The global epidemic of HIV-1 is the result of a cross-species infection of humans by a chimpanzee lentivirus, simian
immunodeficiency virus (SIV_cpz), which occurred in west central Africa.¹³¹ Chimpanzees, in turn, appear to have acquired SIV_cpz sometime after their divergence into multiple subspecies. Pan troglodytes and P.t. schweinfurthii are naturally infected, whereas other closely related species are not, suggesting that SIV_cpz is unlikely to have coevolved with its natural host.³⁶⁶ How the virus initially arose as a new virus in chimpanzees remains unclear, but it likely occurred following transmission of two other ancestral SIV species from monkeys,³⁶⁶ perhaps because chimpanzees hunt monkeys for food. Of note, SIV_cpz is not an asymptomatic infection in chimpanzees; the epidemiologic evidence documents decreased survival in chimpanzee troops infected by SIV_cpz.²⁰⁹ In contrast, other SIV infections appear to be largely asymptomatic in their natural monkey hosts (i.e., SIV_smm in sooty mangabeys) and do not lead to immune deficiency or decreased life span. Experimental transmission of these viruses to susceptible non-natural hosts can result in progressive and profound immunodeficiency and acquired immunodeficiency disease (AIDS).³⁵⁴ In fact, the transmission of SIV_smm in West Africa is the origin of the HIV-2 epidemic in humans.³⁵⁴ Although the reasons for differences in transmissibility and pathogenicity during cross-species transmission are not fully understood, it is clear that innate host defenses such as those imparted by apolipoprotein B messenger RNA (mRNA)–editing enzyme, catalytic polypeptide-like 3G (APOBEC3G), and tripartite motif 5 alpha (TRIM-5α) play a role and create barriers that viruses must overcome in order to become established and survive within a new host species (see Chapter 49).

At least three distinct cross-species transmissions from chimpanzees to humans are thought to have occurred, giving rise to three highly divergent genetic lineages of infection, termed groups M, N, and O.¹⁵⁸,³⁰²,³⁷² Group M is by far the most widespread, accounting for most cases globally. Because the greatest genetic diversity in group M is found in Kinshasa, Zaire, the epidemic likely began in this region of Africa. The approximate date of introduction of the M group into the human population is estimated to be around 1931, based on supercomputer analysis of large numbers of sequenced HIV isolates, assuming a constant rate of evolution.²¹⁸ This approach has been validated by successfully timing the earliest historic isolate, which was sequenced from a stored plasma sample obtained in 1959 from a person who died of AIDS in Manchester, England.⁷⁷

The most heavily affected continent at the present time is Africa, with an estimated 25 million persons currently infected with HIV, and at least 3 million with advanced disease in need of treatment. HIV infections have increased markedly over the last decade, despite knowledge of the mechanisms of transmission. Expanding epidemics in Eastern Europe, South Asia, and China may ultimately eclipse the African epidemic in terms of sheer numbers of infected persons. The viruses fueling these epidemics vary according to geographic region, with clade C
virus being the most prevalent worldwide and clade B being currently the most prevalent in the United States and Europe. Clades differ one from another by up to 35%; within a single clade variation between isolates can be as great as 20%. Mistakes in reverse transcription, which occur because the viral reverse transcriptase lacks an editing function, leads to variability of individual virus sequences within a single infected person that can exceed what is seen with other viruses (e.g., influenza) over the course of a global epidemic.¹³⁵

Most cases of HIV infection worldwide are the result of sexual transmission, but important modes of transmission also include parenteral transmission and transmission from mother to infant. The actual risk of transmission per exposure varies widely (Table 50.1). For sexual transmission, the risk of male to male transmission is greater than the risk of heterosexual transmission, and is highest in persons practicing receptive anal intercourse.³⁴⁸ Overall, the greatest risk of transmission is from contaminated blood transfusions, with 95 of 100 persons becoming infected, although transfusion-related infections are rare now that blood is routinely screened for this pathogen.³⁶¹ The rate of transmission depends on the inoculum size delivered, but perhaps also on local mucosal defense mechanisms that modulate infection but remain to be defined. Population studies performed in Africa suggest that there is a threshold viral load (∼1,500 RNA copies/mL plasma) below which the likelihood of sexual transmission is greatly reduced.¹⁵⁰,³²⁵

Mathematical modeling suggests that the published risks of sexual transmission, which come from longitudinal studies of persons in identified risk groups, are too low to account for the huge numbers of infections worldwide and for the rapid increase in infections that occur after puberty. In South Africa, for example, 4% of 16-year-old girls are HIV seropositive, whereas by age 21 this number increases to 31%.³⁰⁸ Coital frequency estimates in these same age groups fail to account for this rapid increase in infections. More recent data suggest that transmission is amplified during periods of increased genital tract viral shedding.¹²⁸ Increased shedding certainly occurs at the time of acute infection, which has been shown through a study of more than 15,000 persons in 56 villages in Uganda to be a particularly important period for transmission and, thus, has important public health implications.⁴¹¹ An important aspect of this study was that it followed couples who were both seronegative at the time of enrollment, so it was not potentially selecting for discordant couples who might have already demonstrated some intrinsic resistance to transmission. Other factors that may amplify the risk of transmission include other co-infections (e.g., malaria and tuberculosis, sexually transmitted diseases [STDs], or in particular co-transmission of STD and HIV).⁷¹ Estimated transmission probability likely also depends on genetic factors, such as density of the HIV co-receptor CCR5 on cervical cells.⁵⁸ In addition to using epidemiologic data, much has been learned about transmission across a mucosal barrier by studying the SIV system, where it has been shown that CD4 T cells are the first cells to be infected (as opposed to antigen-presenting cells), and that there is a small temporal window of opportunity in which antiviral therapy or immune responses can block viral dissemination.¹⁵⁹,⁴³³

Ever since the analysis of viral load in the Multicenter AIDS Cohort Study (MACS) dataset in 1996,²⁶⁷ it has been clear that the course of HIV infection can be predicted within the first 6 to 12 months based on plasma viral load, and that CD4 count at this time also independently helps to determine whose disease will progress rapidly and whose will not.²⁴⁶ The median viral load in plasma at the time of peak viremia during acute infection is approximately 10⁶ to 10⁷ RNA copies/mL, which drops to a mean set point of 30,000 copies/mL within the first 6 to 12 months of infection. Longitudinal studies from the MACS, which were initiated before the availability of effective antiviral therapy, revealed the interquartile ranges of viral load and the relationship to risk of disease progression (Fig. 50.2). Interestingly, at 1 year after infection, only a fivefold difference
was seen in plasma viremia between the quartile that does the best and the quartile that does the worst.²⁴⁶

Studies in both humans infected with HIV⁴³,²⁶⁴ and the SIV macaque model²³⁴,²⁶¹ provide important new insights into why the period of acute infection plays such a defining role in ultimate disease progression, and why the immune system ultimately fails to control HIV in most infected persons. Infection and depletion of large numbers of both resting and activated memory CD4 cells in the gut during primary infection may be what fuels the initial peak viremia, and likely places constraints on subsequent adaptive responses⁴³,²⁶⁴ (Fig. 50.3). Details of what is happening at a cellular level in acute infection have come from experimental infection of macaques with SIV. At the time of acute SIV infection, massive infection occurs of gut-associated lymphoid tissue, with between 30% and 60% of all gut-associated CD4 cells becoming productively infected, leading to massive depletion of these cells in only 4 days.²⁶¹ The memory cell population, including both activated and resting memory cells, seems most vulnerable.²³⁴ In the earliest stages of infection, more than half of all memory CD4+ T cells are lost, and the body is left to fight the prolonged chronic phase of infection with a severely constricted repertoire of CD4 T helper cells. These experiments in the SIV model are almost certainly representative of what is happening in acute HIV infection in humans, where the initial plasma viral load averages 1 to 10 million RNA copies/mL. Whether the subsequent decline to a steady state results from active inhibition by innate and adaptive immune responses to HIV, or a decrease in the number of cells available to support viral replication, remains an area of dispute.³¹²

Most of the CD4 T cells that are depleted from the gut during acute HIV or SIV infection are T helper cells that produce interleukin 17 (IL-17) (Th17 cells). These cells are crucial for maintaining the integrity of the mucosal barrier, and their depletion by HIV or SIV leads to loss of epithelial integrity within the gut mucosa, thereby allowing translocation of gut-associated microbial products into the systemic circulation.¹¹⁰ These products are a major driver of the systemic immune activation that is a predominant feature of all pathogenic HIV and SIV infections.⁴¹,⁴² The loss of CD4 T cells from the gut during acute infection therefore has two effects upon the immune system: it depletes a major portion of the immune reserve in the form of memory CD4 T cells, and it allows microbial translocation, which establishes a state of chronic immune activation that further promotes systemic HIV infection and replication.

During the period of acute infection, a stable reservoir of HIV-infected resting memory CD4 cells is established that harbors replication-competent, integrated provirus.⁶³,¹¹⁶ This occurs whether or not persons are started on treatment in the initial symptomatic stage of infection. Because proviruses that form this latent reservoir are not transcriptionally active, no viral proteins or enzymes are produced. These infected cells therefore are protected from the effects of antiviral drug therapy and from immune attack. This latent reservoir is present whether the viral load is high or low, and it persists even after prolonged highly active antiviral therapy.¹¹⁵ A rough estimate is that about 1 in 1 × 10⁶ CD4 cells harbors integrated latent provirus.⁴³⁴ The half-life of infected cells is such that it would take more than 70 years to achieve viral eradication by gradual senescence of these cells even with no blips in viremia on therapy,⁴³⁴ a situation that is yet to be achieved (see below).

The initial immune response to HIV involves innate immune mechanisms. Toll-like receptors (TLRs) on dendritic cells (DCs) are activated by a combination of HIV-1 RNA via TLR7 and microbial products (discussed earlier) via other TLRs, resulting in high levels of interferon alpha (IFN-α), IL-12, tumor necrosis factor alpha (TNF-α), and IL-6, which together result in a profound activation of the immune system.²³,⁴²,¹⁶⁹ These activated DCs can be infected with HIV, which may further contribute to impaired DC function in infected persons.²⁴⁵,³⁵⁵ Natural killer (NK) cells are also part of the initial innate immune response to HIV, and their numbers are greatly increased in acute infection. Because the increase in NK cells coincides with the initial decline in viremia after acute infection, these cells are thought to have a key role as an antiviral defense mechanism. Moreover, NK cells appear to be largely anergic in chronic infection, suggesting that a pool of nonfunctional NK cells accumulates over time.⁷ The fact that expression of a killer cell inhibitory receptor, KIR3DS1, and a particular sequence in its ligand human leukocyte antigen (HLA) Bw4, is associated with slower disease progression²⁵⁷ also argues for an important role of these cells in infected persons.

HIV infection is associated with the development of antibodies that appear within weeks of initial infection. Although most of these antibodies bind, but do not neutralize,³¹⁹ a subset of these can neutralize virus, either by binding directly to the envelope glycoprotein trimer on the surface of free virions or following CD4-gp120 binding, preventing the fusion of the viral and cell membranes that is essential for viral entry.⁴⁷ The envelope is heavily glycosylated, and these sugars prevent antibody binding to the underlying peptidic structure. These neutralizing antibody responses nevertheless are sufficiently strong to influence viral evolution; indeed, the virus population in vivo is rapidly replaced by mutant virus that escapes recognition.³³⁹,⁴¹² This selection is an iterative process: new antibody responses develop to neutralize the mutant virus, which then escapes again. These responses are primarily narrowly directed
and do not appear to be able to broadly cross-neutralize many strains of HIV. During the chronic phase of infection, there is little evidence to indicate an ongoing contribution of neutralizing antibodies to immune control, as shown by depletion of antibody-generating B cells in an animal model of AIDS virus infection.³⁵⁷ Having said this, it is clear that sera from chronically infected individuals can potently neutralize multiple strains of HIV broadly.²⁷,⁹⁹,²³⁵,³⁸²,⁴⁰⁹ Isolation of antibodies from these individuals has proven to be a rich source of monoclonal antibodies with broad and potent activity, and has led to the identification of potential targets of neutralizing antibodies for vaccine development.

The optimal production of antibodies and the function of CD8 T cells depends on the presence of virus-specific CD4 T cells.⁹² The magnitude of these responses, as measured by IFN-γ production, is consistently much lower than the CD8 T-cell responses, and a small number of peptides are recognized by most individuals.²⁵,³²⁶ In contrast to the wide spectrum of responses mediated by CD8 T cells, the adaptive CD4 T-cell response is primarily directed against the Gag protein, for reasons that are unclear. In the early stage of acute infection, a robust induction of HIV-specific CD4 T cells occurs, and these cells clearly provide help to CD8 T cells, as evidenced by the ability of CD8 T cells in acute infection to proliferate in vitro in response to cognate epitope recognition.²³⁶ The ability of CD4 and CD8 T cells to proliferate in vitro is lost over the first few months of untreated infection, but the acquired defect in CD8 T-cell proliferation can be restored in vitro by the addition of CD4 T cells obtained at the time of acute infection or by IL-2.¹⁸⁵,²³⁶ Although HIV-specific CD4 T-cell function has been augmented through therapeutic interventions (vaccination, treatment interruptions, and IL-2 therapy), none of these approaches has led to improved virologic or clinical outcomes.³,¹⁸⁵,²⁰⁷,³⁴⁰,³⁴⁴,³⁴⁶

CD4 T cells can be classified based on their major functions into multiple lineages, including T helper 1, T helper 2, T helper 17, follicular T helper, and T regulatory. Most HIV-specific CD4 T cells fall within the T helper 1 category based on their secretion of IFN-γ, but their overall function appears to be impaired.¹⁶¹,¹⁸⁵ The fact that HIV preferentially infects and depletes HIV-specific CD4 T cells may partially explain the low frequency of HIV-specific CD4 T cells and their inability to adequately coordinate humoral and CD8 T-cell responses to HIV.¹⁰² In addition, the phenotype and function of opportunistic pathogen-specific CD4 T cells may affect their susceptibility to infection and depletion by HIV, thereby explaining some of the variation in onset of different opportunistic infections.⁵²,¹³⁹ Specifically, CD4 T cells that express β chemokines (which block HIV entry by binding to CCR5) are more resistant to HIV infection and depletion than those that do not.⁵²,¹³⁹

One of the strongest predictors of disease progression is the HLA type of the host. The HLA class I alleles B*27 and B*57 are associated with low viral load and prolonged asymptomatic
infection,²⁰¹ although additional alleles have also been shown to affect viral load (Fig. 50.4). Other HLA alleles have been associated with more rapid disease progression, in particular a subtype of the HLA B35 allele referred to as HLA B35 px.¹³³ It is now clear that the HLA B alleles are dominant in terms of having an impact on viral load.²¹³ On a population level, clear evidence indicates viral imprinting by host CD8 T-cell responses: Persons with certain HLA alleles have a significantly increased prevalence of certain viral polymorphisms that appear to be driven by immune escape.²⁰⁸ Animal studies have shown clearly that the major histocompatibility complex (MHC) type can influence both disease progression and the relative contribution of cellular and humoral immune responses to viral control.²⁵¹ Immunization of humans with candidate HIV vaccines has also shown HLA-related differences in immunogenicity.²⁰² Remaining to be identified are the precise mechanisms whereby these alleles mediate a protective or detrimental effect, or enhance immunogenicity of candidate vaccines, or whether this represents a linkage to some other genetic factor that modulates outcome. In addition to HLA alleles, other genetic factors may also have an impact on disease progression, and are discussed below.

Acquisition of HIV-1 infection is accompanied by relatively nonspecific symptoms of an acute viral illness in approximately
50% to 70% of infected individuals.¹⁹⁸ Symptoms, which usually begin 2 weeks after exposure, frequently include fever, pharyngitis, headache, arthralgias, myalgias, malaise, and weight loss.⁸⁵,¹⁶⁷ A nonpruritic, maculopapular rash on the face and trunk occurs in up to 70% of cases.³⁹⁵ In addition, generalized lymphadenopathy is a frequent finding. Mucocutaneous ulceration and weight loss help distinguish primary HIV-1 infection from other viral syndromes.¹²⁵,¹⁶⁷ Aseptic meningoencephalitis is the most common neurologic manifestation of primary HIV-1 infection.³⁹⁵ Symptoms of primary infection resolve within 3 to 4 weeks in most patients. Persistence of symptoms beyond 8 to 12 weeks, along with a severely depressed CD4+ T-lymphocyte count and high plasma HIV-1 RNA levels may predict more rapid progression of disease.

Laboratory characteristics of primary HIV-1 infection include lymphopenia and a decrease in the absolute CD4+ T-lymphocyte count, usually accompanied by an increase in circulating activated CD8+ T cells.⁷⁵ Other hematologic abnormalities are unusual, except for mild thrombocytosis. Modest elevations in serum aspartate transaminase and alkaline phosphate may be present, but clinically significant hepatitis is uncommon.³⁹⁵ Plasma HIV-1 RNA titers generally peak 1 week following the onset of symptoms, averaging 10⁶ to 10⁷ copies/mL, and decline to steady-state levels (between 10³ and 10⁵ copies/mL) by 2 months after infection.²³⁷,²⁴⁶

Following resolution of primary infection and establishment of a virologic quasi-steady state,¹⁷⁵ a prolonged period of asymptomatic chronic infection ensues. Although most patients remain asymptomatic during much of this phase, ongoing viral replication and CD4 lymphocyte depletion make the term clinical latency inappropriate. The loss of CD4 T cells proceeds at an average rate of 30 to 60 cells/mm³/year, although CD4 cell counts can remain stable for several years before a period of rapid decline.²⁶⁶ A small proportion of patients (<1%) experience progression to AIDS within 1 to 2 years.²⁸⁰,³¹⁰ Rapid progression may be associated with transmission of syncytium-inducing (SI) variants of HIV-1 (i.e., viruses that use the CXCR4 coreceptor, also known as X4 viruses).⁶⁴,²²⁴,²⁸⁷

Fatigue and lymphadenopathy are noted by many patients during this otherwise asymptomatic phase of HIV-1 infection. Minor clinical events, such as oral hairy leukoplakia (caused by Epstein-Barr virus infection of oral epithelial cells), oral and vaginal candidiasis, herpes zoster, and a variety of other dermatologic conditions may be early signs of clinical progression. With advancing disease, night sweats and weight loss become more common.

A variety of systemic manifestations of HIV-1 infection involving nearly every organ system can occur during the chronic phase of disease. Remission in response to antiretroviral therapy suggests a direct role for HIV-1 in the pathogenesis of these disorders. Dermatologic conditions are particularly common among HIV-1-infected individuals, including seborrheic dermatitis, papular pruritic eruption, and eosinophilic folliculitis.⁷⁶ Neurologic manifestations include disorders of both the central nervous system (CNS) and peripheral nervous system, in addition to opportunistic infections and malignancies of the CNS. AIDS dementia complex and distal symmetric polyneuropathy, which occur in up to 27% and 35% of patients, respectively, are among the most frequently encountered HIV-related neurologic complications.³⁷³,³⁷⁷ Wasting associated with HIV infection remains highly prevalent in resource-limited settings and is an independent predictor of mortality.²⁵⁰ Anorexia, malabsorption, and inappropriate nutrient utilization all contribute to the loss of lean body mass in patients with AIDS.¹⁵⁴,²²⁰

Other organ system–specific complications of HIV infection include nonspecific interstitial pneumonitis and lymphocytic interstitial pneumonitis, disorders of the gastrointestinal tract, endocrine dysfunction, anemia and neutropenia, immune-mediated thrombocytopenia, and a variety of rheumatologic syndromes. In addition, HIV-associated nephropathy leading to renal insufficiency is particularly common among African Americans and injection drug users.³⁹⁰,⁴²²

Opportunistic infections and malignancies are rare in HIV-infected persons with CD4 counts above 500 cells/mm³, but increase as the CD4 count declines below this benchmark. Oral candidiasis, pneumococcal infections, tuberculosis, and reactivation of herpes simplex and varicella zoster viruses become more common. The risk of life-threatening complications, including Pneumocystis jirovecii (formerly, P. carinii) pneumonia, candida esophagitis, disseminated histoplasmosis and other systemic fungal infections, toxoplasma encephalitis, and cryptococcal meningitis increases substantially once the CD4 lymphocyte count drops below 200 cells/mm³.²⁵⁸ Opportunistic infections, such as disseminated Mycobacterium avium complex infection; reactivation of cytomegalovirus (CMV) infection, cryptosporidiosis, and microsporidiosis; and progressive multifocal leukoencephalopathy caused by JC virus reactivation, are all indicative of a profound defect in cellular immunity and usually occur at CD4 counts below 50 cells/mm³. Malignancies associated with AIDS generally are related to underlying viral infection, including Kaposi sarcoma (KS) caused by infection with human herpes virus 8; lymphomas associated with Epstein-Barr virus infection; and cervical and anal carcinoma associated with human papilloma virus infection. Although potent antiretroviral therapy has clearly reduced the risk of KS and non-Hodgkin lymphoma, the effect on Hodgkin lymphoma is less clear.³²,³³,³⁷⁰ Whereas the incidence of lung cancer is also increased in persons with HIV infection, age-adjusted rates of other malignancies (e.g., breast and prostate cancer) are comparable to those found in the general population.¹⁵²,³⁶⁹

A growing number of end-organ complications not traditionally considered “AIDS-defining” events have been recognized to occur more frequently in HIV-1–infected patients.²⁵³,²⁸⁶ These include an increased risk of cardiovascular disease, non-AIDS defining malignancies, HIV-associated neurologic dysfunction (HAND), and HIV-associated nephropathy (HIVAN), and may also include loss of bone mineral density (BMD) and other changes typically associated with increasing age, suggesting that HIV-1 infection may accelerate the aging process more generally.

With respect to cardiovascular risk, a number of studies suggest increased risk of myocardial infarction in HIV-1–infected patients as compared to control populations matched for traditional cardiovascular risk factors such as age, family history, hyperlipidemia, hypertension, diabetes, and smoking.¹⁷⁷,²⁸⁶,³⁹⁸ More striking is the finding that interrupting antiretroviral
therapy significantly increases the risk of cardiovascular events, independent of CD4 cell count.¹⁰⁶,³¹³ This increased risk was correlated with an increase in plasma levels of inflammatory markers such as IL-6 and high-specificity C-reactive protein (hsCRP).²²³,³⁴⁷ In addition, a number of studies have shown adverse changes in surrogate markers associated with cardiovascular risk such as flow-mediated vasodilatation, carotid intima-media thickness, and coronary artery calcification in patients with HIV-1 infection.⁸²,¹⁵³ The extent to which these changes can be prevented or reversed by antiretroviral therapy is an active area of current research.

Patients with HIV-1 infection also show lower BMD and an increased risk of fractures compared to age-matched, uninfected controls.³⁹⁷ The causes of HIV-associated loss of BMD are poorly understood and most likely are multifactorial, including increased immune activation and inflammation, renal tubular dysfunction, low vitamin D levels, and other endocrine abnormalities.¹⁰ Whether vitamin D levels should be monitored in all HIV-infected patients, and at what level to offer supplementation, remains an area of controversy.

Numerous viral and host factors contribute to determining the rate of HIV disease progression. The plasma HIV-1 RNA level, or viral load, reflects the rate of virus replication and is a powerful independent prognostic factor for the risk of disease progression.²⁶⁷,³⁰⁴ Data from the MACS show that in the absence of antiretroviral therapy individuals with plasma HIV-1, RNA levels greater than 100,000 copies/mL 6 months after seroconversion are 10 times more likely to progress to AIDS within 5 years than are those with lower steady-state levels of viremia.²⁶⁵ Similarly, for patients with established HIV-1 infection, a steady-state viral load of more than 30,000 copies/mL is associated with a more than fivefold greater risk of disease progression within 3 years as compared with patients with viral loads of 3,000 to 10,000 copies/mL.²⁶⁶

According to a Poisson regression model based on data from the Concerted Action on Seroconversion to AIDS and Death in Europe (CASCADE) collaboration of 22 cohorts from Europe, Canada, and Australia, in the absence of combination antiretroviral therapy the 6-month risk of AIDS for a 25-year old patient with a CD4 cell count of 350/mm³ ranges from 0.6% at a viral load of 3,000 copies/mL to 2.5% at a viral load of 300,000 copies/mL.³¹¹ At a CD4 count of 100 cells/mm³, the AIDS risk at the same viral loads increases to 3.7% and 14.5%, respectively. The risk of disease progression has been reduced substantially since the introduction of potent combination antiretroviral therapy.¹⁰⁵ Progression to AIDS among persons infected with HIV-2 proceeds at a significantly slower rate as compared with infection with HIV-1.¹⁸⁸,²⁶⁰ Epidemiologic and cohort studies show that HIV-2 infection generally results in lower steady-state levels of viremia (10³ copies/mL) and a more gradual decline in CD4 cell counts.¹⁵,²⁴,¹⁸⁸

Age at the time of infection is an independent risk factor for disease progression, with older persons being at significantly greater risk, perhaps as a consequence of diminished thymic reserve.⁸⁹,³⁰⁹ The role of gender in determining disease progression is less clear. Most studies show that HIV-1–infected men and women progress to AIDS at similar rates.¹³⁴,²⁷³ Early in the course of infection, however, women tend to have significantly lower plasma HIV-1 RNA levels men.¹¹,¹²,¹¹¹ This difference disappears as disease progresses. Because overall progression rates are comparable between the sexes, these observations imply that compared with men, women experience HIV-1 disease progression at lower viral loads. Whether to initiate antiretroviral therapy at lower levels of viremia in infected women than in men remains controversial (see below).

Although virus replication, and consequently viral load, is the engine that drives progression to AIDS, the CD4 cell count is the most useful marker for predicting the immediate risk of developing particular opportunistic infections.¹⁰⁵ Moreover, differences in viral load explain only a small fraction of the variability in rates of CD4 cell decline in patients not receiving antiretroviral therapy,³⁴³ suggesting that other factors such as immune activation triggered by translocation of microbial products across a damaged intestinal mucosa drive CD4 cell loss in HIV infection (see Pathogenesis). Indeed, the proportion of activated CD8+ T cells, measured as the percentage of cells expressing CD38, predicts the risk of disease progression independently of viral load and CD4 count.³¹,¹⁴²,¹⁴³,²⁴⁰ Levels of IL-6 and hsCRP are elevated in patients with HIV infection and independently predict the development of opportunistic diseases.³⁴²

The role of chemokine receptor tropism in determining the rate of disease progression remains unresolved. The prevalence of X4 variants increases with decreasing CD4 cell count, and several studies show a significantly increased risk of disease progression among patients with X4 (SI) virus.⁴⁵,²¹⁷,²⁷⁸,³⁶² Macaques infected with a simian-HIV (SHIV) (SIV/HIV chimera) that expresses an X4 HIV-1 envelope show rapid depletion of CD4 cells, suggesting a causal role of X4 viruses in rapid disease progression.¹⁷⁸,²⁸⁸ X4 variants, however, emerge in only half of patients who progress to AIDS.³⁸,³³⁶ The long interval between infection and emergence of X4 viruses in most patients argues for strong selection against X4 viruses early in the course of HIV disease. Therefore, the possibility that emergence of X4 variants is a consequence, rather than a cause, of advancing immunodeficiency remains a plausible alternative explanation for the apparent association of X4 virus with disease progression. The development of chemokine receptor antagonists as a novel class of antiretroviral agents may provide new tools to address this important pathogenesis question.

The risk of disease progression is moderated also by a variety of host genetic factors. Among the most important are polymorphisms in the genes encoding the chemokine co-receptors and their ligands, and in the HLA genes. HIV-1 uses one of two chemokine receptors as co-receptors for virus entry into CD4 cells: CCR5 or CXCR4.⁹⁷,¹⁰³,¹¹³ Approximately 10% of Caucasians carry a defective allele that has a 32-base pair (bp) deletion in the CCR5 gene (ccr5Δ32); 1% are homozygous for this deletion and resist infection by R5 viruses.²³⁹,³⁵² Infected individuals who are heterozygous for the ccr5Δ32 allele have a slower rate of disease progression than do those who are homozygous for the wild-type allele.⁹⁴ This effect is limited to patients with R5 strains of HIV-1, presumably because syncytium-inducing viruses use the CXCR4 receptor.⁴⁰,²⁷⁰ A mutation in the CCR2 gene, CCR2–64I, likewise reduces the rate of disease progression,²¹⁹,³⁷⁶ possibly by delaying emergence of X4 virus.⁴⁰³ By contrast, mutations in the promoter region of CCR5 (CCR5 P1) are associated with accelerated disease progression.²⁵⁶ Other analyses have failed, however, to show a significant effect of CCR5 and CCR2 genotype after
controlling for CD4 cell count, plasma HIV-1 RNA level, and viral tropism.¹⁴¹,¹⁸⁴

Copy number of the gene encoding the natural ligand for CCR5 (CCL3L1, previously known as MIP-1α) also influences the rate of disease progression. Higher CCL3L1 copy number is associated with a lower risk of progression, which is most pronounced in patients with polymorphisms that reduce the level of functional CCR5 on the cell surface.¹⁴⁵ Variable effects on progression have been noted for the SDF1-3′A polymorphism, which affects the untranslated region of the mRNA encoding stromal-derived factor (SDF-1, also known as CXCL12), the natural ligand of CXCR4.⁸⁶,¹⁴⁵,⁴²¹

Several studies have related host HLA haplotype to the rate of HIV disease progression. For example, presence of the HLA-B27 and B57 alleles is associated with slow progression.⁸,¹¹²,³⁰⁵ Conversely, patients carrying class I alleles B*35 or Cw*04 progress to AIDS significantly more rapidly than do those lacking these alleles.⁵¹,¹³³ Maximal heterozygosity at HLA class I loci A, B, and C is associated with delayed onset of AIDS.²⁰¹ Given that HLA antigen class I molecules play an essential role in antigen presentation to CD8+ CTL, these findings provide strong, albeit indirect, evidence for the importance of CTL in moderating the rate of HIV disease progression.

Variation in the killer immunoglobulin-like receptor (KIR) genes also affects the course of disease. These receptors are found on NK cells and regulate NK activity by recognition of certain HLA class I molecules on the surface of target cells.⁴⁰⁴ The effect of KIR alleles appears to be related to the presence or absence of specific HLA-B or HLA-C alleles.¹³⁷ For example, when present together with the HLA-B Bw4-80Ile allele, the activating KIR allele KIR3DS1 is associated with delayed disease progression.²⁵⁷ By contrast, in the absence of HLA-B Bw4-80Ile, the presence of KIR3DS1 is associated with more rapid disease progression. Similarly, presence of the inhibitory KIR allele KIR3DL1 together with HLA-B*57S is highly protective against disease progression.²⁴⁴ The KIR2DS2 allele appears to be associated with more rapid CD4 decline over time, but has no effect on viral load, whereas HLA-B*5701 and B*2705 alleles are associated with significantly lower levels of viremia.¹³⁶