AIDS RESEARCH AND HUMAN RETROVIRUSES Volume 18, Number 16, 2002, pp. 1175– 1179 © Mary Ann Liebert, Inc. Short Communication Rapid Disease Progression in Human Immunodeficiency Virus Type 1-Infected Seroconverters in India SANJAY M. MEHENDALE, 1 ROBERT C. BOLLINGER, 2 SMITA S. KULKARNI, 1 REBECCA Y. STALLINGS, 2 RONALD S. BROOKMEYER, 2 SANGITA V. KULKARNI, 1 ANAND D. DIVEKAR, 1 RAMAN R. GANGAKHEDKAR, 1 SMITA N. JOSHI, 1 ARUN R. RISBUD, 1 MADHURI A. THAKAR, 1 BHARATI A. MAHAJAN, 1 VARSHA A. KALE, 1 MANISHA V. GHATE, 1 DEEPAK A. GADKARI, 3 THOMAS C. QUINN, 4 and RAMESH S. PARANJAPE 1 ABSTRACT To determine if the early immunological and virological events of HIV infection are unique in a setting with limited access to health care and HIV-1 subtype C infection, we undertook a prospective cohort study to char- acterize the early natural history of HIV viral load and CD4 1 T lymphocyte counts in individuals with re- cent HIV seroconversion in India. CD4 1 T lymphocyte counts were prospectively measured for up to 720 days in 46 antiviral drug-naive persons with very early HIV infection, documented by HIV antibody seroconver- sion. HIV viral RNA levels were measured subsequently on reposited plasma samples from these same time points. The median viral load “ set point” for Indian seroconverters was 28,729 RNA copies/ ml. The median CD41 cell count following acute primary HIV infection was 644 cells/ mm 3. Over the first 2 years since pri- mary infection, the annual rate of increase in HIV viral load was 18274 RNA copies/ml/year and the annual decline in CD4 cell count was 2120 cells/year. Although the viral “ set point” was similar, the median trajec- tory of increasing viral load in Indian seroconverters was greater than what has been reported in untreated HIV seroconverters in the United States. These data suggest that the more rapid HIV disease progression de- scribed in resource-poor settings may be due to very early virological and host events following primary HIV infection. A rapid increase in viral load within the first 2 years after primary infection may have to be con- sidered when applying treatment guidelines for antiretroviral therapy and opportunistic infection prophy- laxis. 1175 OVER 90% OF HIV-INFECTED INDIVIDUALS live in countries of Africa and Asia, where the natural history of HIV in- fection is reported to be more rapidly progressive than in the United States and Europe. 1,2 Possible explanations suggested include lack of access to health care, 3 coinfection with other pathogens,4 malnutrition, 5 and other endemic infections before development of severe immunodeficiency. 6 A large number of studies from the United States and Eu- rope have demonstrated that plasma HIV RNA copies (viral load) measurements and CD4 counts are highly predictive of clinical progression to AIDS 7,8 and the rate of clinical pro- gression may depend upon early postinfection immunological and virological events. 9–11 Little is known about the early changes in CD4 1 T lymphocyte counts and viral load due to nonsubtype B HIV-1 infection. This study describes early changes in HIV viral load and CD4 1 T lymphocyte counts in individuals with recent primary HIV infection, most commonly affected with HIV-1 subtype C, and with very limited access to antiretroviral therapy. Between 1995 and 2000, following informed consent, pa- 1National AIDS Research Institute, Pune, India. 2Johns Hopkins University, Baltimore, Maryland. 3National Institute of Virology, Pune, India. 4National Institute of Allergy and Infectious Diseases, NIH, Bethesda, Maryland. tients attending two STD clinics in Pune, India and a Women’ s Health Clinic in the “ red-light” area of Pune were offered sero- logical screening for HIV-1 and HIV-2 infection. Individuals found to be HIV seronegative were invited to enroll in a cohort study. Study procedures, the incidence rates, and risk factors for HIV seroconversion for this cohort have been previously re- ported.12 At initial screening and follow-up, all patients were treated with standard therapy for STDs based on clinical im- pression, using Centers for Disease Control and World Health Organization guidelines. 13 All HIV-infected and HIV-unin- fected individuals were provided intensive risk-reduction pretest and posttest counseling at each visit. HIV-infected in- dividuals were encouraged to return for follow-up every 6 months or whenever they required medical attention and were provided access to the highest standard of locally available clin- ical care, as well as follow-up and referral services. Highly ac- tive antiretroviral therapy (HAART) is not yet a standard of care in India. Specimens that were positive by both the ELISA and the rapid test for HIV-1 and/or HIV-2 were confirmed by respec- tive HIV-1 or HIV-2 Western blot assay. From HIV serocon- verters (HIV-uninfected at screening who became HIV-infected during the follow-up) who agreed to continued prospective fol- low-up, blood was drawn at each visit. CD4 1 cell counts were estimated on the whole blood sample and plasma at each of these visits was stored frozen at 270°C. CD41 T lymphocyte percentages were measured, at each visit, from freshly collected blood samples by standard two-color fluorescent antibody com- binations using a flow cytometer. Ethylene diamine tetraacetic acid (EDTA)-treated plasma specimens, stored at 270°C, were used for quantitation of HIV-1 RNA performed by reverse tran- scription polymerase chain reaction (RT-PCR). All viral load assays were performed in duplicate. The date of primary HIV infection was estimated as the mid- point between the last HIV antibody-negativ e date and the first HIV antibody-positive date. The days since primary infection were calculated as the difference between this estimated infec- tion date and the date of sample/ data collection. For 4 of the 47 study participants, primary HIV infection was identified by the presence of p24 antigen prior to detection of anti-HIV an- tibody. For these individuals, the date of infection was esti- mated to be the mid-point of the p24 antigen-positive/ V an- tibody-negative date and the prior p24 antigen-negative/ V antibody-negative test date. For estimating the HIV viral load “ set point” and corre- sponding CD4 1 cell count, only data collected more than 90 days since the estimated infection date were included to limit the impact of data collected during the very acute primary in- fection period. Median values of the viral load and log 10 viral load in the window of 91 to 720 days after seroconversion were considered as “ set points.” Results on 47 individuals (10 females and 37 males) with documented HIV antibody seroconversion who agreed to return for regular follow-up and had samples sufficient for CD4 1 cell counts and/or viral load measurements are reported. They were derived from a total of 244 documented seroconverters attend- ing these three clinics from 1995 through 2000. Characteristics at the baseline screening visit and at the time of first positive HIV antibody test were compared between participants and non- participants. The two groups were comparable relative to age, gender, occupation, marital status, and risk group. However, the nonparticipating HIV seroconverters were less likely to have received any formal education and more likely to report fever that has been previously reported to be associated with acute HIV infection. The median age of the participants was 28 years. Of the 47 participants, 31 (65.9%) had education below high school level and nearly half were unmarried. Only eight (17.02%) reported sex work as a profession. There were no significant gender differences in initial me- dian CD4 counts (677/ mm 3 in women and 625/ mm 3 in men, respectively) or median viral loads (23,530 RNA copies/ml in women and 29,396 RNA copies/ml in men, respectively). The median time of the first viral load test ( .90 days since esti- mated time of infection and up to 720 days) for participants was 178 days (mean 5226 days) and the corresponding me- dian time for CD4 count estimation was 182 days (mean 5 213 days). Table 1 shows the median and interquartile ranges for the first postseroconversio n observations of HIV RNA viral load and CD4 1 cell counts. The median viral load of the ini- tial observation (viral load setpoint) following HIV serocon- version was 28,729 RNA copies/ ml of plasma (interquartile range 6,362– 133,661). The initial median CD4 1 lymphocyte count was 644 cells/ mm 3 (interquartile range 431– 782). We also analyzed the data by narrowing the window of time since infection to 91– 360 days and observed that the median viral load (29,092 copies/ml), log 10 viral load (4.46), and median CD4 count (631/mm 3) did not change substantially. However, between the two periods of 91– 360 days and 361– 720 days, the median viral loads showed an increase from 29,902 copies/ml to 33,803 copies/ml and median CD4 counts dropped from 631/mm3 to 497/mm3. Estimated slopes for HIV viral load and CD4 cell counts were computed in individual simple linear regression models for persons with two or more valid data points falling within the 91- to 720-day period postestimated day of infection. These MEHENDALE ET AL. 1176 TABLE 1. VIRAL LOAD SETPOINT AND CD4 CELL COUNT ESTIMATES IN DOCUMENTED SEROCONVE RTERS IN PUNE, INDIA (1995– 2000) 25th 75th Parameter N Median percentile percentile HIV RNA viral load a (copies/ml) 46 28,729 6362 133,661 Log10 HIV RNA viral load a (copies/ml) 46 4.46 3.80 5.13 CD4 cell count a (cells/mm3) 42 644 431 782 aFirst observation .90 days postestimated date of infection. individual slope estimates were combined to derive the de- scriptive statistics as shown in Table 2, which shows the cal- culated median change (slopes) in viral load and CD4 counts. The mean number of observations available for viral load and CD4 estimations was 3.4 and 3.2, respectively, with a median of 2 for both. The median annual change in HIV viral load was 10.17 log10 copies/ ml/ year. Based on slopes derived from raw units, the median annual increase was 8,274 RNA copies/ ml/year. The median decrease in CD4 cell count was 120 cells/year. There were two extreme values in this data set. One case had an initial viral load of 2,762,376 and another had a viral load slope 24,579,870. Analysis done by excluding these cases greatly affected the means and standard errors, but not the me- dians. We have therefore used median values for interpretation of the central tendency. A previous report on the natural history of early HIV infec- tion in U.S. seroconverters has confirmed the importance of early HIV viral load set points and early viral load trajectory in predicting the risk of progression to AIDS. 11 Our study pre- sents the first such data from HIV seroconverters in India, a country with an estimated HIV prevalence in excess of 3.8 mil- lion persons. In this study in Pune, India, we observed a me- dian HIV viral load “ set point” of 28,729 RNA copies/ml and a corresponding median CD4 count of 644 cells/mm 3 in indi- viduals with recent HIV infection. These values in Indian se- roconverters were similar to what have been described in un- treated HIV-infected seroconverters from the U.S. Multicentric AIDS Cohort study (MAC study). The median viral load and CD4 T lymphocytes count 6 months after the first HIV anti- body-positive visit in U.S. seroconverters were 33,759 RNA copies/ml and 599 cells/ mm 3, respectively. 11 The comparable time points since infection for viral load and first CD4 1 mea- surements were 7 months in our Indian study. Studies of viral load and CD4 count from countries other than the United States, Australia, and European countries are limited. A recent study from Uganda reported a median plasma viral load of 31,000 RNA copies/ml in subjects infected with HIV for less than 6 years compared with 87,000 RNA copies/ml for persons in- fected for greater than 6 years 14 and a lower median viral load of 15,000 RNA copies/ml among asymptomatic patients (WHO stage I) when compared with 65,000 RNA copies/ml among AIDS patients (WHO stage 4). There is evidence suggesting that HIV disease progression is more rapid in HIV-infected persons from resource-poor set- tings15 such as Sub-Saharan Africa and Thailand, 16,17 which could be due to lack of access to optimal health care. HIV-in- fected Africans in London have been shown to have no differ- ence in crude survival compared with non-African patients, when adjusted for CD4 count, disease stage, and year of diag- nosis.3 Low socioeconomic status and limited access to care have been associated with poorer survival in Canada. 18 Con- versely, in some settings with a better general health care in- frastructure, such as South Africa, untreated HIV patients with severe immunodeficiency have a median survival similar to un- treated patients in the United States. 19 HIV-associated survival and disease progression in resource-poor settings are reported to be associated with common endemic disease, such as tuber- culosis and bacterial infections, coinfection with helminthes or other endemic microorganisms, 20–24 immune activation leading to more efficient HIV viral replication, 25 and malnutrition or micronutrient deficiency. 5,26 Although the initial viral load set points and CD4 counts in this study in India were similar to those reported for U.S. se- roconverters, the early trajectories of viral load and CD4 count changes in this study in India are suggestive of a more rapid early disease progression in Indian seroconverters. The median early increase in plasma viral load of 10.17 log10 RNA copies/ml/year ( 18274 RNA copies/ ml/ year) observed in our study was higher than what was reported (median of 10.029 log10 copies/ ml/ year) in the U.S. seroconverters. 11 In fact, the overall viral load trajectory for the Indian seroconverters was similar to the median change of 10.18 log10 RNA copies/ ml/year seen in the subgroup of most rapid progressors ( ,3 years to AIDS) in the U.S. study that represented only 11% of the U.S. cohort of seroconverters. The median decline of CD4 count in Indian seroconverters was 2120 cells/ year and was slightly more rapid than that observed in the U.S. study (de- cline of 109 CD4 cells/year). The overall median loss of CD4 cells in the Indian study is between the median for the most rapid U.S. progressors ( 2249 cells/ year) and for U.S. serocon- verters who progressed to AIDS within 3– 7 years ( 289 cells/ year). These data suggest that the initial control of acute viremia may be similar in patients with primary HIV infection in India and the United States, but more rapid increase in viral load in the subsequent early infection period may contribute to a more rapid clinical disease progression. While it would be useful to statistically compare the median estimated slopes for plasma HIV viral load and CD4 cell count obtained in our Indian sam- ple against those reported in the MAC study, it is not compu- tationally possible to do so without the relevant raw data from the MAC Study. Previous studies have demonstrated an asso- ciation of symptomatic HIV infection with a more rapid dis- ease progression in U.S. patients. 27 As nonparticipating sero- converts were more likely to report symptoms of acute infection RAPID PROGRESSION IN INDIAN HIV SEROCONVERTERS 1177 TABLE 2. VIRAL LOAD AND CD4 CELL COUNT SLOPESa PER YEAR IN DOCUMENTED SEROCONVE RTERS IN PUNE, INDIA (1995– 2000) 25th 75th Parameter N Median percentile percentile HIV RNA viral load (copies/ml) 34 8274 222,212 78,380 Log10 HIV RNA viral load (copies/ ml) 34 0.17 20.60 0.88 CD4 cell count a (cells/mm3) 34 2120 2501 234 aBased on simple linear regression models with two or more observations 91– 720 days postestimated date of infection. than our cohort participants, the rapid increase in early viral load in this cohort might underestimate the actual trajectory of the HIV viral load increase in the STD patients. However, it is possible that higher viral loads seen in our study patients could be partly due to selection bias arising from self-referral due to the presence of some unmeasured clinical conditions. The ex- tent and rate of CD4 decline need to be assessed in the context of clinical outcomes in carefully designed prospective studies with larger sample sizes. It is also important to assess if changes in early plasma viral load and CD4 counts are HIV-1 subtype C specific and are influenced by potentially treatable host fac- tors. Some of the participants had initial measurements made far from their estimated date of infection and although it would have been ideal to analyze the two groups separately, we pre- ferred not to do this due to small sample size. Finally, important issues are emerging with improved access to antiretroviral therapy for HIV-infected patients in resource- poor settings. Even in the United States, recommendations about when to initiate HAART therapy have recently been mod- ified to support delaying HAART until significant immune dys- function occurs. 28 In addition to the tremendous infrastructure and training investments required for the safe and effective use of these expensive and potentially toxic drugs, additional stud- ies of the natural history of changes in CD4 counts and HIV viral load are critically needed to characterize unique virolog- ical and immunological characteristics of HIV infection that may be present in India. A rapid increase in viral load and CD4 decline within the first 2 years of primary infection may have important implications in decisions related to initiation of HAART and opportunistic infection prophylaxis in India and similar settings. ACKNOWLEDGMENTS We would like to thank Dr. Mrs. Phadke, Dr. Naik, Dr. Sule, Dr. Tolat, and Dr. Dharmadhikari from the B. J. Medical Col- lege and Dr. Ravetkar and Dr. Jadhav from the Health Depart- ment of the Pune Municipal Corporation for their guidance, co- operation, and help in setting up our study clinics. We would like to gratefully acknowledge the contribution of the whole HIVNET and Pathogenesis of Acute HIV Infection Project teams in their valuable contribution in data and specimen col- lection, laboratory procedures, and data management. This work was supported by National Institute of Allergy and Infectious Diseases, National Institutes of Health NIAID (AI 33879-02), NIH-NCRR OPD-GCRC (5M01RR00722), the NIH-Fogarty International Center (D43TW0000), and intramural research grants from the Indian Council of Medical Research. This arti- cle does not necessarily represent the position of the NIH or the ICMR. REFERENCES 1. Nunn AJ, Mulder DW, Kamali A, Ruberantwari A, Kengeya- Kayondo FJ, and Whitworth J: Mortality associated with HIV-1 in- fection over five years in a rural Ugandan population: Cohort study. BMJ 1997;315:767– 771. 2. Kitayaporn D, Tansuphaswadikul S, Lohsomboon P, Pannachet K, Kaewkungwal J, Limpakarnjanarat K, et al.: Survival of AIDS pa- tients in the emerging epidemic in Bangkok, Thailand. J Acquir Immune Defic Syndr Hum Retrovirol 1996;11:77– 82. 3. Del Amo J, Petruckevitch A, Phillips A, Johnson AM, Stephenson J, Desmond N, et al.: Disease progression and survival in HIV-1- infected Africans in London. AIDS 1998;121:1203– 1209. 4. Lucas SB, Odida M, and Wabinga H: The pathology of severe mor- bidity and mortality caused by HIV infection in Africa. AIDS 1991;5(Suppl 1):S143– 148. 5. Camp WL, Allen S, Alvarez JO, Jolly PE, Weiss HL, Phillips JF, et al.: Serum retinol and HIV-1 RNA viral load in rapid and slow progressors. J Acquir Immune Defic Syndr Hum Retrovirol 1998;18:401– 406. 6. Gilks CF: The clinical challenge of the HIV epidemic in the de- veloping world. Lancet 1993;324:1037– 1039. 7. Vlahov D, Graham N, Hoover D, Flynn C, Bartlett JG, Margolick JB, et al.: Prognostic indicators for AIDS and infectious disease death in HIV-infected injection drug users: Plasma viral load and CD41 cell count. JAMA 1998;279:35– 40. 8. Anastos K, Kalish LA, Hessol N, Weiser B, Melnick S, Burns D, et al.: The relative value of CD4 cell count and quantitative HIV- 1 RNA in predicting survival in HIV-1-infected women: Results of the women’ s interagency HIV study. AIDS 1999;13:1717– 1726. 9. Mellors JW, Munoz A, Giorgi JV, Margolick JB, Tassoni CJ, Gupta P, et al.: Plasma viral load and CD4 1 lymphocytes as prognostic markers of HIV-1 infection [see comments]. Ann Intern Med 1997;126:946– 954. 10. Mellors JW, Rinaldo CR, Jr., Gupta P, White RM, Todd JA, and Kingsley LA: Prognosis in HIV-1 infection predicted by the quan- tity of virus in plasma. Science 1996;272:1167– 1170. 11. Lyles RH, Munoz A, Yamashita TE, Basmi H, Detels R, Rinaldo CR, et al.: Natural history of human immunodeficiency virus type 1 viremia after seroconversion and proximal to AIDS in a large co- hort of homosexual men. J Infect Dis 2000;181:872– 880. 12. Mehendale SM, Rodrigues JJ, Brookmeyer RS, Gangakhedkar RR, Divekar AD, Gokhale MR, et al.: Incidence and predictors of hu- man immunodeficiency virus type 1 seroconversion in patients at- tending sexually transmitted disease clinics in India. J Infect Dis 1995;172:1486– 1491. 13. Centers for Disease Control and Prevention: Sexually transmitted diseases treatment guidelines. MMWR 1993;42:19– 95. 14. Morgan D, Rutebemberwa A, Malamba S, Ross A, Whitworth J, Kaleebu P, et al.: HIV-1 RNA levels in an African population- based cohort and their relation to CD4 lymphocyte counts and World Health Organization clinical staging. J Acquir Immune Defic Syndr 1999;22:167– 173. 15. Boerma JT, Nunn AJ, and Whitworth JA: Mortality impact of the AIDS epidemic: Evidence from community studies in less devel- oped countries. AIDS 1998;12[Suppl 1]:S3– 14. 16. Morgan D, Maude GH, Malamba SS, Okongo MJ, Wagner HU, Mulder DW, et al.: HIV-1 disease progression and AIDS-defining disorders in rural Uganda. Lancet 1997;350:245– 250. 17. Okongo M, Morgan D, Mayanja B, Ross A, and Whitworth J: Causes of death in a rural, population-based human immunodefi- ciency virus type 1 (HIV-1) natural history cohort in Uganda. Int J Epidemiol 1998;27:698– 702. 18. Hogg RS, Strathdee SA, Craib KJP, O’ Shaughnessy MV, Mon- taner JSG, and Schechter MT: Lower socioeconomic status and shorter survival following HIV infection. Lancet 1994;344: 1120–1124. 19. Maartens G, Wood R, O’ Keefe E, and Byrne C: Independent epi- demics of heterosexual and homosexual HIV infection in South Africa— survival differences. Q J Med 1997;90:449– 454. 20. Conlon CP, Pinching AJ, Perera CU, Moody A, Luo NP, and Lu- cas SB: HIV-related enteropathy in Zambia: A clinical, microbio- MEHENDALE ET AL. 1178 logical, and histological study. Am J Trop Med Hyg 1990;42: 83–88. 21. Gopinath R, Ostrowski M, Justement SJ, Fauci AS, and Nutman TB: Filarial infection enhances susceptibility to HIV infection. J Infect Dis 2000;182:1804– 1808. 22. De Cock KM, Soro B, Coulibaly IM, Lucas SB: Tuberculosis and HIV infection in sub-Saharan Africa. [Review]. JAMA 1992;268: 1581–1587. 23. Kalinkovich A, Weisman Z, Leng Q, Borkow G, Stein M, Green- berg Z, et al.: Increased CCR5 expression with decreased beta chemokine secretion in Ethiopians: Relevance to AIDS in Africa. J Hum Virol 1999;2:283– 289. 24. Martin DJ, Sim JG, Sole GJ, Rymer L, Shalekoff S, van Niekerk AB, et al.: CD41lymphocyte count in African patients co-infected with HIV and tuberculosis [see comments]. J Acquir Immune Defic Syndr Hum Retrovirol 1995;8:386– 391. 25. Bentwich Z, Kalinkovich A, and Weisman Z: Immune activation is a dominant factor in the pathogenesis of African AIDS. Immunol Today 1995;16:187– 191. 26. Baum MK, Shor-Posner G, Lai S, Zhang G, Lai H, Fletcher MA, et al.: High risk of HIV-related mortality is associated with sele- nium deficiency. J Acquir Immune Defic Syndr Hum Retrovirol 1997;15:370– 374. 27. Hogg RS, Strathdee SA, Craib KJP, O’ Shaughnessy MV, Mon- taner JSG, and Schechter MT: Lower socioeconomic status and shorter survival following HIV infection. Lancet 1994;344: 1120–1124. 28. Panel on Clinical Practices for Treatment of HIV Infection: Guide- lines for the Use of Antiretroviral Agents in HIV-Infected Adults and Adolescents. Department of Health and Human Services and the Henry J. Kaiser Family Foundation, February 5, 2001. Address reprint requests to: S. M. Mehendale National AIDS Research Institute, Pune, India Plot No 73 “ G” Block MIDC Bhosari, Pune 411026 India E-mail: hivnet@vsnl.com or sanjaymehendale@ya hoo.com RAPID PROGRESSION IN INDIAN HIV SEROCONVERTERS 1179