunodeficiency syndrome malignancies. Semin Oncol
10 Newton R, Ziegler J, Beral V et al.
A case-control study of human immunodeficiency virus infection and cancer in adults and children residing in Kampala, Uganda. Int J Cancer
11 Mbulaiteye SM, Parkin DM, Rabkin CS. Epidemiology of AIDS-related malignancies: an international perspective. Hematol Oncol Clin North Am
12 Chan YM, Ng TY, Ngan HY, Wong LC. Screening for HIV infection in women with newly diagnosed cervical cancer. Gynecol Oncol
13 Gichangi PB, Bwayo J, Estambale B et al.
Impact of HIV infection on invasive cervical in Kenyan women. AIDS
14 Moodley M, Moodley J, Kleinschmidt I. Invasive cervical cancer and human immunodeficiency virus (HIV) infection: a South African perspective. Int J Gynecol Cancer
15 Aareleid T, Pukkala E, Thomson H, Hakma M. Cervical cancer incidence and mortality trends in Finland and Estonia: a screened vs unscreened population. Eur J Cancer
16 Bekkers RL, Massuger LF, Bulten J, Melchera WJ. Epidemiological and clinical aspects of human papillomavirus detection in the prevention of cervical cancer. Rev Med Virol
17 Moodley M. Update on pathophysiologic mechanisms of human papillomavirus. Curr Opin Obstet Gynecol
18 Massad LS, Seaberg EC, Watts DH et al.
Low incidence of invasive cervical cancer among HIV-infected US women in a prevention program. AIDS
19 Lomalisa P, Smith T, Guidozzi F. Human immunodeficiency virus infection and invasive cervical cancer in South Africa. Gynecol Oncol
Copyright © 2006 Blackwell Publishing Ltd.