Malignancies have been detected in approximately 40% of all patients with acquired immunodeficiency syndrome (AIDS) sometime during the course of their illness.
Malignancies have been detected in approximately 40% of all patients with acquired immunodeficiency syndrome (AIDS) at some time during the course of their illness. These cancers have been both a primary cause of death in some patients and a source of considerable morbidity. In the current era of highly active antiretroviral therapy (HAART), patients infected with the human immunodeficiency virus (HIV) are surviving longer than ever. HAART appears to have substantially reduced the incidence of, and mortality from, Kaposi sarcoma (KS) and non-Hodgkin lymphoma (NHL), and it may enhance the efficacy of treatment for patients who do develop these tumors. Unfortunately, HAART has not shown a similar effect on the development of other types of neoplasms, and caring for patients who develop malignancies in the setting of HIV infection remains a challenge.
Sidebar: A review of AIDS-defining cancers (ADCs) from the French ANRS CO4 cohort during four calendar periods (1992–1996; 1997–2000; 2001–2004; 2005–2009) showed significant reductions in the incidence of ADCs across the calendar periods. The risk of developing these cancers remained consistently higher in HIV-infected persons than in the general population, including those with restored immunity (eg, CD4 cell counts > 500/µL) (Hleyhel M et al: Clin Infect Dis 57:1638–1647, 2013).
KS has been the most common tumor associated with HIV infection, but it currently develops in fewer than 10% of homosexual men with AIDS in the United States and in 1% to 2% of other HIV-infected persons. The incidence of KS has declined substantially, from 4,800 per 100,000 person-years in 1990 to 200 per 100,000 person-years in 2005. In 2003, a European study found that the incidence of KS among HIV-infected individuals was less than 10% of the incidence seen a decade earlier, in 1994.
Among AIDS patients in the United States, the incidence of KS is higher in males than in females. There is also a higher incidence of KS in men than in women in Africa (male to female ratio, 2:1), despite the equal prevalence of HIV infection among men and women.
The age distribution of AIDS-related KS follows the distribution of HIV infection. As such, AIDS-related KS can occur in all age groups. In American men, the most common age at onset of AIDS-related KS is 30 to 40 years, although a growing proportion of cases are now being seen in older individuals.
No racial or ethnic differences in the incidence of AIDS-related KS have been observed.
In the United States, KS is seen in fewer than 10% of homosexual men with AIDS. The proportion of KS cases among patients with AIDS-defining diagnoses is lower in parts of Europe, where there are proportionately fewer cases of AIDS in homosexual men (eg, 6.8% of Italian AIDS patients), and higher in parts of Africa, where KS is endemic in the non–HIV-infected population. Among patients with AIDS in the United States, the proportion with KS has declined from the beginning of the AIDS epidemic, possibly as a result of changes in high-risk sexual behavior among homosexual men and the wider use of more effective antiretroviral combination regimens.
In 1994, unique viral DNA sequences were identified in tumor tissues from patients with AIDS-related KS, which led to the identification of a new virus called human herpesvirus type 8 (HHV-8). HHV-8 has been identified in more than 90% of AIDS-KS tumors, as well as in classic KS, endemic African KS, and post–organ transplant–related KS. It has also been identified in body cavity–based lymphoma/primary effusion lymphoma, multicentric Castleman disease, and angio-immunoblastic lymphadenopathy with dysproteinemia in HIV-infected patients.
HHV-8 may be transmitted through sexual contact, blood products, or organ transplant. The seroprevalence of HHV-8 in AIDS-related KS is nearly 100%. HHV-8 has been found in high concentration in the saliva of patients with KS.
HHV-8 is critical in the pathogenesis of AIDS-related KS. The mechanism by which HHV-8 induces KS in susceptible individuals is the subject of intense current investigation.
Various environmental and host factors, including HIV- and HHV-8–induced cytokines, AIDS-associated infections, the degree of immunosuppression, and antiretroviral therapy, may induce or suppress the development of KS and alter its growth.
The manifestations of KS in patients with AIDS are variable and range from small, innocuous-looking cutaneous lesions to symptom-producing visceral or oral lesions, which may be troublesome and even life-threatening. Although KS occasionally may involve just about every internal organ, it is rarely seen in the bone marrow or central nervous system (CNS).
KS tumors typically begin as flat or raised lesions that may progress to plaque-like or nodular tumors. Lesions vary in size and shape but are generally nonpruritic and painless. They range in color from light pink to red to deep purple. KS lesions may be cosmetically disfiguring and may result in social stigmatization that far exceeds any actual physical impairment.
Involvement of the skin and lymph glands with tumor can result in edema of the extremities, periorbital areas, and genitals and may be complicated by skin breakdown and bacterial cellulitis. Edema can be marked and may prevent patients from wearing shoes and/or walking.
These are often asymptomatic but can produce pain and swallowing difficulties.
KS can involve the gastrointestinal (GI) tract. Most lesions are asymptomatic; however, obstruction, bleeding, or enteropathy can occur occasionally.
Lung involvement usually presents as dyspnea without fever and may become severely debilitating and rapidly fatal if untreated.
Currently, there are no screening tests for KS. Although most KS lesions are readily recognized, early lesions may be difficult to diagnose, and the lesions of other diseases (eg, bacillary angiomatosis) may mimic those of KS. Once clinically suspected, the diagnosis of KS is made by biopsy and histologic examination of skin lesions, an excised lymph node, or other tissue or by presumptive diagnosis based on the bronchoscopic or endoscopic appearance of a visceral lesion. Staining for HHV-8 antigen in a biopsy specimen may confirm the diagnosis of KS.
These lesions have a typical red, raised appearance and are difficult to diagnose by biopsy because most are located submucosally.
In patients with pulmonary KS, chest radiographs typically demonstrate diffuse, reticular-nodular infiltrates, mediastinal enlargement and, sometimes, pleural effusion. Bronchoscopy may reveal endobronchial involvement with tumor. Definitive diagnosis requires transbronchial or open lung biopsy. Transbronchial biopsies, however, often yield negative results. In the absence of fever and after exclusion of infections, a presumptive diagnosis of pulmonary KS may be made based on radiographic and endobronchial findings associated with KS-appearing lesions.
Thallium and technetium-99m scanning may help differentiate pulmonary KS from other pulmonary diseases. Patients with KS have been found to have thallium- and technetium-avid scans, whereas pulmonary lymphomas and infections are more typically gallium-avid.
Cutaneous KS is a lesion of the dermis comprising a proliferation of aberrant vascular structures lined by abnormal-appearing, spindle-shaped endothelial cells, and with extravasated erythrocytes and leukocytes within the structures. These spindle cells are generally sparse in the early stages of the disease but become more numerous and “stack up” between the vascular structures as the tumor advances. Infiltration of mononuclear leukocytes, including with plasma cells, T cells, and monocytes, is more prominent in earlier lesions. The histologic appearance of KS in AIDS patients is similar to that seen in non–HIV-infected patients.
The KS tumor cell is believed to be of mesenchymal, endothelial origin. Several endothelial cell markers are positive in KS, including stains for CD31, CD34, and EN-4. In addition, the tumor stains with factor VIIIa, CD68, and alpha-actin but not with pathologische anatomie leiden endothelium.
Although it is difficult to predict from the initial presentation which patients are most likely to have rapidly progressive tumors, several retrospective studies have shown a correlation of survival with the degree of T-cell immunodeficiency, as reflected in the absolute number of T-helper cells in blood. Prior opportunistic infections or the presence of fevers and weight loss also portend a poor prognosis. Patients who develop KS or whose tumor growth accelerates after an opportunistic infection often have a more aggressive clinical course. Patients with pulmonary involvement generally have a poor prognosis.
A tumor classification system has been proposed for AIDS-related KS by the oncology committee of the AIDS Clinical Trials Group. This system (TIS; see Table 1) segregates patients into good or poor prognostic groups based on tumor characteristics, immune system function, and systemic illness. A retrospective analysis of 294 patients with AIDS-related KS has shown that the TIS system is a valid predictor for survival.
The treatment of AIDS-related KS requires an individualized approach, based on the extent and location of the lesions, the wishes and treatment needs of the patient, the presence of tumor-associated symptoms (eg, pain, bleeding, edema), the presence of other AIDS-associated illnesses, and the patient’s tolerance of medications. Nevertheless, the following general statements can be made:
• Patients with widespread symptomatic disease or life-threatening visceral involvement require prompt, cytoreductive treatment with one or more chemotherapeutic drugs.
• Even in the absence of symptomatic visceral disease, the disfigurement and emotional distress wrought by these visible reminders of AIDS may mandate treatment for psychological reasons.
• For patients with asymptomatic indolent lesions, aggressive treatment is not mandatory, but these patients may derive substantial benefits from local treatment or investigational therapies that are directed against HIV or HHV-8 or that may interrupt the pathogenesis of KS and/or restore immune competence.
• Given the heterogeneity and unpredictable growth of this tumor, it is often difficult to gauge objective responses. The peculiarities of this multicentric tumor make some subjectivity almost inevitable in gauging treatment responses.
Staging classification for AIDS-related KS
With the introduction of protease inhibitors and non-nucleoside reverse transcriptase inhibitors for management of HIV infection, cases of KS regression with combination antiretroviral therapy have been reported. Because KS seems to be influenced by the state of HIV infection, it is particularly important for patients with AIDS-related KS to have their HIV infection under optimal control. There is no one best anti-HIV regimen, and oncologists should consult with infectious disease specialists familiar with the treatment of HIV infection. Occasionally, a flare in KS tumor progression may be seen when antiretroviral therapy is initiated. Eventually, even with good anti-HIV therapy, some patients with AIDS-related KS will require some form of treatment for their tumor.
Local treatments, including cryotherapy, topical 9-cis retinoic acid, intralesional chemotherapy, and local irradiation, can produce good local control of tumors. Interferon-alfa (Intron A, Roferon-A) and cytotoxic chemotherapy are effective systemic treatments for patients with more extensive or symptomatic disease. Single-agent or combination chemotherapy is effective in controlling tumors, even in patients with extensive disease and severe immune deficiencies (Table 2). The use of hematopoietic growth factors has facilitated the administration of myelosuppressive treatments, such as interferon-alfa and chemotherapy.
Interferon-alfa. The first treatment licensed for AIDS-related KS was recombinant interferon-alfa. Tumor responses have been seen in approximately 30% of patients treated with subcutaneous (SC) interferon given daily or three times weekly. Current practice is to administer SC interferon-alfa three times weekly. Unmaintained response durations in trials of interferon-alfa monotherapy have ranged from 12 to 24 months in complete responders and from 8 to 12 months in partial responders.
• Duration of therapy-The optimal duration of interferon-alfa treatment is unknown; however, many patients relapse within a few months after discontinuation of therapy. Reinduction of second responses with interferon-alfa after relapse may be unreliable and often is of short duration. It is therefore generally recommended that treatment with interferon-alfa be continued for as long as drug tolerance and tumor responses continue.
• Dose-The optimal dose of interferon-alfa also has not been clearly established. Interferon-alfa is generally administered at either 3 million units or 5 million units SC three times weekly in conjunction with antiretroviral therapy.
• Major dose-limiting toxicities-Major dose-limiting toxicities of interferon-alfa include fever, chills, rigor, and other flu-like symptoms. They are dose-related and often observed at the initiation of treatment but lessen somewhat with continued use. Neutropenia, elevation of transaminase levels, depression, peripheral neuropathy, and other neuropsychiatric abnormalities may also occur.
• Other side effects-Other side effects include headaches, cognitive impairments, paresthesias, and mild thrombocytopenia. Because the subjective side effects of interferon-alfa are also common in HIV-related or other conditions, care must be taken to avoid ascribing all of these symptoms to drug toxicity and overlooking treatable infections and other conditions.
Retinoids. Alitretinoin gel 0.1% (9-cis-retinoic acid [Panretin]) has received approval from the US Food and Drug Administration (FDA) for the topical treatment of localized cutaneous KS. This compound inhibits the growth of KS and induces apoptosis of KS cells by binding to retinoic acid receptors on the cell surface.
Chemotherapy for AIDS-related KS
Phase III clinical trials of alitretinoin three or four times daily compared with placebo gel demonstrated a 35% rate of complete and partial responses in the alitretinoin-treated patients, vs a rate of 18% in controls. The median time to response to alitretinoin was 29 to 34 days, with a median duration of response of 12 to 16 weeks. Responses were seen in both previously untreated and previously treated KS patients and were not dependent on patients’ CD4 cell count.
Local cutaneous adverse reactions to alitretinoin include erythema, skin irritation, skin cracking, flaking, peeling, and desquamation. The severity of these reactions can be mitigated by less frequent dosing, thinner application, or use of topical vitamin E.
Alitretinoin should be reserved for patients who do not require systemic treatment for visceral disease. However, it may be used in conjunction with other treatments for cutaneous disease.
Bexarotene (Targretin), an oral retinoid X receptor–selective agonist, has been studied in patients with AIDS-related KS, with an overall response rate [ORR] of 33% reported in one study.
Chemotherapy. For patients with more widely disseminated, rapidly progressive, or symptomatic disease, systemic chemotherapy is generally warranted. Chemotherapy drugs are included in Table 2.
• Antiretroviral drugs-A total of 32 anti-HIV drugs and combinations of drugs have received FDA approval, and more are in various stages of clinical development. When evaluated by an oncologist, the majority of HIV-infected individuals will be taking some anti-HIV drugs. The interactions between cytotoxic chemotherapy and the various anti-HIV drugs have not been fully studied. Thus, oncologists treating patients with AIDS-related KS should continue to monitor them frequently for side effects. Consultation with pharmacists or pharmacologists familiar with these agents is recommended.
• Combination regimens-The three most frequently used combination chemotherapy regimens are doxorubicin, bleomycin, and vincristine (ABV); bleomycin and vincristine (BV); and etoposide and vincristine (EV). These regimens were initially reported to yield tumor response rates in excess of 70% to 90%, with good palliation of symptoms, including decreased edema, decreased pain and, in patients with pulmonary KS, respiratory improvement and alleviation of obstructive symptoms. A beneficial effect of these combinations on survival has not been clearly demonstrated, however.
High response rates cited in early reports of studies using these combination regimens have not been reproduced by later multicenter trials. A conservative response rate of 50% to 60% has been reported in more contemporary phase III trials. The discrepancy in response rates most likely stems from differences in the response criteria used.
• Liposomal anthracyclines-Liposomal anthracyclines (eg, liposomal doxorubicin [Doxil] and liposomal daunorubicin [DaunoXome]) are also effective in inducing tumor regression in KS. Clinical trials have shown that liposomal anthracyclines administered as single agents can achieve a response rate equal to or better than that obtained with the ABV combination regimen. As such, the liposomal anthracyclines have become first-line chemotherapy for AIDS-related KS.
The dose-limiting toxicity is neutropenia, and many patients will require the use of granulocyte colony-stimulating factor (filgrastim [Neupogen]) after several cycles of treatment. Other common side effects include nausea, fatigue, anemia, and thrombocytopenia. A palmar-plantar syndrome, characterized by acute painful erythematous swelling of the hands and feet, has been reported with the use of liposomal doxorubicin. Once the symptoms resolved, readministration of liposomal doxorubicin did not necessarily reproduce the syndrome.
• Paclitaxel-Paclitaxel has been shown to produce responses in both chemotherapy-naive KS patients and patients with refractory tumors, including those refractory to liposomal anthracyclines. The dosage is typically 100 to 135 mg/m2 IV given over 3 hours every 2 to 4 weeks. This agent is widely considered the primary second-line chemotherapy for KS.
The dose-limiting toxicity is neutropenia. Other reported toxicities include anemia, stomatitis, alopecia, and fatigue. Neuropathy has not been a major problem with this low-dose approach.
• Investigational agents-Other drugs under investigation for the treatment of KS include a number of other antiangiogenesis compounds. Compounds that may affect HHV-8 gene expression (eg, bortezomib) or signal transduction pathways in KS-infected cells (eg, imatinib [Gleevec] and sunitinib [Sutent]) and a mammalian target of rapamycin inhibitor (eg, sirolimus [rapamycin]) have been or are currently being investigated in clinical trials. A phase II study of bevacizumab in patients with HIV-associated KS who were receiving antiretroviral therapy showed that it was tolerated and had activity in a subset of patients (with an ORR of 31%).
Radiation therapy. Although radiation therapy can easily produce sufficient regression of KS to be useful for palliation of symptomatic disease or cosmetic improvement of disfiguring lesions, this practice has become less common because HAART has changed the natural history of AIDS. More than 90% of lesions will respond (complete responses [CRs] and partial responses [PRs]). Local radiation therapy commonly alleviates pain and bleeding, reduces edema, and shrinks obstructing lesions.
• Treatment technique-For most lesions, a single, shaped, en face x-ray beam of relatively limited penetration or low-energy electron beam (eg, 6 MeV) can be used.
For patients with more widespread tumors of the legs with edema, parallel opposed megavoltage x-ray beams and overlying bolus material are often used to provide homogeneous irradiation to the entire area.
• Dose fractionation regimens-Several dose fractionation regimens have proved effective in AIDS-related KS.
For most cutaneous lesions, a single treatment of 800 cGy will produce a short-term response. For lesions on sensitive structures (eg, penis, hands, conjunctivae), some radiation oncologists attenuate treatment to a total dose of 2,000 cGy administered in 300 cGy increments (accepting a 50% decrease in CR rate), whereas 3,000 cGy delivered over 2 weeks is more typically used for lesions in general.
Prospectively acquired data clearly demonstrate a dose-response relationship for radiation therapy in AIDS-related KS. When compared with 2,000 cGy given in 10 fractions or 800 cGy given in 1 fraction, a dose of 4,000 cGy delivered in 20 fractions over 4 weeks was significantly more effective, as measured by a higher response rate, longer duration of tumor control, and the absence of residual hyperpigmentation. However, the short- and medium-term effects of moderately intense but briefer regimens, such as 3,000 cGy in 10 fractions over 2 weeks, probably are equivalent to those of higher-total-dose and more protracted regimens, and the moderately intense regimens require only half the time to deliver.
For patients with an anticipated survival time of less than 3 months, in whom the response duration may be of less overall importance, a single fraction of 800 cGy is likely to provide the same benefit as the more intensive regimens. In contrast, small lesions in patients who are expected to survive for at least 1 year should be treated with fractionated radiation therapy, such as 3,000 cGy in 10 fractions over 2 weeks.
The incidence of NHL is 60 times higher in individuals with HIV infection than in the general population. The overall occurrence of lymphoma as a manifestation of AIDS has declined somewhat as treatment of HIV infection has improved.
Although NHL currently accounts for less than 5% of all initial AIDS-defining conditions, it accounts for as many as 28% of all AIDS-related deaths. The majority of patients present with advanced-stage, high- or intermediate-grade, B-cell lymphoma and have a high frequency of extranodal involvement. Primary CNS lymphoma, which previously was fairly common in AIDS patients, is now seen very infrequently in HIV-infected patients because of the availability of better drugs and earlier initiation of antiretroviral therapy.
The majority of patients with AIDS-related lymphoma have advanced HIV disease. Median CD4 cell counts in patients with systemic lymphoma range from 100/Î¼L to 200/Î¼L, although NHL can occur at any CD4 cell count.
NHL occurs with approximately equal frequency in all population groups infected by HIV, including intravenous drug users, homosexual and bisexual men, transfusion recipients, and patients with hemophilia.
AIDS-related NHL is seen more frequently in men than in women and occurs more often in whites than in blacks.
The age distribution of AIDS-related NHL follows the distribution of HIV infection.
Current data do not indicate any geographic differences in the incidence of AIDS-related NHL.
AIDS-related NHL is believed to arise as a consequence of continued stimulation of B-cell proliferation as a result of HIV infection, Epstein-Barr virus (EBV) infection, and other infections, all of which occur in the setting of profound T-cell immunodeficiency. HIV also induces the expression of a number of cytokines (eg, interleukin [IL]-6 and IL-10) that can further increase B-cell activation.
Genetic errors are increased in the setting of chronic B-cell proliferation, and a variety of chromosomal translocations resulting in oncogene activation can lead to polyclonal and monoclonal B-cell expression. Other molecular biological abnormalities associated with small noncleaved lymphomas include expression of an abnormal TP53 (also known as p53) tumor-suppressor gene and presence of the c-myc or ras oncogene.
The pathogenesis of AIDS-related immunoblastic lymphoma appears to be distinct from that of small noncleaved cell lymphoma and is more likely related to EBV infection without c-myc dysregulation. Clonal integration of EBV within tumor cells, with expression of various latent EBV proteins, has been demonstrated in essentially all cases of AIDS-related primary CNS lymphoma and in as many as two-thirds of systemic lymphomas.
The specific molecular aberrations described in patients with AIDS-related diffuse large cell lymphoma appear distinct as well, with recent descriptions of abnormal BCL6 expression in approximately 40% of cases.
This appears to be highly associated with HHV-8 and EBV. The tumor cells stain positive for CD45. The disease appears to occur predominantly in males and may coexist with KS in patients with AIDS.
These events (ie, fever, weight loss, and night sweats) are seen in approximately 80% of patients with systemic AIDS-related NHL. In these patients, it is mandatory to exclude the presence of occult opportunistic infections before ascribing B symptoms to the lymphoma itself.
Advanced-stage disease is expected in the majority of patients, with extranodal involvement reported in 60% to 90% of patients in most series. Common sites of extranodal involvement include the CNS (occurring in approximately 30% of patients), GI tract (25%), and bone marrow (25%). Essentially any other site in the body can also be involved, including the rectum, soft tissue, oral cavity, lungs, and heart.
CNS lymphoma. Patients with primary CNS lymphoma often present with focal neurologic deficits, seizures, and/or altered mental status. Any site in the brain may be involved, and one to four space-occupying lesions are usually seen on magnetic resonance imaging (MRI) or computed tomography (CT) scan.
Other sites. Changes in bowel habits, GI bleeding, weight loss, pain, and hepatomegaly are common presenting symptoms in patients with GI involvement. Pancytopenia may indicate bone marrow involvement.
Primary effusion lymphoma. Patients usually present with pleural or pericardial effusion without an identifiable mass. Pain, shortness of breath, and B symptoms are the main initial complaints.
Diagnosis of NHL in patients with AIDS requires histologic confirmation by biopsy with immunophenotypic and/or molecular gene-rearrangement studies.
A complete staging evaluation should be done. This should include:
• CT or MRI of the head
• Positron emission tomography/CT scan of whole body
• Bone marrow aspiration and biopsy
• Liver function studies
• Spinal fluid analysis
The presence of EBV DNA in cerebrospinal fluid, as determined by polymerase chain reaction, appears to have a high specificity and sensitivity for the diagnosis of primary CNS lymphoma.
More than 95% of AIDS-related NHL cases are of B-lymphocyte origin. Most AIDS-related NHL tumors are high-grade types, including the immunoblastic and small noncleaved cell lymphomas. Diffuse large cell lymphoma constitutes up to 30% of AIDS lymphomas.
Although not considered part of the AIDS epidemic, several cases of T-cell lymphoma occurring in HIV-infected patients have been described. In addition, cases of Ki-1–positive, large cell anaplastic lymphoma and plasmablastic lymphoma have been reported in HIV-infected patients. The clinical and pathologic characteristics of these forms of lymphoma are similar to those seen in non–HIV-infected individuals.
These are typically of the immunoblastic or large cell type.
Large cell or immunoblastic lymphomas are also more likely to involve the GI tract and oral cavity than are small noncleaved lymphomas.
The cells are large and pleomorphic with prominent nucleoli and immunoblastic morphology. Clonal immunoglobulin DNA rearrangement demonstrates clonality of the tumor cells but not surface immunoglobulin expression.
Staging of AIDS-related NHL is the same as that for non–AIDS-related NHL. The Ann Arbor classification system for staging of NHL is used (see “Non-Hodgkin Lymphoma” chapter), and the staging workup includes imaging studies as well as bone marrow and CNS evaluation for lymphomas.
Five factors have been shown to correlate most closely with a shorter survival time in patients with systemic AIDS-related NHL:
• A history of opportunistic infection before the lymphoma
• CD4 cell count less than 100/Î¼L
• High International Prognostic Index score
• Karnofsky Performance Status (KPS) score less than 70
• Stage IV disease, especially if due to bone marrow or meningeal involvement
Chemotherapy for AIDS-related NHL
In patients without these findings, the median survival is typically 11 to 12 months, compared with a median survival of approximately 4 to 5 months in those with one or more of these adverse prognostic features.
Three factors correlate with better survival in patients with primary CNS lymphoma:
• KPS score greater than 70
• Age less than 35 years
• Adequate dose of radiation therapy
To date, no major differences have been seen in response or survival among the various pathologic types of systemic AIDS-related NHL. Patients with polyclonal lymphomas appear to have better tumor responses to chemotherapy and better survival. Patients with primary CNS lymphoma have an extremely poor prognosis. The prognosis for patients with primary effusion lymphoma is also poor, with a median survival time of only 5 months.
Chemotherapy. The mainstay of treatment for patients with systemic AIDS-related NHL is chemotherapy. Because the likelihood of tumor dissemination is great, AIDS patients who develop NHL must be assumed to have widespread disease at presentation and should be treated with systemic chemotherapy, even if tumor dissemination is not confirmed on routine staging evaluation.
Some of the commonly used regimens designed for AIDS-related NHL are listed in Table 3. No regimen appears to be superior to any other, although recent findings show that the R-EPOCH regimen (rituximab, etoposide, prednisone, vincristine [Oncovin], cyclophosphamide, doxorubicin) gives the best results to date.
CNS prophylaxis with either intrathecal cytarabine (50 mg) or intrathecal methotrexate (10–12 mg) every week for four treatments has been shown to be effective in reducing the incidence of CNS relapse.
The role of rituximab in AIDS-related lymphoma is still somewhat debated, given that it induces greater remissions but can cause more cellular and humoral immunodeficiency and may predispose patients to life-threatening infections. Prospective trials evaluating the use of rituximab in AIDS lymphoma have generally been positive. Given these data, clinicians should consider using rituximab in patients with CD20-positive NHL who have CD4 cell counts greater than100/Î¼L and using it in combination with multiple prophylactic antibiotics.
Dose intensity. Standard-dose chemotherapy (eg, cyclophosphamide, doxorubicin, vincristine, and prednisone [CHOP]) is also effective for most patients with AIDS-related NHL. Results from trials using continuous infusion therapy (eg, R-EPOCH) have shown better CR rates and long disease progression–free and overall survival, and this therapy is now generally considered the best first-line treatment for AIDS-related diffuse large B-cell NHL.
Certain subsets of patients with high CD4 cell counts (> 100/Î¼L), no B symptoms, lower disease stage at presentation (stage I or II), and good performance status (0 or 1) have enjoyed prolonged relapse-free survival (> 2 years) when treated with either a standard-dose or intensive, high-dose regimen.
The AIDS Malignancy Consortium conducted a phase II study of 34 patients with HIV-associated adult Burkitt lymphoma (BL) and atypical BL, to address concerns about BL-treatment toxicity in HIV-positive patients. Patients in the study received modified CODOX-M/IVAC-rituximab (cyclophosphamide, vincristine, doxorubicin, high-dose methotrexate/ifosfamide, etoposide, and high-dose cytarabine with rituximab). Despite five early treatment terminations, the 1-year overall survival rate was 72% (95% CI, 53–84), 2-year overall survival was 69% (95% CI, 50–82), and 1-year progression-free survival was 69% (95% CI, 51–82). A total of 27 patients experienced at least one grade 3 to 5 toxicity, including 20 hematologic, 14 infectious, and 6 metabolic events. No patients had grade 3 to 4 mucositis. The authors noted that the 1-year survival rate was comparable to that observed in two BL studies with no HIV-positive patients, and that modifications of the CODOX-M/IVAC regimen yielded a grade 3 to 4 toxicity rate of 79%, whereas the parent (unmodified) regimen had a 100% toxicity rate.
Growth factor and other support. The major dose-limiting toxic effect of multiagent chemotherapy regimens is myelosuppression. Studies of methotrexate with leucovorin, bleomycin, doxorubicin, cyclophosphamide, vincristine, and dexamethasone (m-BACOD); R-EPOCH; or CHOP chemotherapy demonstrated that coadministration of myeloid hematopoietic growth factors enhanced patient tolerance of these regimens. In addition, prophylaxis with trimethoprim/sulfamethoxazole, azithromycin, fluconazole, valganciclovir (Valcyte), and ciprofloxacin can also reduce the risk of infection during intensive chemotherapy and rituximab therapy.
Salvage chemotherapy. Patients in whom initial treatment fails or relapse occurs after initial remission rarely achieve a prolonged second remission. Studies of various salvage regimens with or without autologous stem cell support have demonstrated encouraging results.
Second-line chemotherapy (eg, etoposide, methylprednisolone, high-dose cytarabine, cisplatin [ESHAP]) has been shown to produce a CR of up to 31% and a PR of 23%, with a median survival of 7.1 months, in patients with refractory or relapsed AIDS-related NHL.
A pooled analysis of AIDS Malignancy Consortium trials evaluating rituximab plus CHOP versus rituximab plus EPOCH provided additional level 2 evidence supporting the use of concurrent R-EPOCH in patients with HIV-associated lymphoma and a CD4 cell count greater than 50/Î¼L.
Radiation therapy. The role of radiotherapy in systemic lymphoma is limited to consolidation of the effects of chemotherapy. Treatment principles are similar to those used for aggressive NHL in the non-HIV setting and typically involve the use of involved or extended fields only.
• Lymphomatous meningitis-For patients with lymphomatous meningitis and/or radiographically detectable cerebral deposits, “step-brain” irradiation (including the covering meninges) is administered along with intrathecal chemotherapy to control microscopic spinal disease. Focal radiation therapy may be required for known tumor deposits in the spine. Unfortunately, many such patients develop multiple deposits anywhere along the spinal axis, either synchronously or metachronously.
Fractionated doses of 3,000 to 4,500 cGy may be used to control local lymphoma deposits in nodal areas. Patients who have lymphomatous meningitis typically have a poor prognosis and are best treated with regimens that do not unduly occupy their time (eg, 3,000 cGy in 10 fractions over 2 weeks).
An effective therapy for patients with AIDS-related CNS lymphoma has not yet been found, although many AIDS oncologists find high-dose methotrexate with leucovorin rescue more effective than radiation therapy.
Cervical carcinoma in the setting of HIV infection has been recognized as an AIDS-defining malignancy since 1993. Unfortunately, in some women, cervical carcinoma may be the first indication that they have HIV infection.
Cervical intraepithelial neoplasia (CIN) is also seen in association with HIV infection. These premalignant lesions, also known as squamous intraepithelial lesions (SILs), may foretell a higher incidence of cervical carcinoma among HIV-infected women. SILs have been associated with human papillomavirus (HPV), particularly those subtypes with greater oncogenic potential, such as serotypes 16, 18, 31, 33, and 35.
The risk of HIV infection in women with an abnormal Pap smear varies with the prevalence of HIV infection in the given population. Screening in clinics in high-prevalence areas has yielded HIV-positivity rates of between 6% and 7% (and up to 10% in parts of Africa). In such high-prevalence areas, among women younger than 50 years with cervical carcinoma, up to 19% were found to be HIV-positive. HIV-positive women have up to a 10-fold increased risk of abnormal cervical cytology. Several centers have reported abnormal cytology rates of 30% to 60% in HIV-positive women and Pap smears consistent with cervical dysplasia in 15% to 40%. The prevalence of cervical dysplasia increases with declining CD4 cell counts in HIV-infected women.
Nationwide, invasive cervical carcinoma was found in 1.3% of women with AIDS. In New York, invasive cervical carcinoma constitutes 4% of AIDS-defining illnesses in women. Recent findings from linkage studies in the United States and Italy clearly have shown increased rates of cervical cancer in women with HIV infection.
The prevalence of invasive cervical carcinoma among American Hispanic and black women is higher than that in white women. However, this difference may stem from a difference in access to health care. The southern and northeastern sections of the United States have a higher reported number of cases of HIV-associated invasive cervical carcinoma.
The severe cellular immunodeficiency associated with advanced HIV infection may allow oncogenic viruses to flourish and may also compromise the body’s immunologic defenses that control the development of these tumors.
There is abundant evidence that HPV infection is related to malignant and premalignant neoplasia in the lower genital tract. HPV serotypes 16, 18, 31, 33, and 35 are the most oncogenic strains and have been associated with invasive cervical carcinoma and progressive dysplasia. The prevalence of cervical SILs among HIV-infected women may be as high as 20% to 30%, with many having higher cytologic and histologic grade lesions.
The majority of cervical SILs are detected on routine cytologic evaluation of Pap smears in women with HIV infection.
Postcoital bleeding with serosanguineous and/or foul-smelling vaginal discharge is usually the first symptom of more advanced invasive disease. Lumbosacral pain or urinary obstructive symptoms may indicate advanced disease.
Because the majority of patients with cervical dysplasia or early invasive cancer are asymptomatic, frequent cytologic screening of women at risk for HIV infection must be undertaken. The role of newly developed HPV vaccines in preventing HPV infection or disease progression in HIV-infected women has yet to be determined.
Current screening recommendations call for women with HIV infection to have pelvic examinations and cytologic screening every 6 months during the first year after HIV diagnosis and then annually if the test results are normal. Pap smears indicating cervical SILs must be taken seriously, and abnormalities justify immediate colposcopy. Although abnormalities are sometimes missed by relying solely on cytologic screening, recommendations for routine colposcopy have not yet been established.
For women who have a history of cervical SILs, more frequent reevaluation and cytologic screening should be undertaken. Since these women are at high risk for recurrence or development of lesions in other areas of the lower genital tract, post-therapy surveillance with repeated colposcopy also is warranted.
For women with invasive carcinoma, complete staging should be undertaken; this should include pelvic examination, CT of the pelvis and abdomen, chest radiography, and screening laboratory tests for hepatic and bone disease. In addition, full evaluation for and treatment of HIV infection and related complications should be initiated.
Most cases of cervical carcinoma are of the squamous cell type.
The staging classification for cervical carcinoma (see “Cervical Cancer” chapter), as adopted by the International Federation of Gynecology and Obstetrics, also applies to AIDS patients.
Cervical dysplasia in HIV-infected women is often of higher cytologic and histologic grade. These women are more likely to have CIN II–III lesions with extensive cervical involvement, multisite (vagina, vulva, and anus) involvement, and endocervical lesions.
HIV-infected women with cervical carcinoma typically present with more advanced disease and appear to have a more aggressive clinical course. Tumors are typically high-grade with a higher proportion of lymph node and visceral involvement at presentation. Mean time to recurrence after primary treatment is short, and many patients have persistent disease after primary therapy. The median time to death in one series was 10 months in HIV-infected women, compared with 23 months in HIV-negative patients.
Cryotherapy, laser therapy, cone biopsy, and loop electrosurgical excision procedure have all been used to treat preinvasive disease in HIV-infected patients. Short-term recurrence rates of 40% to 60% have been reported.
Determinants of recurrence. Immune status of the patient seems to be the most important determining factor for recurrence. Close surveillance after initial therapy is critical, and repetitive treatment may be necessary to prevent progression to more invasive disease.
The same principles that guide oncologic management of the immunocompetent patient with cervical carcinoma (see “Cervical Cancer” chapter) are used in AIDS patients with this cancer.
Resection. Resection can be undertaken for the usual indications, and surgical decisions should be based on oncologic appropriateness and not on HIV status.
Radiation therapy. Because most AIDS patients with cervical cancer present with advanced disease, radiation therapy is indicated more often than surgery. If the patient’s overall physical condition permits, treatment regimens are identical to those used for the same stage disease in uninfected individuals (see “Cervical Cancer” chapter). It is important to note that the standard of care for advanced carcinoma of the cervix (stages III–IV, without hematogenous dissemination) now includes a combination of irradiation and concurrent cisplatin-based chemotherapy. At present, there is insufficient evidence to suggest that irradiation or other treatments for cervical carcinoma in AIDS patients is any less effective than in similar non–HIV-infected individuals.
Chemotherapy. Antineoplastic regimens, such as cisplatin (50 mg/m2) or carboplatin (200 mg/m2), bleomycin (20 U/m2; maximum, 30 U), and vincristine (1 mg/m2), have been used in patients with metastatic or recurrent disease. Vigorous management of side effects and complications of these treatments and of AIDS itself must be provided.
A prospective cohort study of 1,123 women in South Africa was performed to determine the impact of HAART on incidence, regression, and progression of HPV-related cervical lesions in HIV-infected women. Women receiving HAART had increased regression and a decreased incidence of cervical lesions. These findings support the positive impact of HAART on the natural history of HPV-related cervical disease in HIV-infected women.
Although anal carcinoma is not currently an AIDS-defining illness, the incidence of this tumor is increasing in the population at risk for HIV infection. The incidence of anal carcinoma in homosexual men in a San Francisco study was estimated at between 25 and 87 cases per 100,000, compared with 0.7 cases per 100,000 in the entire male population.
Precursor lesions of anal intraepithelial neoplasia (AIN), also known as anal SILs, have been found to be associated with HPV infection, typically with oncogenic serotypes, eg, types 16 and 18. Cytologic abnormalities occur in nearly 40% of patients, especially those with CD4 cell counts less than 200/Î¼L. Abnormal cytology may predict the later development of invasive carcinoma.
Rectal pain, bleeding, discharge, and symptoms of obstruction or a mass lesion are the most frequent presenting symptoms.
Studies to evaluate the usefulness of anoscopy with frequent anal cytology have been undertaken to determine whether early detection of AIN may result in interventions that would prevent the development of invasive tumors.
For patients with anal carcinoma, determination of the extent of local disease, as well as full staging for dissemination, should be undertaken (see “Colon, Rectal, and Anal Cancers” chapter).
The majority of anal carcinomas are of the squamous cell type.
The grading for AIN is similar to that for CIN, with AIN-1 denoting low-grade dysplasia and AIN-2 and AIN-3 referring to higher-grade dysplastic lesions. The gross appearance of lesions on anoscopy does not predict histologic grade. Higher-grade dysplastic lesions are seen in patients with lower CD4 cell counts.
The staging of squamous cell carcinoma of the anus in HIV-infected individuals is the same as that in non–HIV-infected patients (see “Colon, Rectal, and Anal Cancers” chapter). Patients with severe immunosuppression (ie, CD4 cell counts < 50/Î¼L) may present with more advanced, more aggressive disease. The true natural history of this tumor in the AIDS population has yet to be defined, however.
Treatment of patients with local AIN is similar to that of women with CIN. Ablative therapy may be used.
Anal cancer can be controlled with chemotherapy and radiation therapy despite HIV infection. However, patients who have low CD4 cell counts appear to be more likely to experience severe toxicity and to require colostomy for salvage therapy than those with higher CD4 cell counts. For patients with squamous cell carcinoma of the anus, chemotherapy with mitomycin (10 mg/m2 on day 1) and fluorouracil (5-FU; 1,000 mg/m2 by continuous infusion on days 1 to 4) combined with radiation therapy can produce high rates of complete remission.
Concomitant treatment with radiation therapy, 5-FU, and mitomycin has been reported to produce a CR in 9 of 11 patients with AIDS-associated invasive anal carcinoma (median CD4 cell count at diagnosis of 209/Î¼L) and a 60% 2-year actuarial survival rate. Two patients remain alive more than 8 years following treatment, but severe toxicity (three grade 3 hematologic, one grade 3 dermatologic, and one grade 4 and one grade 5 GI toxicity) and one death resulted from treatment.
Evidence appears to suggest that 5-FU plus cisplatin may be better tolerated than 5-FU plus mitomycin in HIV-infected patients. Tolerance to treatment seen in patients who have relatively intact immune systems is similar to that seen in HIV-infected patients with anal cancer. The role of cetuximab (Erbitux) and other target therapies in this disease population also has not yet been determined. In addition, the appropriate dose of radiation therapy for patients with anal carcinoma in the context of HIV infection remains unsettled. Patients who are unable to tolerate chemoradiation therapy and those in whom treatment fails (with failure defined as a positive biopsy after a CR) should be considered for abdominoperineal resection.
The outcomes of HAART for anal cancer treatment in HIV-infected patients appear to be comparable to those of anal cancer treatment in the general population.
Because more effective control of HIV infection has become possible, patients are living longer and are being affected by types of cancers that are not directly induced by HIV. A growing number of reports from around the world now indicate that other malignant tumors, including Hodgkin lymphoma (HL), nonmelanomatous skin cancers, lung cancer, germ cell tumors, myeloid or lymphoid leukemias, multiple myeloma, renal cell carcinoma, head and neck cancer, brain tumors, squamous tumor of the conjunctiva, and leiomyosarcoma in pediatric patients, are occurring in HIV-infected individuals.
Rates of other non–AIDS-defining cancers appear to be increasing among HIV-infected individuals. In an Australian cancer database, 196 cases of non–AIDS-defining cancers were noted in 13,067 individuals (1.5%), including 8,351 notified with HIV infection and 8,118 registered with AIDS. Because HIV-infected individuals are surviving longer with currently available combination anti-HIV drugs, clinicians should anticipate seeing the development of more tumors in these patients. Greater vigilance for these tumors is warranted.
These non–AIDS-defining cancers are generally treated with the same stage-adjusted strategies as would be used in individuals not infected with HIV.
As reported by Baeyens et al, because HIV infection has yielded to better therapies, unrelated tumors occurring in infected individuals raise the question of tolerance to and efficacy of otherwise standard therapies in this setting. To some degree, the concern is fueled by in vitro studies, such as a report that demonstrated heightened sensitivity of HIV-infected T lymphocytes to damage from radiation, but the in vivo data are not uniformly supportive.
Researchers in Japan concluded that damage to the skin and mucosa, including stomatitis and diarrhea, tended to occur after low-dose radiation therapy; however, no severe acute adverse effects were seen in other organs, such as the brain, lung, and bone. Acute adverse effects often occurred earlier in HIV-positive patients and became severe more frequently than in the general population, especially disorders of the mucosa, which tended to arise rapidly. It was shown by Kaminuma et al that radiation therapy is safe when it is performed carefully and that it is a very useful treatment for cancer in HIV-positive patients.
Studies showing a possible increased incidence of HL in HIV-infected individuals have been reported from Europe, Australia, and the United States. The most common histologic types are mixed cellularity and lymphocyte-depleted HLs. Male predominance, a higher prevalence of B symptoms, and more extranodal disease on presentation are the main characteristics of HL in HIV-infected patients.
Chemotherapy is recommended for this group of patients, because of the high proportion of stage III or IV disease. Standard treatments include doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) or ABVD alternating with vincristine mechlorethamine (Mustargen), vincristine, procarbazine (Matulane), and prednisone (MOPP). More recently, early results with bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone (BEACOPP) and the Stanford V regimen (mechlorethamine, doxorubicin, vinblastine, vincristine, bleomycin, etoposide, and prednisone) look promising. The addition of brentuximab vendotin to doxorubicin, vinblastine, and dacarbazine is also being tested.
As in the general population, basal cell carcinoma is more common than squamous cell carcinoma in the setting of HIV infection. The risk factors for the development of these tumors are the same as in the general population: namely, fair skin, history of sun exposure, and family history.
Patients with HIV infection appear to have a higher relative risk of developing lung cancer than do age-matched controls (relative risk = 4.5; 95% CI, 4.2–4.8). These tumors tend to present at later stages than do tumors with a similar histologic distribution in the general population. Retrospective cohort studies have also suggested poorer survival in HIV-infected individuals with lung cancer than in those without HIV infection.
Sidebar: With the use of HAART, patients who have HIV-associated lung cancers are living longer than they did previously, and the lung cancer itself is now the predominant cause of death (Bearz A et al: Eur Rev Med Pharmacol Sci 18:500–508, 2014).
Cases of aggressive leiomyosarcoma developing in HIV-positive children have been reported. Leiomyosarcoma is a rare tumor, occurring in fewer than two cases per 10 million non–HIV-infected children. However, a much higher than expected frequency of leiomyosarcoma has been reported in HIV-infected children. Visceral sites (eg, the lungs, spleen, and GI tract) are commonly involved.
Hleyhel M, Belot A, Bouvier AM, et al: Risk of AIDS-defining cancers among HIV-1-infected patients in France between 1992 and 2009: results from FHDH-ANRS CO4 cohort. Clin Infect Dis 57:1638–1647, 2013.
Polizotto MN, Uldrick TS, Hu D, Yarchoan R: Clinical manifestations of Kaposi sarcoma herpesvirus lytic activation: Multicentric Castleman disease (KSHV-MCD) and the KSHV inflammatory cytokine syndrome. Front Microbiol 3:73, 2012.
Strother RM, Gregory KM, Pastakia SD, et al: Retrospective analysis of the efficacy of gemcitabine for previously treated AIDS-associated Kaposi’s sarcoma in western Kenya. Oncology 78:5–11, 2010.
Uldrick TS, Wyvill KM, Kumar P, et al: Phase II study of bevacizumab in patients with HIV-associated Kaposi’s sarcoma receiving antiretroviral therapy. J Clin Oncol 30:1476–1483, 2012.
Unemori P, Leslie KS, Hunt PW, et al: Immunosenescense associated with presence of Kaposi’s sarcoma in antiretroviral treated HIV infection. AIDS 27:1735–1742, 2013.
Barta SK, Lee JY, Kaplan LD, et al: Pooled analysis of AIDS malignancy consortium trials evaluating rituximab plus CHOP or infusional EPOCH chemotherapy in HIV-associated non-Hodgkin lymphoma. Cancer 118:3977–3983, 2012.
Bayraktar UD, Ramos JC, Petrich A, et al: Outcome of patients with relapsed/refractory acquired immune deficiency syndrome-related lymphoma diagnosed 1999-2008 and treated with curative intent in the AIDS Malignancy Consortium. Leuk Lymphoma 53:2383–2389, 2012.
Cesarman E: Pathology of lymphoma in HIV. Curr Opin Oncol 25:487–494, 2013.
Gopal S, Patel MR, Yanik EL, et al: Temporal trends in presentation and survival for HIV-associated lymphoma in the antiretroviral era. J Natl Cancer Inst 105:1221–1229, 2013.
Levine AM, Noy A, Lee JY, et al: Pegylated liposomal doxorubicin, rituximab, cyclophosphamide, vincristine, and prednisone in AIDS-related lymphoma: AIDS Malignancy Consortium Study 047. J Clin Oncol 31:58–64, 2013.
Noy A, Lee JY, Cesarman E, et al: AMC 048: Modified CODOX-M/IVAC-rituximab is safe and effective for HIV-associated Burkitt lymphoma. Blood 126:160–166, 2015.
Adler DH, Kakinami, L, Modisenyane T et al: Increased regression and decreased incidence of HPV-related cervical lesions among HIV-infected women on HAART. AIDS 26:1645–1652, 2012.
Denny L, Adewole I, Anorlu R, et al: Human papillomavirus prevalence and type distribution in invasive cervical cancer in sub-Saharan Africa. Int J Cancer 134:1389–1398, 2014.
Denslow SA, Rositch AF, Firnhaber C, et al: Incidence and progression of cervical lesions in women with HIV: A systematic global review. Int J STD AIDS 25:163–177, 2014.
Fraunholz I, Rabeneck D, Gerstein J, et al: Concurrent chemoradiotherapy with 5-fluorouracil and mitomycin C for anal carcinoma: Are there differences between HIV-positive and HIV-negative patients in the era of highly active antiretroviral therapy? Radiother Oncol 98:99–104, 2011.
Martellotta F, Berretta M, Cacopardo B, et al: Clinical presentation and outcome of squamous cell carcinoma of the anus in HIV-infected patients in the HAART-era: A GICAT experience. Eur Rev Med Pharmacol Sci 16:1283–1291, 2012.
Mungo C, Cohen CR, Maloba M, et al: Prevalence, characteristics, and outcomes of HIV-positive women diagnosed with invasive cancer of the cervix in Kenya. Int J Gynaecol Obstet 123:231–235, 2013.
Robbins HA, Shiels MS, Pfeiffer RM, Engels EA: Epidemiologic contributions to recent cancer trends among HIV-infected people in the United States. AIDS 28:881–890, 2014.
Baeyens A, Slabbert JP, Willem P, et al: Chromosomal radiosensitivity of HIV positive individuals. Int J Radiat Biol 86:584–592, 2010.
Bearz A, Vaccher E, Martellotta F, et al: Lung cancer in HIV positive patients: The GICAT experience. Eur Rev Med Pharmacol Sci 18:500–508, 2014.
Deeken JF, Tjen-A-Looi A, Rudek MA, et al: The rising challenge of non-AIDS defining cancers in HIV-infected patients. Clin Infect Dis 55:1228–1235, 2012.
Hentrich M, Berger M, Wyen C, et al: Stage-adapted treatment of HIV-associated Hodgkin lymphoma: Results of a prospective multicenter study. J Clin Oncol 30:4117–4123, 2012.
Kaminuma T, Karasawa K, Hanyu N, et al: Acute adverse effects of radiation therapy on HIV-positive patients in Japan: study of 31 cases at Tokyo Metropolitan Komagome Hospital. J Radiat Res 51:749–753, 2010.
Lambert AA, Merlo CA, Kirk GD: Human immunodeficiency virus-associated lung malignancies. Clin Chest Med 34:255–272, 2013.
Montoto S, Shaw K, Okosun J, et al: HIV status does not influence outcome in patients with classical Hodgkin lymphoma treated with chemotherapy using doxorubicin, bleomycin, vinblastine, and dacarbazine in the highly active antiretroviral therapy era. J Clin Oncol 30:4111–4116, 2012.
Shiels MS, Pfeiffer RM, Gail MH, et al: Cancer burden in the HIV-infected population in the United States. J Natl Cancer Inst 103:753–762, 2011.