Chapter 44: The Acquired Immunodeficiency Syndrome–Related Cancers

The relationship between malignancies and acquired immunodeficiency syndrome (AIDS) changed in 1996 when highly active antiretroviral therapy (HAART) was introduced in industrial nations. Thanks to the United Nations and other philanthropy programs, HAART has also been successfully introduced into a number of developing nations (¹). Africa, the pandemic epicenter, is the exception, due to the epidemic magnitude on that continent and its significant political and social turmoil. Prior to 1996, epidemiologists noted specific malignancies afflicting patients with AIDS, with a risk proportional to host immune status. Before HAART, AIDS patients could be separated into two groups: patients with an opportunistic infection as their first manifestation of AIDS (60%) and those with a malignancy as its mode of presentation (40%) (²).

Of those with an AIDS-related malignancy, up to 90% would have Kaposi sarcoma (KS) and the rest non-Hodgkin lymphomas (NHLs), including primary central nervous system lymphoma (PCNSL) and systemic diffuse large B-cell lymphoma (DLBCL). Despite an increase in human papilloma virus (HPV)-related invasive cervical cancers in women with high-grade uterine cervical dysplasia, recent findings of a lack of clear association between cervical cancer and human immunodeficiency virus (HIV)-related immunosuppression questions the validity of including cervical cancer among AIDS-defining or associated malignancies (³). After HAART, previously obvious relationships between AIDS and some malignancies have been challenged. An example is HIV-related Burkitt lymphoma, initially associated with AIDS-induced immunosuppression. Investigators have found that HAART improvement of immunity is associated with significant reductions in KS, PCNSL, and systemic DLBCL, but this is not the case with Burkitt lymphoma. Together with invasive cervical cancer, the incidence of Burkitt lymphoma has remained stable across the pre- and post-HAART eras, increasing its proportional frequency. Epithelial dysplasia and squamous cell carcinomas of the anal canal, rectum, and oral cavity are also observed in men infected by HIV. There has been a post-HAART era increase in Hodgkin lymphoma (Epstein-Barr virus related), lung cancer, and nonmelanoma skin cancer, with implications related to the complex relationship between immunity, aging, chronic antigenic stimulation, and viral oncogenesis. Overall, the excess risk of a malignancy in HIV disease is observed mostly in cancers with an established or suspected infectious cause (⁴).

In this chapter, we first discuss HIV and its effect on the immune system. We then concentrate on malignancies associated with AIDS immunosuppression. We include a discussion of malignancies (Burkitt lymphoma and HPV-related cancers) not directly associated with HIV immunosuppression occurring with a high enough incidence in HIV-infected patients to merit study.

Historical Significance of the Virus

The AIDS pandemic came into the medical world in 1980 with the publication in the Center for Diseases Control and Prevention (CDC) journal Morbidity and Mortality Weekly Report of a series of patients afflicted by de novo opportunistic infections, mostly Pneumocystis jiroveci (formerly known as Pneumocystis carinii) pneumonia and cytomegalovirus (CMV) infections (⁵). At about that time, these conditions were known to occur in immunodeficient patients. It was followed by a report of KS in 26 homosexual men thought to be immunologically healthy. The number of similar cases increased exponentially throughout the United States and Europe, with reports from other parts of the world confirming a new pandemic with an epicenter in sub-Saharan Africa. The high number of cases with similar clinical presentations gave way to a new syndrome, the acquired immunodeficiency syndrome, or AIDS. Twenty-five years after, much is known and much is to be learned about the cause of the syndrome, HIV, and its complications, including opportunistic infections and malignancies (⁶).

Human immunodeficiency virus disease is caused by infection of a human subject with a retrovirus, HIV. Human immunodeficiency virus is only transmitted through blood or unprotected sexual contact, causing a progressive destruction of the immune system with occurrence of opportunistic infections and malignancies. When an opportunistic infection or a malignancy occurs, it signals a significant degree of immunodeficiency. The association of opportunistic infections and immune suppression is well established in the medical literature and clinical practice. The relationship between immune surveillance dysfunction and development of malignancies has also been described in the medical literature. In the early 1970s, kidney transplantation programs in Canada reported an increased incidence of malignancies in patients treated with the immunosuppressant azathioprine (⁷). Similar reports indicated that pharmacologically immune-suppressed patients had increases of several folds of magnitude of certain types of cancers and opportunistic infections. Typically, they had KS or lymphoproliferative malignancies, although other types of malignancies were also reported. In a sense, these patients had a chemically induced AIDS, whereas HIV patients have a biologically induced AIDS.

Origin of the Disease

Human immunodeficiency virus was discovered in 1983 (⁸), dispelling much of the mystery surrounding AIDS. Acquired immunodeficiency syndrome (AIDS) in humans is caused by a retrovirus, HIV, a lentivirus (Latin lentus, meaning “slow,” + virus), endemic in African primates, which entered the human population as a result of a cross-species transmission or zoonosis. Two types of HIV have infected humans: HIV-1 and HIV-2. The HIV-1 virus originated from chimpanzees infected by a retrovirus, the SIVcpz (the chimpanzee simian immunodeficiency virus), and HIV-2 originated from sooty mangabeys monkeys, endemically infected by another retrovirus, SIVsm. SIVcpz is the product of recombination in chimpanzees of two monkeys’ retroviruses preyed upon by chimpanzees: the SIVrcm and the SIVgsn (the red capped mangabeys, Cerocebus torquatus, and the great spot-nosed monkeys, Ceropithecus nictitans, respectively). These viruses do not cause disease in their natural hosts and were named simian immunodeficiency viruses, or SIV, after their genetic and structural similarities with HIV. Each HIV type (1 and 2) is phylogenetically classified in groups and clades; HIV-1 has three groups, M (major), N (non-M, non-O), and O (outlier), with the predominant group M comprising 11 clades A to K, and HIV-2 has six clades, A to F. The HIV-1 and HIV-2 clades can recombine giving origin to genetically complex viral quasispecies. Transmission from animals to humans occurs when infected animals enter in contact with humans as a result of hunting and butchering (SIVcpz from chimpanzees) or through contact with infected animals used as domestic pets (sooty mangabeys monkeys) (⁹).

The HIV-2 virus is endemic and contained in coastal West Africa. The HIV-1 virus is endemic in west equatorial Africa with several clades from the group M responsible for the majority of worldwide infections. In the Western world, HIV-1B is responsible for over 90% of infections, whereas in Africa, subtypes A, C, D, and G, and in Asia, subtype C and the circulating recombinant forms (CRFs) 01 and 02 can be found. The disease is diagnosed by detecting serum anti-HIV antibodies using an enzyme-linked immunoabsorbent assay (ELISA) test. The initial result can be corroborated by a repeated Elisa test (Explanation: saying “with another Elisa test” implies a different type of ELISA) or by a Western blot blood test. The amount of replicating virus can be measured by quantitative polymerase chain reaction (PCR) or via signal amplification such as branch-DNA HIV viral load test (¹⁰).

Status of the Pandemic and Its Effects on the Immune System

Human immunodeficiency virus disease and AIDS are responsible for a human tragedy of incalculable proportions. Since the pandemic’s beginning, there have been over 25 million deaths and more than 50 million infected persons, including those who have died. The estimated rate of new infections is 7,500 new infections per day. Most of these new infections occur in developing nations, affecting many women and children. The most frequent modes of transmission are unprotected sex and intravenous (IV) drugs. The main epicenter of the pandemic continues to be sub-Saharan Africa, with new epicenters in the former Soviet Union, China, India, and Latin America. In the United States, minority groups such as Hispanics and African Americans as well as younger generations of gay men are disproportionately affected by HIV (¹¹).

Human immunodeficiency virus is an enveloped diploid RNA virus (two RNA ribbons per particle), each with an armamentarium of enzymes essential for the virus life cycle. The viral particle membrane is composed of human leukocyte antigen (HLA) groups, other cell surface membrane proteins, and the viral protein gp120 in a trimer form. Among the enzymes within the viral particle are reverse transcriptase, integrase, and proteases. Eighty percent of transmitted mucosal HIV infections are established by a single Transmitted/Founder (T/F) variant, or Transmitted/Founder virus. The other 20% are transmitted by two to five variants (ie, IV drug users). These T/F viruses have specific phenotypic characteristics, including enhanced infectivity, higher envelope content, more efficient binding of dendritic cells, and relative resistance to interferon (IFN)-α (¹²). Once HIV enters a susceptible host, the viral envelope spikes (gp120 trimers) bind to receptors (CD4) and co-receptors (CCR-5) of cells of the immune system, entering through a membrane fusion. The particle content is released into the cytoplasm, and viral RNA is transcribed into DNA. This viral DNA is then randomly integrated into the genome of the infected cell. Through a complex process of transcription activators and as result of the frequent replication cycles of the immune system’s cells, new viral particles are generated. These new particles are released through budding and lysis, continuing to infect new susceptible cells with this process, occurring at an unusually fast pace. Infection with HIV is characterized in simian models of AIDS and in man by a rapid destruction of memory cells of the gut-associated lymphoid tissue (GALT). This process occurs in days to few weeks in animal models and in weeks in man. GALT harbors the majority of body lymphocytes in comparison to the 2% to 5% of lymphocytes located in the peripheral circulation.

The result of GALT’s destruction, together with the activation of B cells and inhibition of the immune system function by HIV viral gene products such as the vif protein, which inhibits the APOBEC3 gene family, implicated in the control of HIV infection including the production of neutralizing antibodies, inducing a severe immune dysfunction and depletion. (This is an important piece of information that address the mechanistic question of why we have such a difficult time developing neutralizing antibodies against HIV) (¹³). Once HIV is integrated into the genome of the susceptible cells, there is a gradual destruction of the immune system, leading to a progressive status of immunodeficiency, with development of opportunistic infections and tumors. This early senescence of the immune system is implicated in changes at a molecular level resulting in loss of control of oncogenic viruses and associated with the malignant transformations observed in AIDS.

Kaposi Sarcoma

Epidemiology

Kaposi sarcoma was described by Dr. Moritz Kaposi in 1872 as an indolent dermatologic disease characterized by the appearance of purplish nodules or plaques, particularly in the lower extremities of older men of Eastern European, Mediterranean, or Jewish descent (classical KS). Canadian investigators were one of the first ones to note in the 1970s that KS occurred in renal transplant patients exposed to immunosuppressant regimens (ie, azathioprine; transplant or iatrogenic KS). Their observations included reports of remissions when immunosuppressant regimens were temporarily discontinued, suggesting a relationship between immunodeficiency and the malignancies. They also noted a high incidence of NHL. In the 1960s, British investigators reported an aggressive lymphadenopathic form of KS confined to equatorial Africa (African KS). This form occurred in younger patients with aggressive involvement of lymph nodes and appearance of nodules and plaques of the lower extremities that rapidly became ulcerated. With the CDC Morbidity and Mortality Weekly Report of 26 gay men with KS, the disease became one of the hallmarks of the AIDS epidemic (epidemic or AIDS-related KS).

In contrast to the classical form, the AIDS-related KS lesions appear with an aggressive pattern of distribution that includes the trunk, arms, and face in addition to the lower extremities (¹⁴). Prior to antiretroviral therapies, patients died of a combination of tumor progression and opportunistic infections. Despite HAART, KS continues to be a prevalent cancer among HIV-infected patients, although HAART has significantly changed the incidence of the disease. From 1990 to 1995, the incidence of KS in the United States was 1,838.9 cases per 100,000 person-years in contrast to 334.6 cases per 100,000 person-years from 1996 to 2002. In the United States and Europe, AIDS-related KS has been almost exclusively diagnosed in homosexual men, suggesting that its prevalence may vary among different categories of AIDS patients. In Africa, where the human herpesvirus 8 (HHV-8), or KS herpesvirus (KSHV), is endemic, the male-to-female ratio of AIDS-related KS in some countries is 2:1, almost the same ratio observed in transplant- or iatrogenic-related KS. Thus, in the presence of profound immune suppression, factors that made the disease more prevalent in males than in females prior to the AIDS pandemic appear to be of little relevance. What is clear is that the incidence of AIDS-related KS is related to the degree of the immune suppression of the infected hosts, with most afflicted patients having CD4⁺ cell counts of 200 CD4⁺ cells/μL or less.

The Viral Etiology of Kaposi Sarcoma

The etiologic agent of all forms of KS is HHV-8, also called KSHV (¹⁵). Early in the pandemic, other viruses or agents were implicated as the cause of AIDS-related KS, including CMV. In 1994, sequences of a new herpes-like virus were isolated from the lesions of an AIDS-related KS patient. Using a subtractive PCR technique called representative differential analysis, investigators found sequences isolated from KS lesions homologous but not identical to other known herpesviruses, and thus it was named HHV-8, because it became the eighth known herpesvirus. Not all patients infected with HHV-8 develop KS; however, viral DNA and seroconversion can be detected in patients prior to the development of KS, confirming the role of HHV-8 as the cause of all forms of KS and the relationship of its pathogenesis to immunosuppression in addition to other cofactors.

Human herpes virus 8 belongs to the γ-herpesvirus subfamily and the subgroup γ-2 or rhadinovirus (from the Latin term rhadino, referring to the tendency of the viral genome to break apart when it is isolated) and is the first human virus of this subfamily identified. The detection of the infection relies on the presence of antibodies against viral antigens using immunofluorescence assays based on the use of B lymphocytes as the antigen source or ELISA with recombinant immunogenic proteins or peptides of HHV-8. The infection seroprevalence mirrors the geographic distribution of AIDS-related KS, with the highest infection rates in central African countries (80%), rates of 25% to 50% among homosexual men in the Western world, and an intermediate level in the Mediterranean regions. The adult general population of blood donors in North America and Europe has an HHV-8 seroprevalence ranging from 0% to 8%. In addition to being the etiologic agent of AIDS-related KS, HHV-8 has been associated with two other lymphoproliferative disorders: primary effusion lymphoma (PEL, a subset of body cavity–based lymphomas [Fig. 44-1], subsequently called PELs) and multicentric Castleman disease (¹⁵).

Pathogenesis of Kaposi Sarcoma

Human herpes virus 8 incorporates a significant number of host genes such as cyclin D and growth factor interleukin (IL)-6 (¹⁶). These genes participate in the replication, survival, and transformation of the infected tumor cells. The viral K1 gene kaposin and viral G protein–coupled receptor (vGPCR) have transformation potential. Others deregulate cell growth and lead to transformation including viral IL-6, viral IL-10, viral cc-class chemokines, and viral FLICE-inhibitory protein (vFLIP). The expression of the different key genes is related to the latency and lytic cycles of HHV-8. During the latency phase, genes such as LANA-1 (latency-associated nuclear antigen 1), in addition to the maintenance of latency, inactivate p53, inhibiting apoptosis. In addition, a viral cyclin prevents cell cycle arrest by cyclin-dependent kinases, pRB and vFLIP, avoiding the activation of the Fas death receptor pathway. During the lytic phase, homologues of replication genes including the K1 kaposin gene, a Bcl-2 homologue, a viral G protein–coupled receptor gene (vGPCRP), a viral homologue of IL (IL-6), and viral macrophages and IFN regulatory factors become active. Some of these genes have immunosuppression functions, such as the inhibition by vFLIP including cytotoxicity of T cells against HHV-8–infected cells and the inhibition of HHV-8 class 2 major histocompatibility complex (MHC)-mediated T-cell activation by K1 (¹⁷). Finally, other viral proteins such as K3 and K5 downregulate the presentation of MHC class 1 molecules on the cell surface.

Pathology of Kaposi Sarcoma

The histology of KS is characterized by the abundance of spindle cells in a matrix of neovascular formation and a rich background of mononuclear inflammatory cells and collagen. Vascular spaces are dilated and contain extravasated erythrocytes. Involvement of the reticular dermis, reflected by patchy lesions and the involvement of all the layers of the skin, clinically presents as nodular or plaque lesions that can coalesce, interfering with the lymphatic circulation, and are histologically and clinically associated with surrounding hemorrhage and subcutaneous edema. The spindle KS cells are rich in endothelial factor VIII. Recent microarray studies have demonstrated that the origin of the KS cell is from a virally transformed lymphatic endothelial cell. Kaposi sarcoma spindle cells express angiogenic/inflammatory cytokines and growth factors, including vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), IL-1, and IL-6, among others. Kaposi sarcoma cells also overexpress receptors for cytokines, suggesting growth through autocrine or paracrine mechanisms. They also proliferate in response to IL-1, IFN-γ, IL-6, and tumor necrosis factor (TNF), which are abundantly present in the serum of patients with poorly controlled HIV infection, and matrix metalloproteinases (MMPs), enzymes involved in the destruction of extracellular matrix proteins required for angiogenesis and metastasis. In AIDS-related KS, Tat (the trans-activator of transcription protein) stimulates KS spindle and endothelial cell replication, promoting an increase in the concentrations of bFGF. This, in turn, upregulates the integrins α₅β₁ and α_vβ₃, receptors for fibronectin and vitronectin, which are highly expressed in AIDS-related KS.

Clinical Features of Kaposi Sarcoma

Most patients with AIDS-related KS have CD4⁺ cell counts of 200 CD4⁺ cells/μL of blood, with an increased number and aggressiveness of lesions in those who are more severely immunosuppressed. There are periods of exacerbation alternated with quiescence related to oscillations of the patient’s immunity. Kaposi sarcoma rarely invades the central nervous system. This is of biological interest because many other human malignancies are characterized by the invasion of the central nervous system (¹⁸).

The distribution of skin lesions in patients with AIDS-related KS often follows the Langer’s folds of the skin. The occurrence of lesions in acral regions of the body such as the tip of the nose is common. The evolution of the skin lesions correlates with the patient’s immune system status. In the pre-HAART era, severe involvement of the skin of the face by raised purplish lesions was frequent. With KS progression, there is frequent involvement of the gastrointestinal tract. Lesions of the palate and gums are often the first ones to be noted, and diarrhea and occasional bleeding can suggest KS involvement of the gastrointestinal tract. In the case of involvement of the lower extremities, progressive edema with nodular and coalescing plaque lesions can cause significant discomfort and pain. Prior to HAART, this was a frequent and serious complication of KS, because “elephantiasis” secondary to the progression of KS was extremely difficult to treat. Advanced cases often involve ulceration of lesions, particularly when located in the lower extremities. Lymph node involvement is frequent, and when there is only generalized lymph node involvement, a biopsy is required for confirmation of diagnosis. In advanced cases, lung involvement is manifested by bilateral basilar infiltrates mixed with a nodular appearance; however, severe hemoptysis or gastrointestinal bleeding is infrequent. During the early years of the epidemic, a significant number of patients experienced involvement by large masses of KS in vital organs such as the liver and heart, and death was due to from progression of their KS tumors and associated opportunistic infections.

Staging and Prognostic Factors for Kaposi Sarcoma

The usual TNM system used in other solid tumors is not easily applicable to AIDS-related KS. A system based on the work of Chachoua and colleagues was proposed in 1989 by the AIDS Clinical Trial Group (ACTG) (Table 44-1). This staging system included the extent of tumor involvement, the immune status measured by the level of CD4⁺ cell count, and presence or absence of any systemic illness (B symptoms). In addition to a complete physical examination, it included a complete blood count, serum chemistries, HIV viral load, panendoscopy of the gastrointestinal tract, computed tomography (CT) of the abdomen and pelvis, and when indicated, the performance of bronchoscopy when pulmonary involvement by KS was suspected. Biopsies of skin or lymph nodes were also suggested when indicated to rule out entities that could be similar in presentation, such as bacillary angiomatosis or pyoderma gangrenosum. After HAART, the extent of disease and the presence of HIV systemic symptoms became the most important prognostic factors; however, pulmonary involvement by KS still carries a particularly poor prognosis. Correlations with the levels of HIV viral load and the status of HHV-8 infection are under study in relationship to their impact on survival (¹⁹).

Therapy of Kaposi Sarcoma

Highly Active Antiretroviral Therapy

Highly active antiretroviral therapy brought a dramatic decrease in the incidence of AIDS-related KS. Highly active antiretroviral therapy consists of the administration of a combination of agents with anti-HIV activity, including inhibitors of HIV reverse transcriptase and protease inhibitors. The consensus of experts in the field about frontline components of HAART is periodically published in Guidelines for the Use of Antiretroviral Agents in HIV-1 Infected Adults and Adolescents by the Department of Human and Health Services (DHHS; available online by visiting the DHHS website). For patients in whom KS is part of the initial diagnosis of AIDS, HAART therapy should be started irrespective of the extent of the disease. For patients with minimal tumor burden, HAART initiation constitutes frontline therapy of their AIDS-related KS. This approach can control these lesions for long periods of time, often more than 1 year, and in many instances results in complete disappearance of KS lesions (²⁰). Patients with extensive disease or visceral involvement can receive systemic chemotherapy in addition to HAART.

Radiation Therapy for Kaposi Sarcoma

Radiotherapy can be useful for treatment of minimal local disease and when the use of systemic treatment other than HAART is not indicated. It can also be used as an adjunct treatment modality for patients in whom the administration of chemotherapy leads to incomplete results, enhancing the beneficial effects of the systemic treatment. Depending on the general condition of the patient and the size of the lesions to be treated, doses range from the administration of a single fraction to fractionated doses over periods of 2 to 4 weeks. For single lesions and frail patients, the administration of a single 800-cGy dose can be used. Radiotherapy can be used for cosmetic reasons, although this should be done carefully to avoid secondary side effects such as postradiation cataracts in the case of periorbital lesions. For larger lesions or when the therapeutic intent is cosmetic, fractionated doses between 200 and 4,000 cGy are effective and carry less risk. For patients receiving systemic therapy, radiotherapy can be an adjuvant for the treatment of complicated single lesions, particularly when they are bleeding, ulcerated, or painful, or when they affect the well-being of the patient. Such is the case of patients with disseminated disease receiving systemic treatment and in whom oral lesions may affect eating due to local pain or size.

Local Therapy Other Than Radiotherapy for Kaposi Sarcoma

In the modern era, the use of local therapies such as cryotherapy and laser therapy may have a role for patients with few and small lesions. The use of surgery may be appropriate in selected cases such as large skin lesions or when there are complications (bleeding of obstruction of a hollow viscus). Other treatments, such as intralesional injection of chemotherapeutic agents, particularly of the oral cavity, and application of alitretinoin gel, have been abandoned and replaced by a more sophisticated use of radiotherapy techniques, HAART, and systemic chemotherapy.

Immunomodulators in Therapy of Kaposi Sarcoma

After the demonstration of the activity of IFN-α in hairy cell leukemia and renal cell cancer in 1984 (²¹), there was an impetus to use the same doses of IFN in patients with AIDS-related KS. Low doses of IFN-α effective against hairy cell leukemia and renal cell carcinoma were ineffective against AIDS-related KS (Rios A, personal observation). A dose-response study unequivocally demonstrated the therapeutic effect of IFN-α in KS when used at doses of 20 to 30 MU/m² (^22,23). A different situation was observed with IFN-γ. Under the angiogenic stimuli of IFN-γ, KS has the capacity to replicate, resulting in a deleterious impact on patients treated with this agent in pilot studies. As a result of these trials, recombinant IFNs α-2a (Roferon-A) and α-2b (Intron-A) were approved for the systemic treatment of patients with AIDS-related KS. Expanded use of these agents has revealed their true activity to be in the range of 15% to 20%.

Interferon can block the synthesis of viral proteins and the budding of viral particles from infected cells in addition to other complex pleiotropic effects. Interferon actions are accompanied by significant systemic side effects including tiredness, fatigue, anorexia, hepatotoxicity, and severe myelosuppression. With the development of HAART and the use of more effective systemic chemotherapy regimens in the treatment of AIDS-related KS, the interest in the use of IFN-α in the treatment of AIDS-related KS has declined.

Chemotherapy for Kaposi Sarcoma

Indications for the use of systemic chemotherapy in AIDS-related KS includes extensive skin, mucocutaneous, and visceral involvement by tumor. In patients who require systemic chemotherapy, local radiotherapy is used to treat local complications in addition to systemic disease. The introduction of HAART has resulted in better and more durable responses with increased tolerability and durability than those observed prior to HAART.

Before the discovery of antiretrovirals, a variety of chemotherapeutic agents had modest to significant activity as monotherapy for KS. These agents included etoposide, vinblastine, vincristine, bleomycin, doxorubicin, vinorelbine, and epirubicin, which induced responses in 40% to 69% of patients. After HAART, ABV (doxorubicin 20 mg/m², bleomycin 10 U/m², and vincristine at maximum doses of 1 to 2 mg) became the first standard treatment of AIDS-related KS. It produced a response rate of 60%, with complications and tolerance depending on the performance status and general condition of the patient (²³). Antiretroviral and other supportive therapies with growth factors (granulocyte-macrophage colony-stimulating factor [CSF] and granulocyte CSF [G-CSF]) paired with vigorous prophylaxis of opportunistic infections reduce the risks of treatment. Complications of ABV were the expected ones with systemic chemotherapy including the potential for cardiac toxicity induced by doxorubicin.

The ABV regimen was followed by the introduction of agents considered today the standard of care for AIDS-related KS, including liposomal encapsulated anthracyclines (doxorubicin and daunorubicin) and taxanes (paclitaxel). The last of these promotes apoptosis and downregulates Bcl-2 protein expression in KS cells in vitro and in KS-like lesions in mice. In addition, it has an important antimitotic effect associated with its capacity for the disruption of tubulin activity during mitosis.

The current treatment of AIDS-related KS is based on the combination of an anthracycline (liposomal doxorubicin 20 mg/m² or liposomal daunorubicin 40 mg/m² but not both together) with paclitaxel 25 mg/m² with or without bleomycin or vincristine. Escalation of the dose of liposomal doxorubicin is not recommended due to a syndrome of desquamation of the skin of the palms and soles of the feet, known as palmar-plantar erythrodysesthesia. In contrast, the dose of liposomal-encapsulated daunorubicin can be increased to up to 60 mg/m² or even higher for patients who tolerate lower doses. This is of particular relevance in patients with advanced disease or significant pulmonary involvement and for whom prompt control and achievement of a quick therapeutic response is of great importance (Fig. 44-2).

Future Therapies for Kaposi Sarcoma

Only patients with early disease and relatively good performance status consistently achieve durable remissions with current therapies for KS. For the rest of the patients, only palliation and stabilization of disease is achieved with current treatments. For these reasons, efforts are under way to develop new therapies based on the knowledge of the pathophysiology of the disease. For example, because angiogenesis is an important component of AIDS-related KS, agents such as thalidomide and anti-VEGF agents such as bevacizumab are of great interest in the therapy of this disease. Metalloprotease inhibitors are also of great interest, and active clinical trials are in progress. Viruses associated with the production of malignancies tend to constitutively activate the nuclear factor-κB (NF-κB) pathway, and agents that can inhibit this pathway such as bortezomib may be of some value. Inhibition of signaling cell receptors implicated in the stimulation of angiogenesis such as platelet-derived growth factor receptor (PDGFR) and C-kit receptor by agents such as imatinib, an orally administered tyrosine kinase inhibitor, approved by the US Food and Drug Administration (FDA) for treatment of chronic myeloid leukemia and gastrointestinal stromal tumor, is being investigated.

There is significant interest in the development of therapies against the latent phase of HHV-8, the most common form of HHV-8 in KS cells, which does not respond to standard antiherpetic drugs such as foscarnet and cidofovir. This area of research has led to potential development of a vaccine against HHV-8. Despite all these new potential therapeutic developments, the impact of HAART in the incidence of AIDS-related KS cannot be overemphasized. The development of more potent and less toxic HAART regimens and the acceptance of earlier therapeutic intervention against HIV seem to be the main paths to control the epidemic of AIDS-related KS.

Acquired Immunodeficiency Syndrome–Related Non-Hodgkin Lymphoma: Systemic Non-Hodgkin Lymphoma

Epidemiology

Non-Hodgkin lymphoma is the second most frequent AIDS-associated malignancy. Both KS and NHL were occurring with an incidence almost linear in its relationship to the patient’s immunodeficiency status (²⁴). In 1985, high- or intermediate-grade B-cell NHLs were considered part of the spectrum of AIDS-related malignancies. Eighty percent of AIDS-related NHLs were systemic (peripheral) lymphomas, involving nodal or extranodal sites, with 15% to 20% originating in the primary central nervous system (PCNSL). A small proportion, less than 3%, of systemic AIDS-related NHL patients had PELs, known as body cavity lymphomas. In general, the risk of AIDS-related NHL in patients with HIV appears to be higher in those who have poor immune function with average CD4⁺ cell counts of 150 CD4⁺ cells/μL of blood.

A viral relationship is implicated in the development of AIDS-associated lymphomas. Epstein-Barr virus (EBV) contributes to the development of most of these tumors, although HHV-8 is associated with the development of PEL (²⁴). There is no relationship between the risk of development of AIDS-related NHL and modes of HIV transmission. The incidence of pre-HAART AIDS-related NHL was 60 to 200 times higher than in a matched HIV-seronegative population; the relative risk was higher for PCNSL. Age, nadir of CD4⁺ cell count, and absence of anti-HIV therapy were critical factors that predicted the development of AIDS-related NHL. In the pre-HAART era, 80% of these NHLs, including systemic and PCNSL cases, were immunoblastic variants associated with CD4⁺ cell count depletion and EBV infection. In the post-HAART era, there has been a 30% reduction of peripheral cases and a 70% reduction in PCNSL, indicating the impact of immune reconstitution in the incidence of immunosuppression-related lymphomas. In contrast, the incidence of Burkitt lymphoma and of centroblastic DLBCL has remained stable without significant change from the pre- to the post-HAART eras (²⁵). When comparing the AIDS-related lymphomas with non–AIDS-related NHL, the former tend to be of higher histologic grade, with increased frequency of B symptoms, extranodal presentations, and an increased incidence of leptomeningeal and primary CNS involvement (²⁶). In the post-HAART era, the World Health Organization (WHO) has expanded the categories of lymphomas that can occur in HIV patients to include extranodal marginal zone B-cell lymphoma of mucosa-associated lymphoid tissue type (MALT lymphoma), peripheral T-cell lymphoma (PTCL), and classical Hodgkin lymphoma (HL), as well as lymphomas that more specifically occur in AIDS patients, including plasmablastic lymphomas of the oral cavity, polymorphic B-cell lymphomas (posttransplant lymphoproliferative disorder–like), and PEL (²⁷). Finally, the demographics of AIDS-related NHL patients have changed in the last decade, reflecting changes in the demographics of the AIDS epidemic, with increasing incidence in Hispanic and African American patients and patients who have acquired HIV through heterosexual contact (Table 44-2).

Pathogenesis of Acquired Immunodeficiency Syndrome–Related Non-Hodgkin Lymphomas

The development of NHLs in HIV patients is similar to that of malignancies associated with other congenital or posttransplant immunodeficiency disorders (^28,29). In such conditions, most of the malignancies consist of NHL and KS. In the case of HIV, immunodeficiency and cofactors, including oncogenic viruses, chronic antigenic stimulation, and cytokine overproduction, are responsible for the development of AIDS-related NHL malignancies. In contrast to AIDS-associated KS, no one has yet found HIV sequences in tumor cells of AIDS-related NHLs (³⁰), although PCR analysis has revealed the presence of HIV in infiltrating T cells. For patients with severe HIV immunodeficiency, the oncogenic nature of both EBV and HHV-8 is responsible for the development of the immunoblastic subtype of DLBCL, PCNSL, plasmablastic lymphoma of the oral cavity, and PEL. The last of these often results from coinfection with HHV-8 and EBV. These lymphomas are the result of active oncogenic viruses released from control by an effective immune surveillance.

Epstein-Barr virus is central to the pathogenesis of AIDS-related NHLs, including those that are related to immunodeficiency and those that occur with a reconstituted immune system, such as centroblastic DLBCL and Burkitt lymphoma. The EBV genome presence is very high in immunodeficiency-associated AIDS-related NHLs (100%) (³¹), although it can only be detected in approximately 60% of centroblastic DLBCLs and 30% of Burkitt lymphomas. This suggests that other factors, including other common latent or chronic viral infections, may be involved in the development of these tumors.

In EBV-infected cells, the EBV virus is, for the most part, in a state of latency with brief periods of lytic activity. The malignant transformation of B cells occurs in the latent phase, requiring multiple molecular events (³²). Epstein-Barr virus contributes to the cellular transformation process through expression of genes with oncogenic activity such as LMP-1, LMP-2, EBNA-1, and EBNA-2. There is also expression of small EBV-encoded, nonpolyadenylated nuclear RNAs (EBERs), all of which participate in the oncogenic transformation phenomenon. These proteins can rescue cells from apoptosis by mimicking cell receptors such as CD40 and B-cell receptor. For example, LMP-1 (latent membrane protein-1) is capable of replacing the function of CD40 in germinal B cells which otherwise would follow an apoptotic fate.

The Impact of Highly Active Antiretroviral Therapy on Distribution of Acquired Immunodeficiency Syndrome–Related Non-Hodgkin Lymphoma

In the post-HAART era there has been a decline in the incidence of NHLs associated with immunodeficiency due to AIDS. In contrast, the incidence of centroblastic DLBCL and Burkitt lymphoma appears to be similar before and after introduction of HAART. Factors influencing the pathogenesis of these diseases include an increase in regulatory cells of the immune system, associated with recovery of the immune status of the host, and effects of chronic antigenic stimulation by HIV with a resultant overproduction of cytokines. The existence of more than one pathogenic mechanism for the occurrence of AIDS-related NHLs can be inferred from the variety of genetic abnormalities displayed by the malignant cells (³³). The number and type of these genetic abnormalities vary according to the anatomic site and tumor histology. They include c-myc rearrangement, bcl-6 gene rearrangement, ras gene mutations, and p53 mutations/deletions (³⁴).

Pathology of Systemic Non-Hodgkin Lymphomas

The hallmark of AIDS-related NHL is a high-grade histology, regardless of the histologic subtype, including diffuse large cell, immunoblastic, and small noncleaved cell lymphomas and Burkitt and Burkitt-like lymphomas. The cells of PEL express CD45, activation-associated antigens such as HLA-DR, CD30, CD38, CD71, epithelial membrane antigen, and CD 138/syndecan-1. Primary effusion lymphoma cells often lack B-cell antigens and c-myc gene rearrangements and mutations and uniformly contain HHV-8 and frequently also contain EBV (^33,34). Other hematologic neoplasms, including low-grade B-cell lymphomas and lymphocytic leukemia, multiple myeloma/plasmacytomas, T-cell neoplasms, and various acute myeloid leukemias and myeloproliferative disorders, have been reported in patients with HIV infection. However, there is no evidence that the incidence of these neoplasms has increased in parallel with the AIDS epidemic (^{35,36,37,38,39}).

Clinical Features of Systemic Non-Hodgkin Lymphomas

Patients with AIDS-related NHLs usually present with advanced stages of the disease and B symptoms, including fever, loss of weight, night sweats, and enlarged lymph nodes or masses. Over 60% of the patients will present with stage III or IV disease. Frequent extranodal sites of involvement are bone marrow, CNS parenchyma and meninges, lungs, and spleen. Patients with PEL present with ascites or a pleural effusion and less frequently with a pericardial effusion. Masses are typically absent in the presentation of PEL, although occasionally a mass may accompany the development of the effusion (⁴⁰) (see Fig. 44-1).

The staging of patients with AIDS-related NHL is similar to non-HIV patients with NHL and should be reported according to the Ann Arbor system (Table 44-3). The International Prognostic Index has been validated in pre-HAART studies, and significant changes in treatment outcomes have occurred since the initiation of HAART (⁴¹). Complete blood count, β₂-microglobulin, lactic dehydrogenase, and complete blood chemistries should be performed, and a radiologic staging should include magnetic resonance imaging (MRI) of the brain and positron emission tomography (PET)/CT scan. Patients in remission after two courses of treatment will tend to remain in remission for the duration of induction therapy. Patients with Burkitt types of lymphoma should have a bone marrow aspiration and biopsy and a diagnostic lumbar puncture. All patients should be screened for hepatitis B, because the exacerbation of this virus by the use of rituximab or chemotherapy can be prevented by screening patients for hepatitis B surface antigen and hepatitis B core antibody and treating those found to have positive results. Hepatitis B surface antigen–positive patients or patients who have a history of hepatitis B with a positive e antigen should be treated for hepatitis B prior to the administration of rituximab. For those with only positive core antibodies, measurements of HBV DNA and of the presence or absence of anti–hepatitis B surface antibodies are important in guiding the decision to intervene (⁴²).

Before HAART, the presence of an opportunistic infection, less than 100 CD4⁺ cells/μL of blood, bone marrow involvement, and increased age predicted for a poor survival, with patients often suboptimally treated due to poor tolerance to standard doses of chemotherapy. After HAART, two factors have become predictors of poor survival: a CD4⁺ cell count of less than 100 cells/μL and high-intermediate International Prognostic Index scores (⁴³).

Therapy for Acquired Immunodeficiency Syndrome–Related Systemic Non-Hodgkin Lymphoma

Prior to HAART, all regimens had in common reduced doses of chemotherapy. After 1996, it became clear that patients on HAART can receive standard doses of chemotherapy. The outcomes of treatment in the presence of HAART are related to the subtype of lymphoma and the specific treatment rather than the immunodeficiency status of the patient (Table 44-4). Vigorous prevention of infections with prophylactic antibiotics, aggressive use of growth factors (G-CSF and pegylated G-CSF), and rituximab where indicated have improved the outcomes for these patients (Table 44-5). Although investigators were originally concerned that rituximab might compromise the immune status of patients with CD20⁺ B-cell lymphomas, studies have demonstrated that rituximab can safely be used for patients with ≥50 CD4⁺ cells/μL of blood.

Regimens commonly used to treat patients with AIDS-related DLBCL include CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone) or rituximab plus CHOP (R-CHOP), R-CDE (rituximab, cyclophosphamide, doxorubicin, and etoposide), and dose-adjusted EPOCH (etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin hydrochloride) or dose-adjusted EPOCH plus rituximab (EPOCH-R). Progression-free survival rates at 2 years for these different treatment regimens are 70% for R-CHOP and R-CDE and approximately 90% for dose-adjusted EPOCH-R (^{44,45,46,47,48,49,50}). In the case of Burkitt lymphoma, CHOP and similar regimens are not recommended because the response is poor. Burkitt lymphoma should be treated with R-HyperCVAD (rituximab plus cyclophosphamide, vincristine, doxorubicin, and dexamethasone) (^51,52) or R-CODOX-M/IVAC (rituximab plus cyclophosphamide, doxorubicin, vincristine, methotrexate/ifosfamide, etoposide, high-dose cytarabine) (⁵³). Either protocol can achieve remissions of more than 92% and a 2-year overall survival rate of 49% (Figs. 44-3 and 44-4). Recent data suggest that dose-adjusted EPOCH-R also has excellent activity against Burkitt lymphoma, establishing this regimen as a potential new standard of care. This regimen known as SC(for short)-RR (Rituximab days 1 and 5)-DA-EPOCH for HIV-related Burkitt lymphoma consists of a 4-day infusion of etoposide (50 mg/m²/d), vincristine (0.4 mg/d), and doxorubicin (10 mg/m²/d) admixed in the same solution, along with prednisone, at a dose of 60 mg/m²/d orally on days 1 to 5. Cyclophosphamide is given at a dose of 750mg/m² as 2-hour infusion on day 5. Rituximab is given at a standard dose of 375 mg/m² on days 1 and 5. Filgrastim is given subcutaneously starting on day 6 until the absolute neutrophil count is 5,000/μL (post nadir). Cyclophosphamide is increased or decreased from the previous course dose if the absolute neutrophil count nadir is over 500/μL or less than 500/μL ANC, respectively. Importantly, only cyclophosphamide is dose-adjusted for hematologic toxicity. Patients receive one cycle after complete remission is established for a minimum of three cycles and a maximum of six cycles (Table 44-6). All patients receive intrathecal chemotherapy (⁵⁴). For patients with relapsed or refractory lymphoma, R-ICE (rituximab, ifosfamide, carboplatin, and etoposide) or R-ESHAP (rituximab, etoposide, methylprednisolone, cytarabine, and cisplatin) can be of value. Patients with disease responsive to salvage therapy can be considered for high-dose chemotherapy and autologous stem-cell transplantation or experimental therapies (⁵⁵).

Acquired Immunodeficiency Syndrome–Related Non-Hodgkin Lymphoma: Primary Central Nervous System Lymphoma

Epidemiology and Pathogenesis

Primary CNS lymphoma became an AIDS-defining malignancy in 1983. Primary CNS lymphoma accounted for up to 15% of NHLs in HIV-infected patients compared to only 1% of NHLs in the general population. In the pre-HAART era, typical patients were men who were younger (median age, 40 years) than their immunocompetent counterparts, with CD4⁺ cell counts of less than 50 cells/μL of blood. The impact of HAART was evidenced by a significant decrease in the disease incidence after 1996 (from 313.2 per 100,000 person-years to 77.4 per 100,000 person-years) (³). The development of PCNSL is related to the effect of HIV immunodeficiency on the activity of EBV. The virus does not replicate in CNS tissue. Thus, infected B cells most likely reach the CNS in increased numbers as a result of the progression of the HIV infection. There is loss of capacity by specific T cells for the production of IFN-γ in response to EBV peptides with increased expression of EBNA-2, LMPs, and EBERs (EBV latency type III). This pattern is seen when EBV transforms primary B cells in vitro. The expression of type III latency upregulates genes involved in transformation, including Bcl-2 and IRF-7, and inactivation of p53 and Rb tumor suppressor gene products (⁵⁶).

Clinical Presentation and Diagnosis

Patients with AIDS and PCNSL present with acute organic brain syndrome, in contrast to immunocompetent patients, in whom neurologic deterioration can be slow and progressive. Headaches, seizures, and focal neurologic signs and symptoms are common. Personality changes are frequent, and nausea and vomiting indicate an increased intracranial pressure. When the rare comatose state occurs, it indicates an acute intracranial catastrophe, such as intratumoral hemorrhage.

Patients with a primary lesion in the brain and no history of systemic lymphoma usually have a brain lesion as the sole manifestation of the disease (Fig. 44-5). Radiologic diagnostic methods paired with analysis of the cerebrospinal fluid are the cornerstones of the diagnosis of PCNSL. Diagnostic MRI of the brain is preferred to CT scan because of the capacity of the former to detect small lesions. Whenever MRI is not available, CT scan is acceptable because it allows detection of larger lesions (≥1 cm) and provides information about safety of performing a lumbar puncture. Prior to and after HAART, the most important differential diagnosis of PCNSL has been cerebral toxoplasmosis, the most frequent cause of cerebral infection and masses in severely HIV-immunosuppressed patients (⁵⁷). In both PCNSL and cerebral toxoplasmosis, multiple lesions can occur, and both can also have enhancing ring lesions, so distinction between the two can be difficult. A positive serology for toxoplasmosis can be helpful if titers are >1:256. If the toxoplasmosis serology is negative, the presence of EBV DNA by PCR in the cerebral spinal fluid and a positive single-photon emission CT–thallium scan of the brain have high specificity for diagnosis of the PCNSL (^58,59). More recently, flow cytometry examination of the cerebrospinal fluid has been used to detect occult disease (⁶⁰).

During the early years of the AIDS pandemic, it was difficult to perform invasive procedures in patients afflicted by an infectious process of which little was known and much was feared. These difficulties resulted in a series of practices borne out of necessity rather than rational approaches to the management and treatment of these patients such as the routine empirical treatment of brain lesions with anti-toxoplasmosis therapy and gauging the diagnosis of the patient based on the clinical response to the anti-toxoplasmosis treatment. Today, there is little room for continuation of such practices, and they should be avoided. They cause unnecessary delays in diagnosis and complicate the patient’s management because there are potential side effects associated with treatment for an illness the patient may not have. Therefore, unless there is an absolute contraindication, the standard of care for a patient with a brain lesion and AIDS is the performance of a stereotactic biopsy of the brain, particularly in patients with a negative serology for toxoplasmosis.

Treatment of Primary Central Nervous System Lymphoma

Highly active antiretroviral therapy is the first step in the treatment of these patients. There is a clear correlation between the immune status of the patient and prognosis. The use of steroids and anticonvulsants is debated among some investigators concerned with the potential of steroids to confound the histologic diagnosis. However, a few days of steroids (4-5 days) can be of clinical benefit, particularly when there is an obvious mass effect. Anticonvulsive therapy administered for a few days can allow the stabilization of the neurologic condition of the patient controlling the risk of focal or grand mal seizures. Even solitary lesions of PCNSL tend to infiltrate surrounding tissues. Thus, there is no role for surgical resection in the treatment of this disease.

Before HAART, whole-brain radiotherapy was standard of care for PCNSL, until replaced by best comfort measures, as the brains of severely immunocompromised patients tolerated poorly the administration of radiotherapy. Similarly to immunocompetent patients with PCNSL, high-dose methotrexate and rituximab have replaced the administration of radiotherapy in patients with HIV and PCNSL. Investigators have suggested a modification of a standard regimen used for the treatment of patients without HIV, which includes rituximab 500 mg/m² on day 1 and methotrexate 3.5 g/m² and vincristine 1.4 mg/m² given only on day 2. Leucovorin rescue is given in standard fashion, and HAART treatment is initiated or continued. This modified regimen is administered for five to six cycles depending on the tolerance of the patient followed by four courses of monthly maintenance with radiotherapy administered in a stereotactic manner at the discretion of the treating physician (⁶¹).

Hodgkin Lymphoma: An Aids-Defining Illness?

Epidemiology, Clinical Features, and Therapy

People infected with HIV have a 10-fold higher risk of developing HL than do HIV-seronegative persons. However, in contrast to KS and NHL, the risk is more pronounced in patients with HIV who only have moderate immunosuppression. In general, patients with HIV have a higher incidence of an unfavorable histology, including mixed cellularity and lymphocyte depletion subtypes of HL, when compared with that seen in patients without HIV infection. Instead of observing a decrease in HL in patients in the post-HAART era, as with certain NHL subtypes, investigators have noted an increase in the incidence of HL in HIV patients (⁶²). This observation has made the relationship between immunodeficiency and HL uncertain. Despite the WHO inclusion as an AIDS-defining malignancy, HL is not considered by most experts as a true AIDS-associated disease. From the pathogenesis point of view, EBV is often associated with HIV-related HL, in the range of 80% to 100%. The Reed-Sternberg cells of HIV-related HL express the EBV-encoded LMP-1, known to have oncogenic properties (⁶³). In the post-HAART era, it has been postulated that an increase in CD4⁺ cells as a result of antiretroviral therapy fosters the development of the appropriate cellular milieu seen in HL in patients without HIV infection. These CD4⁺ cells, generated as a result of immune reconstitution by HAART, produce ligands for membrane receptors in the Reed-Sternberg cells that activate the classical NF-κB pathway (⁶³).

Clinically, patients with HIV and HL are young and have stage III or IV disease with B symptoms (fever, night sweats, and loss of >10% of body weight). Bone marrow involvement is frequent at the time of diagnosis (⁶⁴). In the pre-HAART era, the immunodeficiency of the patients limited the use of standard chemotherapeutic regimens in patients with HL. In the post-HAART era, the standards of care applicable to patients without HIV disease and HL have been applied successfully to patients with AIDS and HL once the HIV disease is controlled. Prior to HAART, Levine and colleagues in the ACTG evaluated the efficacy of ABVD (doxorubicin, bleomycin, vinblastine, and dacarbazine) with G-CSF in 21 HIV-seropositive patients. There was an overall response rate of 62%, with 43% complete response and 19% partial response. The median survival in this cohort was 1.5 years. Almost half the patients experienced grade 4 neutropenia, and 29% of patients developed opportunistic infections (⁶⁵). In this study, patients did not receive antiretroviral therapy while on treatment. Following the introduction of HAART during treatment, investigators reported that ABVD induced a 91% complete remission rate and a median time to relapse of over 36 months (⁶⁶). Other investigators used the Stanford V regimen, administering only short-term chemotherapy (12 weeks) with adjuvant radiotherapy. Of 59 patients who received this therapy, 69% completed the treatment without dose reduction or delays in the administration of the chemotherapy. Eighty-one percent of the patients achieved a complete remission, and with a median follow-up of 17 months, 33 (56%) of 59 patients were alive and free of disease (⁶⁷). After the introduction of HAART, the response rates for patients treated with ABVD or BEACOPP (bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, prednisone) became essentially similar, irrespective of the patient’s stage of the disease. However, more treatment-related mortality occurred in patients who received BEACOPP (⁶⁸). The available data suggest that ABVD with HAART should be the initial treatment of choice for HIV-related HL. High-dose chemotherapy and autologous stem cell transplantation are being explored for patients who have disease progression while on treatment or relapse after remission induction. Thus, because of HAART, patients with HL and HIV disease can be treated with standard-of-care options that are similar to those used for HL patients who do not have HIV.

Cervical Neoplasms

Epidemiology

In the early 1980s, reports appeared signaling an increased association between HIV infection and cervical intraepithelial neoplasia in HIV-infected women. It was not until 1993 that cervical cancer was officially added to WHO recommendations as an AIDS-related malignancy (⁶⁹). Women with HIV disease have a higher incidence of infection with multiple types of oncogenic HPV and higher incidence of dysplastic changes of the cervix than women without HIV disease, events that can culminate in the development of cervical cancer (Table 44-7). There is no obvious association between the level of CD4⁺ cell count and cervical cancer, and the statistical correlations between the association of HIV and HPV-induced cervical cancer remains moderately strong at best. The decline observed for KS and PCNSL after HAART has not been seen in the incidence of HPV-related malignancies.

There are reasons why HPV, the etiologic agent of cervical cancer, causes this disease regardless of the immune status of the infected host. Human papillomavirus infects the basal keratinocytes of the stratified epithelium, and its replication is coupled to the process of keratinocyte differentiation in the infected squamous epithelium. From an initial low-copy number episome in basal keratinocytes, there is a dramatic increase in the concentrations of proteins E1 and E2 by the time the keratinocytes differentiate and enter the stratum spinosum layer of the epithelium. In addition, for oncogenic strains of HPV such as 16, 18, and 31, there is also an increase in the expression of E6 and E7, which have a high oncogenic capacity manifested by the functional inactivation of p53 and Rb (^70,71). Regardless, this is a slow process, and it takes several years for an HPV-induced cervical lesion to transform into cancer. For these reasons, there is uncertainty regarding the relationship between HIV and cervical cancer as an AIDS-related malignancy.

Despite controversies regarding the relationship between HIV, HPV, and cervical cancer, the increased incidence of infections by oncogenic HPV types in patients with HIV in contrast to non–HIV-infected women highlights the importance of mandatory cervical screening of HIV-infected women. However, recent studies have suggested that HIV-infected women are not being offered cervical screening, even though older women without HIV for whom the relevance of the screening may not be considered as important are routinely screened. Clearly, educational efforts are necessary as the number of HIV-infected woman increases in the United States and elsewhere.

Cervical Cytology and Screening

Papanicolaou tests in HIV-infected women have a high prevalence of cytologic abnormalities, ranging from 20% to 40%. Women with negative Pap smears will, over the course of 3 to 5 years, develop cytologic abnormalities at a higher rate than will HIV-negative women, and there is evidence of a higher rate of progression to cervical cancer following the longevity induced by HAART (⁷²). Therefore, it is imperative that patients with HIV infection be screened appropriately with Pap smears, colposcopy, and biopsy when needed for the early detection of cervical cancer. Pap smears interpreted as demonstrating “atypical squamous cells” cannot exclude a high-grade squamous intraepithelial lesion (HGSIL) and must be evaluated with colposcopic examination (see Fig. 44-4). Current US Public Health Service (USPHS) and Infectious Diseases Society of America (IDSA) guidelines recommend Pap smears every 6 months during the first year after HIV diagnosis; if both tests are normal, annual screening is suggested (⁷³) (Fig. 44-6).

Treatment for Cervical Cancer

Because they rarely progress and often go away on their own, the American Society for Colposcopy and Cervical Cytopathology recommends close observation for cervical intraepithelial neoplasia (CIN) grade 1 lesions. However, treatment of invasive cervical cancer is the same as for HIV-seronegative women: surgery in early stages and a combination of surgery and chemoradiotherapy in intermediate stages. For CIN grade 2, the preferred mode of treatment is the loop electrosurgical excision procedure. Cryotherapy and laser surgery can also be used, although these methods are usually reserved for larger lesions. Chemotherapy alone is used for more advanced cases, although the immune status of the patient can influence the response to treatment (⁷⁴).

One of the most important advances against HPV-induced cervical cancer has been the development of preventive HPV vaccines. The HPV vaccines made of viral-like particles display a remarkable structural and antigenic similitude to HPV virions and can induce the production of high titers of neutralizing antibodies. Two versions of HPV viral-like particle vaccines are available for immunization of humans against HPV. Gardasil (Merck, Sharp, and Dohme) has viral-like particles of HPV subtypes 16 and 18 (oncogenic subtypes) and 6 and 11 (the causes of genital warts). Cervarix (GlaxoSmithKline) contains viral-like particles of HPV subtypes 16 and 18. They are both administered intramuscularly and have demonstrated almost 100% protection in clinical trials prior to their approval by the FDA. They should be used prior to the initiation of sexual activity, and neither have demonstrated therapeutic efficacy in the treatment of preexisting infections. Future work is aimed at increasing the immunogenicity of the viral-like particles and expanding the number of HPV oncogenic subtypes available for vaccination (⁷⁵).

Anorectal Carcinoma

Infection of the anogenital tract in men with oncogenic strains of HPV has the same consequence as it does in women. After HAART, the incidence of this disease has increased as patients live longer and the biological characteristics unique to the HPV oncogenic transformation are expressed over time (Table 44-8). The incidence of anal cancer has increased from 19.0 per 100,000 person-years in the pre-HAART era (1992-1995) to 48.3 persons per 100,000 person-years in the immediate post-HAART period (1996-1999) to 78.2 per 100,000 person-years more recently (2000-2003, P < .001) (⁷⁶).

The same subtypes of HPV and pathogenic mechanisms involved in the oncogenic activity of HPV in women apply to men. The screening and treatment of CIN 2 and 3 is of relevance in these patient populations. There has been a greater understanding of the need for anal screening of both men and women with HIV disease (⁷⁷) (Fig. 44-7). The treatment of HGSILs includes the use of local therapies such as podophyllotoxin, liquid nitrogen, and laser surgery (Fig. 44-8). Investigators have recommended that men with HIV infection and anal HPV-related lesions undergo screening with Pap smears every 6 months the first year following diagnosis and yearly thereafter. Invasive lesions are treated as in the general population with the use of chemoradiotherapy followed by salvage surgery when there is no response or relapse after initial treatment (⁷⁸).

Other Malignancies

Patients with HIV can develop other malignancies not necessarily associated with HIV. For example, lung cancer continues to increase in its incidence in this population. In general, smoking is one of the most important negative factors predicting for poor survival in HIV patients even in the presence of HAART. Patients with HIV who develop malignant tumors often do so at an earlier age and tend to have atypical presentations, and frequently their tumors follow a very aggressive course.

In patients with advanced HIV disease (<100 CD4⁺ cells/μL of blood), the initiation of HAART can be accompanied by a paradoxical worsening of established infections or appearance of new ones. This phenomenon is most frequent in patients who have tuberculosis or cryptococcal disease as their opportunistic infection but can happen with any other type of infection. This syndrome is known as immune reconstitution inflammatory syndrome (IRIS). The management of IRIS consists of the administration of specifically indicated therapies and a short course of steroids for 1 to 2 weeks with a rapid taper. The recommended dose of prednisone is 1 to 2 mg/kg/d. Antiretroviral therapy should only be interrupted in severe cases as most patients respond to the use of steroid or anti-inflammatory agents depending on the severity of the IRIS. Because the management of patients with AIDS-related malignancies involves a multidisciplinary team and given the importance of the use of HAART in the management of patients with AIDS and malignancies, treating oncologists must be familiar with this condition to avoid unnecessary delays or interruptions in the HAART of their patients (⁷⁹).

When the AIDS pandemic began, Houston quickly became an AIDS epicenter, occupying the fourth place among cities with the highest number of AIDS cases in the United States for several years. The Department of Epidemiology, under the leadership of Peter W. Mansell and its Director, Guy Newell, took a leading role in studying methods for prevention of AIDS and public education efforts. Their work, together with the collaboration of R. Palmer Beasley, Dean of the University of Texas Health Science Center School of Public Health, the Department of Immunology and Biological Therapy, under the direction of Evan Hersh, and the collaboration of immunologists and virologists including James M. Reuben and Blaine F. Hollinger, was recognized with one of the first AIDS Treatment Evaluation Units Grants awarded for basic science and clinical research in AIDS. This work led to the creation of the Institute for Immunological Disorders under the direction of Peter W. Mansell and Adan Rios. Although the institute was ahead of its time, it opened doors for a humane treatment of AIDS patients and the development of new therapies and strategies for research of the disease. This base of knowledge was instrumental in development of community strategies subsequently applied in management and treatment of AIDS in the state of Texas. This effort extends to current times with the pioneering work done at the institution in the treatment of Burkitt lymphoma in HIV patients by Drs. Jorge Cortes, Debbie Thomas and Houston AIDS community physicians using what is considered today a standard strategy for treatment of AIDS-related malignancies: the use of HAART in combination with the best known strategy for the treatment of the malignancy.

The complexities of the AIDS pandemic are captured in the history of the AIDS-related malignancies (Table 44-9). The introduction of effective anti-HIV therapy or HAART in 1996 brought a profound change in the overall management of cancer in HIV patients. One important predictor of good outcomes in the therapy of cancer and HIV is the administration of HAART (⁸⁰). The common goals for treatment of patients with AIDS and cancer are to treat HIV with HAART and the cancer with the same standards of care that exists for patients without HIV disease, along with vigorous prophylaxis of opportunistic infections, supportive therapy including growth factors, and appropriate nutritional and emotional support.