26 - Human Immunodeficiency Virus | Handbook of Cancer Chemotherapy

Editors: Skeel, Roland T.

Title: Handbook of Cancer Chemotherapy, 7th Edition

> Table of Contents > Section III - Chemotherapy of Human Cancer > Chapter 26 - Human Immunodeficiency Virus

Chapter 26

Human Immunodeficiency Virus

Mary E. Cianfrocca

I. Introduction

Four malignancies are considered acquired immunodeficiency syndrome (AIDS) defining illnesses: Kaposi's sarcoma (KS), primary central nervous system (CNS) lymphoma (PCNSL), non Hodgkin's lymphoma (NHL), and cervical cancer. Although other malignancies appear to have an increased incidence in human immunodeficiency virus (HIV) infected patients (e.g., squamous cell anal cancer and Hodgkin's disease), their precise relationship to HIV infection has not yet been established. The management of any cancer in an HIV-infected patient requires an integrated team approach. Treatment of the underlying HIV infection must be incorporated into the overall treatment plan, in addition to aggressive prophylaxis and treatment of opportunistic infections, maintenance of general health and nutrition, and psychosocial support.

II. Kaposi's sarcoma (KS)

A. Epidemiology

KS is one of the most common HIV associated malignancies. The incidence of KS in both developed and developing countries rose steadily in the 1970s and 1980s, peaked around 1994, and declined dramatically since the introduction of highly active antiretroviral therapy (HAART). As an AIDS-defining illness, the incidence of KS has decreased from approximately 33% of patients in the early 1980s to 14% in the early 1990s to only 6.4% in 1998. In recent years, KS most often presents as a late, non AIDS-defining manifestation of HIV infection. In developing counties, however, the incidence of KS remains high. In some parts of Africa, KS is the most common cancer in men, accounting for up to 50% of all cancers in men. There is significant evidence that points to the importance of human herpes virus type 8 (HHV-8) in the pathogenesis of KS. HHV-8 DNA is identified in more than 90% of KS biopsies. Serologic studies demonstrate detectable antibodies to HHV-8 in a high percentage of KS patients. Furthermore, antibodies to HHV-8 have been associated with the subsequent development of KS.

B. Presentation and detection

The natural history of KS is extremely variable. KS most often presents as a pink or brown purple papule or plaque on the skin or mucous membranes. Lesions are frequently symmetric and follow Langer's lines. KS can present anywhere but has a predilection for the retroauricular areas, soles of the feet, extremities, genitalia, and face. In addition to their cosmetic unacceptability, KS lesions can cause significant morbidity and organ dysfunction. Dermal lymphatic involvement, most commonly involving the lower extremities, genitalia, and periorbital area, can cause painful and disfiguring lymphedema. The edema is often out of proportion to the visible skin involvement. Oral cavity lesions

P.601

occur in approximately 45%of patients with cutaneous KS, and the oral cavity is the first site of disease in 15% of patients. Oral lesions can interfere with speech and eating. Gastrointestinal involvement, present in up to half the number of patients, is often asymptomatic but can cause pain, diarrhea, and bleeding. Pulmonary involvement is the most common life-threatening manifestation of KS. It may be difficult to differentiate from opportunistic infection because the chest radiographmay show a reticulonodular or nodular infiltrate, with or without pleural effusions. Bronchoscopy is useful to visualize the characteristic erythematous plaque-like bronchial lesions of KS. However, biopsy is rarely done because of the risk of bleeding from these highly vascular tumors. Thallium and gallium scans may be useful to differentiate KS from an opportunistic pulmonary infection.

C. Staging

KS is a multicentric tumor and hence does not easily fit the usual TNM categorization. Because the overall prognosis may be more closely related to the degree of underlying immune dysfunction than to sites of disease involvement, the AIDS Clinical Trials Group (ACTG) developed a staging system that reflects various prognostic factors (Table 26.1). Patients are defined as good or poor risk on the basis of tumor burden, sites of involvement, CD4 lymphocyte count, history of opportunistic infections, systemic symptoms, and performance status. In the pre-HAART era, a study validating the ACTG staging system identified tumor burden as a significant predictive factor only in patients with a CD4 count of at least 200 cells/ L. The presence of systemic symptoms was not a predictor of outcome, and a CD4 count of 150 cells/ L, rather than 200 cells/ L, was a better discriminator of outcome. In

P.602

the post-HAART era however, a reassessment of the ACTG staging system was recently reported. Multivariate analysis indicated that tumor burden and presence of HIV-related systemic disease but not CD4 count were predictive of survival. In particular, analysis of the interaction between tumor stage and systemic disease identified two main risk categories with T1S1 patients having a significantly worse survival as compared to all other groups (T1S0, T0S1, T0S0). Furthermore, within the T1 risk category, pulmonary involvement was associated with a significantly worse survival as compared with the other T1 sites.

Table 26.1. AIDS Clinical Trials Group staging for epidemic Kaposi's sarcoma^a

	Relative Risk
Disease status	Good risk (0) (all of the following)	Poor risk (1) (any of the following)
Tumor (T)	Confined to skin, minimal oral disease, or both	Edema; extensive oral ulcers; visceral and gastrointestinal disease
Immune status (I)^a	CD4 count 150/ L	CD4 count <150/ L
Systemic illness (S)	No prior opportunistic infection or thrush; no B symptoms	Prior opportunistic infection or thrush; B symptoms; performance status <70%; other HIV-related illness
B Symptoms, unexplained fever, weight loss >10%, night sweats, diarrhea >2 weeks.
^aModified by validation study.

The initial evaluation of a patient with KS should involve a careful physical examination, including a rectal examination with Hemoccult testing. The history should focus on the rate of progression of KS and lesion-associated symptoms (pain, edema, disfigurement), as well as the history of HIV treatment, HIV-related opportunistic infections, the rate of decline in CD4 lymphocyte counts, and viral load. Even though the lesions are characteristic, a biopsy should be done to exclude other cutaneous processes. A baseline chest radiograph is important to look for asymptomatic visceral disease. Computed tomography (CT) scans are not indicated unless the patient's symptoms suggest abdominal disease. Upper endoscopy and colonoscopy are not indicated in the absence of unexplained gastrointestinal bleeding or symptoms. Medications should be reviewed to ensure that patients are receiving appropriate HAART and prophylaxis for opportunistic infections as well as to identify potentially myelosuppressive drugs that may complicate the use of chemotherapeutic agents (e.g., trimethoprim sulfamethoxazole, sulfadiazine, and zidovudine).

D. Treatment

Effective treatment of KS is dependent on clear communication between the doctor and patient with regard to the risks and benefits of the proposed therapy, its interaction with HIV treatment, and the patient's overall expectations. KS is not a curable tumor. The first decision to be made is whether to initiate therapy. Indications to initiate treatment for KS include prevention of disease progression, cosmesis, palliation of symptoms, and visceral disease. In general, local therapies are used for minimal, primarily cosmetically disturbing KS or for patients with severe multisystem compromise who may not tolerate systemic treatment. Systemic therapy is indicated for bulky, rapidly progressing, symptomatic, or life-threatening disease (Table 26.2). The determination of appropriate therapy is based on an evaluation of both tumor and immune status. Regardless of the extent of KS, effective HIV suppression tends to improve both the response to KS treatment and the durability of the response.

Local therapy. Limited numbers of small lesions can be treated by radiotherapy, topical therapy, cryotherapy, or intralesional injection. Surgery is occasionally useful for an isolated pedunculated lesion. Cryotherapy with liquid nitrogen can lead to hypopigmentation, which may be cosmetically unacceptable to dark-skinned patients. Intralesional therapy with vinblastine or interferon (IFN) can be effective but is limited by the need for multiple injections (Table 26.3). After the injection of oral lesions,

P.603

patients typically slough the oralmucosa in 24 to 48 hours, for which narcotic analgesics should be empirically provided. Intralesional IFN- is occasionally effective even in patients who have failed systemic IFN. Topical alitretinoin has been approved for the local therapy of KS. The 0.1% retinoic acid gel is applied initially twice a day with escalation to three or four daily doses as tolerated. Overall

P.604

response rates for individual lesions range from 27% to 49%. The main adverse effect is local skin irritation and local pain. There is no evidence of systemic absorption. Radiation therapy is effective for local control of KS. Depending on the dose and schedule, radiation therapy results in the most satisfying cosmetic result and longest duration of benefit of all the available local interventions.

Table 26.2. Treatment guidelines for Kaposi's sarcoma^a

Disease Status of KS	HIV Disease Status	Treatment Options
Minimal cutaneous disease	CD4 count <200/ L; prior OI; B symptoms	Local therapy
Minimal cutaneous disease	CD4 count 200/ L; no prior OI; no B symptoms	Interferon and antivirals or local therapy
Isolated, cosmetically disturbing disease	Any	Local therapy
Extensive cutaneous disease	CD4 count <200/ L; prior OI; B symptoms	Chemotherapy
Extensive cutaneous disease	CD4 count 200/ L; no prior OI; no B symptoms	Interferon and antivirals or chemotherapy
Localized bulky or painful disease	Any	Radiation therapy or chemotherapy
Tumor-associated edema	Any	Chemotherapy
Symptomatic visceral disease	Any	Chemotherapy
HIV, human immunodeficiency virus; KS, Kaposi's sarcoma; OI, opportunistic infection.
^aBest antiviral therapy is always a component of KS therapy. Adapted with permission from Susan Krown SE. Highly active antiretroviral therapy in AIDS-associated Kaposi's sarcoma: implications for the design of therapeutic trials in patients with advanced, symptomatic Kaposi's sarcoma. J Clin Oncol 2004;22:399 402.

Table 26.3. Intralesional chemotherapy for Kaposi's sarcoma

Chemotherapy Regimen^a	Dosage
Vinblastine (0.2 mg/mL)	0.1 mL/0.5 cm of surface area of lesion (maximum 4 mL)
Interferon-	3-5 x 10⁶ U three times/week for 4 weeks
^aAppropriate local anesthesia should be given before injection.

Highly active antiretroviral therapy (HAART). Patients with HIV-associated KS should be treated with HAART whenever possible. HAART has been shown to prolong the time to treatment failure and improve survival in patients receiving chemotherapy for HIV-related KS. Furthermore, regression of KS after institution of HAART alone has been well-documented. At the present time, however, the evidence suggests that HAART-induced regression primarily occurs in patients with limited KS who were not previously treated with HAART.

Systemic therapy

Biologic response modifiers. IFN- is the best studied biologic response modifier. A clear-cut dose response relationship has not been established. Responses have been reported with doses ranging from 1 x 10⁶ U/day to 36 x 10⁶ U three times per week (Table 26.4 for recommended dosing). Immune function, as measured by CD4 lymphocyte counts, is the best predictor of response to IFN as a single agent. Patients with CD4 lymphocyte counts of more than 400 cells/ L have an overall response rate of 45%, whereas patients with CD4 lymphocyte counts of less than 100 cells/ L respond less than 10% of the time. Although it may take more than 8 weeks to respond, the responses are often durable, with a median response duration of 1 to 2 years. IFN is best prescribed in combination with antiretroviral agents. IFN plus azidothymidine (AZT) results in a response rate of greater than 40%, with responses seen even in patients with CD4 counts below 100 cells/ L. IFN has been evaluated in combination with protease inhibitor based antiretroviral therapy

P.605

in a recent phase I trial. Dose-limiting toxicities were neutropenia and malaise and the maximum tolerated IFN dose was 5 million IU/day. Most practitioners begin with antiretroviral therapy and relatively low-dose IFN- , 1 to 5 x 106 U SC daily.

Table 26.4. Selected systemic treatment regimens for Kaposi's sarcoma

Regimen	Dosage
Biologic response modifier
Interferon-	1 10 x 10⁶ U/day SC
Chemotherapy
Liposomal doxorubicin (Doxil)	20 mg/m² IV every 3 weeks
Liposomal daunorubicin (DaunoXome)	40 mg/m² IV every 2 weeks
Paclitaxel	100 mg/m² IV every 2 weeks, or 135 mg/m² IV every 3 weeks

Chemotherapy. Chemotherapy provides rapid palliation of KS-related symptoms for most patients (see Table 26.4). Liposomal anthracyclines (e.g., liposomal daunorubicin [DaunoXome], liposomal doxorubicin [Doxil]) are currently considered the chemotherapeutic agents of choice for advanced KS. Multiple randomized trials of bleomycin and vincristine, with or without doxorubicin (Adriamycin), compared with one of the liposomal anthracyclines have consistently demonstrated less toxicity and equal or better response rates and response durations with the liposomal anthracyclines. For second-line therapy, paclitaxel 100 mg/m² every 2 weeks or 135mg/m² every 3 weeks has produced response rates of 50% to 70% associated with significant palliation of tumor-related symptoms. Doses of paclitaxel may need to be reduced for patients who are also receiving HAART with drugs metabolized by the same cytochrome P-450 pathway.

Patients often require growth factor support between cycles. Granulocyte colony-stimulating factor 5 g/kg SC given on days 7 to 12 of a 14-day (or 21-day) treatment cycle is often sufficient to preserve adequate neutrophil counts to prevent infection and allow timely administration of therapy. An absolute neutrophil count of 750/ L is adequate to deliver therapy with liposomal anthracyclines.

The duration of treatment is variable. Treatment should be continued until the maximal response is obtained; however, intercurrent illness may interrupt therapy. Although lesions may progress after therapy is discontinued, an effective maintenance therapy has not yet been defined. Maximal viral suppression with highly active antiretroviral therapy (HAART) may suppress KS tumor regrowth after discontinuation of chemotherapy.

III. Non Hodgkin's lymphoma (NHL)

A. Background

HIV-related lymphomas have increased as a percentage of first AIDS-defining illness since the increased use of HAART. Studies to determine if the use of HAART has led to a decrease in the overall incidence of HIV-related lymphomas have been inconclusive. HIV-associated NHLs are high-grade B-cell lymphomas similar to those seen in other immunocompromised patients. In comparison to NHL in the general population, HIV-associated NHLs are more likely to have advanced disease with bone marrow involvement and leptomeningeal disease and B symptoms. Primary CNS lymphoma accounts for less than 15% of the new AIDS-associated NHL, occurs in patients with severe immunodeficiency, and is universally associated with Epstein-Barr virus (EBV).

B. Presentation and detection

The most common presentations for HIV-associated NHL are constitutional symptoms (fevers, night sweats, and weight loss) or an enlarging mass.

P.606

Extranodal presentations, although once common, have become less frequent since the introduction of HAART. Nodal presentations occur in 60% of patients, and the remainder present with extranodal disease, most commonly involving the gastrointestinal tract, bone marrow, and liver. Very unusual sites of NHL have been seen, including the ear lobe, heart, and bile ducts. Seventy-five percent of patients present with advanced disease (stage III or IV).

C. Staging

The Ann Arbor staging classification is commonly used to stage HIV-related NHL (see Chapter 23); however, the correlation between stage and prognosis is weaker with Hiv cases. In the pre-HAART era, treatment of patients with HIV-related NHL was frequently complicated by severe hematologic toxicity and increased infectious mortality. Complete remissions (CRs) were uncommon and median survival was less than 6 months. In a pre-HAART study, four important adverse prognostic factors were identified for HIV-associated NHL: CD4 lymphocyte counts of less than 100 cells/ L, age more than 35 years, stage III or IV, and use of IV drugs. The presence of three or four of these indicators was associated with a median survival of only 18 weeks as compared with 46 weeks for patients with one or none of these indicators. Similarly, in a retrospective review of 60 patients, two prognostic subgroups were identified on the basis of immune function, performance status, and bone marrow involvement (Table 26.5). Use of HAART, however, has been associated with improved tumor response and survival.

Table 26.5. Prognostic stratification for HIV-associated non Hodgkin's lymphoma

From Levine	From Strauss
Good prognosis: median survival =	11.3 months Good prognosis: Presence of 0 1 indicators, median survival of 46 weeks.
No prior AIDS diagnosis, and
Karnofsky performance status >70%, and
No bone marrow involvement
Poor prognosis: median	Poor prognosis: Presence of 3 4
survival = 4.0 months	indicators, median survival of
Prior AIDS diagnosis, or	18 weeks
Karnofsky performance	Prognostic factors:
status <70%, or	Age <35
Bone marrow involvement	CD4 <100 Stage III or IV Use of IV drugs
HIV, human immunodeficiency virus; AIDS, acquired immunodeficiency syndrome. From Levine AM, Sullivan-Halley J, Pike MC, et al. Human immunodeficiency virus-related lymphoma: prognostic factors predictive of survival. Cancer 1991;68:2466 2472; and Strauss, with permission.

P.607

Complete staging evaluation should include CT scans of the head, chest, abdomen, and pelvis; bilateral bone marrow biopsies; and lumbar puncture with cerebrospinal fluid (CSF) sent for protein and cytologic evaluation. Before the availability of HAART, 40% of patients had CNS involvement; therefore, all patients should undergo CSF evaluation regardless of clinical stage. The incidence of CNS involvement at presentation in the HAART era is unknown but appears to be decreasing.

D. Treatment

General approach. The best therapy for HIV associated NHL remains to be defined. As is the case for KS, aggressive treatment of the underlying HIV infection is an important component of treatment. Fortunately, many clinical trials are ongoing to improve our understanding and treatment of this disease. Therapy must be tailored to the overall condition of each individual patient.

Systemic therapy. Before the availability of HAART, patients with AIDS-related NHL had a high incidence of opportunistic infections, limited bone marrow reserve, and poor outcomes with dose-intensive treatment. Because of this, the ACTG compared standard m-BACOD (Table 26.6) with low-dose (50% doses of cyclophosphamide and doxorubicin) m-BACOD in patients with HIV-associated NHL. There was no difference in the complete response rate or median survival between the two groups, but the low-dose group had significantly less hematologic toxicity. This led to the recommendation of low-dose treatment, particularly for patients with CD4 counts less than 200 cells/ L.

In the HAART era, standard chemotherapy should be recommended for patients with well-controlled HIV infection. The optimal treatment recommendation for patients with poorly controlled HIV infection is unclear. French/Italian randomized studies have shown that for patients with good performance status and relatively preserved immune function, treatment with standard-dose CHOP is feasible and leads to a complete response (59%) and event-free survival rates (33% at 2 years) significantly higher than those obtained with low-dose CHOP.

On the basis of data suggesting that longer exposure of tumor cells to drugs may enhance drug efficacy, a 4-day infusional regimen of cyclophosphamide, etoposide, and doxorubicin (CDE) has been evaluated in HIV-associated NHL. Phase II studies demonstrated a CR rate (46%) similar to that obtained with standard regimens but with prolongation of the median time to progression to 17 months and a near doubling of the 1-year survival rate (48%). Randomized trials comparing infusional versus bolus chemotherapy are underway.

While HIV suppression is consistently associated with improved overall survival for patients with HIV-associated malignancies, it is not clear that HAART and NHL therapy must be delivered concurrently. A possible down-side of chemotherapy plus HAART is the potential for enhanced toxicity. Protease inhibitors interfere with the excretion of drugs metabolized through hepatic cytochrome P-450,

P.608

and indinavir, in particular, decreases cyclophosphamide clearance by 40%. Recommendations for dose-modifications based on these pharmacologic interactions are not yet available. Saquinavir has been associated with significant mucositis in patients receiving myelosuppressive chemotherapy.

Table 26.6. Selected chemotherapy regimens for HIV-associated non Hodgkin's lymphoma

Low-dose m-BACOD
Bleomycin	4 mg/m² IV on day 1
Doxorubicin	25 mg/m² IV on day 1
Cyclophosphamide	300 mg/m² IV on day 1
Vincristine	1.4 mg/m² IV on day 1 (not to exceed 2 mg total)
Dexamethasone	3 mg/m² PO on days 1 5
Methotrexate	500 mg/m² IV on day 15
Leucovorin	25 mg PO every 6 h for four doses, on day 16 beginning exactly 24 h after methotrexate
CHOP
Cyclophosphamide	750 mg/m² IV on day 1
Doxorubicin	50 mg/m² IV on day 1
Vincristine	1.4 mg/m² IV on day 1, not to exceed 2 mg total
Prednisone	100 mg/m² PO on days 1 5
CDE chemotherapy
Cyclophosphamide	200 mg/m²/24 h CIV on days 1 4
Doxorubicin	12.5 mg/m²/24 h CIV on days 1 4
Etoposide	60 mg/m²/24 h CIV on days 1 4
Infusional EPOCH regimen (repeated q3wk x 6 cycles)
Etoposide	50 mg/m²/day CIV x 4 days
Vincristine	0.4 mg/m²/day CIV x 4 days
Doxorubicin	10 mg/m²/day CIV x 4 days
Cyclophosphamide	187 mg/m² IV on day 5 for CD4+ count <100, or 375 mg/m²IV on day 5 for CD4+ count 100
Prednisone	60 mg/m² PO on days 1 5
G-CSF	Starting on day 6
HIV, human immunodeficiency virus; G-CSF, granulocyte colony-stimulating factor; CIV, continuous IV infusion.

Little and colleagues treated patients with infusional EPOCH treatment (see Table 26.6) and withheld HAART until the completion of chemotherapy. The observed CR rate was 77% with an overall survival of 74%, after approximately 30 months median follow-up. The median survival has not yet been reached. Resumption of HAART after

P.609

chemotherapy leads to rapid repopulation of CD4+ cells and suppression of HIV viral load within 3 months.

The role of rituximab in the treatment of HIV-related NHL is controversial. The AIDS Malignancy Consortium (AMC) conducted a phase III trial randomizing patients to standard-dose CHOP versus CHOP with rituximab. Although the CR rate was higher in the R-CHOP arm (57% vs. 47%), this did not reach statistical significance. Death due to lymphoma was also lower in the R-CHOP arm. There were more deaths from infection, however, in the R-CHOP arm (15% vs. 2%; p = 0.027). Interpretation of these results is complicated by the fact that more patients in the R-CHOP arm had a CD4 count of less than 50 cells/mm³ and of the 15 patients on the R-CHOP arm who died of infection, 9 had a CD4 count less than 50 cells/mm³. Since more than 60% of the infectious deaths overall occurred in patients with CD4 counts of less than 50 cells/mm³, it raises the question of whether the higher infectious mortality on the R-CHOP arm was due to the severe immunosuppression of the patients on that arm. Furthermore, other trials have confirmed the high CR rate with rituximab-based chemotherapy without additional infectious complications. A current AMC trial is evaluating EPOCH with concurrent versus sequential rituximab.

For patients with well-controlled HIV infection, who are not eligible for a randomized clinical trial, standard dose CHOP, infusional EPOCH, or infusional CDE plus or minus rituximab are reasonable therapeutic options (see Table 26.6). Careful observation for increased or unusual toxicities is essential. Growth factor support is frequently needed. Aggressive prophylaxis against opportunistic infections is important. Regardless of CD4 lymphocyte count, all patients should receive prophylaxis for Pneumocystis carinii pneumonia during the treatment of a high-grade lymphoma (Table 26.7).

Central nervous system (CNS) prophylaxis/ treatment. CNS involvement at presentation occurs in 10% to 20% of patients with HIV-related NHL. The presence of EBV in the primary tumor is an independent predictor for CNS involvement either at diagnosis or during relapse. CNS prophylaxis is recommended for patients with small noncleaved cell tumors or epidural, bone marrow, paraspinal, paranasal sinus or testicular involvement. CNS prophylaxis should include four weekly treatments of either preservative-free methotrexate (12 mg) or

P.610

cytosine arabinoside (50 mg). In patients with documented meningeal disease, intrathecal therapy should be given three times per week until the CSF clears, then weekly for 8 weeks and monthly for 10 months. An Ommaya reservoir should be placed to facilitate therapy.

Table 26.7. Prophylaxis regimens for Pneumocystis carinii pneumonia

Trimethoprim sulfamethoxazole (Bactrim DS) 1 tablet PO, Monday, Wednesday, Friday, or

Dapsone 100 mg PO daily^a

^aExclude glucose-6-phosphatase deficiency by quantitative spectrophotometry in high-risk patients before initiating therapy.

Burkitt's lymphoma (BL). While survival of patients with HIV-associated diffuse large cell lymphoma has improved in the HAART era, survival of HIV-associated BL remains poor. At present, patients with HIV-associated BL are not treated differently than HIV-positive patients with diffuse large cell lymphoma. However, there is data supporting the use of more intensive chemotherapy regimens such as those used in HIV-negative patients in this population.
Follow-up. In this era of cost containment, repeat staging studies are often controversial. Despite this zeal to minimize testing, documentation of a complete response as early as clinically warranted is important to minimize the duration of therapy in this already-immunocompromised group of patients. Treatment should be continued for a minimum of four cycles or for two cycles after attainment of complete response.
Salvage therapy. Some patients who have a relapse after initial treatment have been successfully treated with an alternative NHL regimen. Suitable patients should be considered for high-dose chemotherapy with autologous stem cell transplantation.

IV. Primary central nervous system lymphoma (PCNSL)

A. Background

The incidence of PCNSL has declined since the introduction of HAART. Unlike systemic HIV-associated NHL, which can occur at earlier stages of HIV infection, PCNSL typically occurs in profoundly immunocompromised patients with CD4 lymphocyte count less than 50 cells/ L. The Epstein-Barr virus (EBV) genome is identified in virtually all investigated cases of HIV-associated PCNSL. This supports the belief that EBV may have a direct etiologic role in the development of this disease.

B. Presentation and detection

The diagnosis of PCNSL is often difficult to make. Most patients present with a focal neurologic deficit. CT scan or magnetic resonance imaging (MRI) of the head typically shows single or multiple contrast-enhancing masses with surrounding edema. These lesions are often in a periventricular location and may be difficult to distinguish from those of toxoplasmosis. CSF cytology is rarely diagnostic, and stereotactic biopsy is often recommended to establish a tissue diagnosis. Because 95% of HIV-positive patients with toxoplasmosis have serologic evidence of Toxoplasma sp. infection, the Toxoplasma titer can be useful to determine a course of action.

If the Toxoplasma titer is negative, stereotactic biopsy should be considered.
If the Toxoplasma titer is positive and the patient is either clinically unstable or the clinician feels uncomfortable without a tissue diagnosis; based on the clinical scenario, stereotactic biopsy should be considered.

P.611

If, however, the Toxoplasma titer is positive and the patient is clinically stable, a 1- to 2-week trial of empiric therapy for a presumptive diagnosis of toxoplasmosis may be appropriate. If there is no evidence of clinical or radiologic improvement or if there is evidence of clinical decompensation, stereotactic biopsy should be considered.

This approach requires close and frequent monitoring for signs of neurologic deterioration or progression. A newer alternative to the above is the use of EBV polymerase chain reaction (PCR) in the spinal fluid as a diagnostic clue. The presence of EBV by PCR in the CSF is a sensitive and highly specific (80% and 100%, respectively, in a recent study) marker for primary PCNSL. Therefore, in the context of either a suggestive MRI or CT of the brain and a positive thallium scan, EBV-detection in the CSF may be considered highly suggestive of PCNSL in the absence of a tissue diagnosis.

C. Treatment

Whole-brain radiation therapy is the standard therapy. Temporary control and improvement of neurologic deficits occur in 70% of patients. A range of radiation doses (2,000 to 6,000 cGy) has been used. In one retrospective review, survival was found to be a function of performance status, not the total radiation dose administered. Median survival with radiation is only 2 to 5 months. Recent data suggests that the addition of HAART to radiation therapy improves survival. Furthermore, tumor regression with HAART alone has been reported. The role of chemotherapy or chemoradiation for HIV-related PCNSL is not well established. Treatment with combined-modality therapy (CHOP plus radiation therapy) based on the experience in non HIV-related PCNSLs was evaluated and did not demonstrate superior results to radiation therapy alone. High-dose methotrexate, with or without intrathecal chemotherapy, has been reported to lead to prolongation of median survival to as much as 40 months; however, studies done to date have been very small. Small clinical trials evaluating preradiation chemotherapy with CHOD/BVAM (cyclophosphamide, doxorubicin, vincristine, and dexamethasone alternated with BCNU, vincristine, ara-C, and methotrexate) have reported prolonged survival but with significant toxicity including dementia and leukoencephalopathy.

V. Cervical cancer

A. Background

It was not until 1993 that cervical cancer became an AIDS-defining illness. As in non HIV-infected women, the development of cervical squamous carcinoma has been directly linked to prior infection with human papilloma virus (HPV). Although both HIV and HPV are sexually transmitted diseases, other immunosuppressed populations also have an increased incidence of HPV infection, suggesting that immune competence is a deterrent to the development of HPV coinfection. The presence of HPV, especially types 16, 18, and 31, is correlated with a higher incidence of cervical intraepithelial neoplasia (CIN).

B. Presentation and detection

Presentation is similar to that of non HIV-related cervical cancer, except that the disease is often more aggressive in HIV-infected patients and advanced disease is more common. Routine Papanicolaou's

P.612

(Pap) smears may not be sensitive enough to detect this aggressive neoplasm. It has been recommended that colposcopy, not Pap smears, be the standard for following up HIV-infected women at high risk of CIN. Because CIN may progress at a faster rate in HIV-infected women, annual screening may not detect curable disease. Women with rapidly progressing CIN should probably be tested for HIV. The effect of HAART on CIN has been examined and the findings have been mixed, with some studies showing no effect and others suggesting modest regression (12.5% regression in one study).

C. Staging

The International Federation of Gynecology and Obstetrics (FIGO) staging system, used for non HIV-infected patients, is used in this population as well (see Chapter 11). Stage for stage, however, the prognosis appears worse in HIV infected women owing to their compromised immune function.

D. Treatment

In a small study, women with CD4 lymphocyte counts of less than 500 cells/ L did significantly worse than women with intact immune function. Until these results are confirmed in larger trials, HIV-positive women with invasive cervical cancer should be treated in the same manner as women without HIV infection.

VI. Other malignancies

Hodgkin's lymphoma, testicular carcinoma, and anal cancers are also seen frequently in HIV-infected patients. Anecdotally, these malignancies appear more aggressive than their usual counterpart in immunocompetent patients, but no large trial has yet confirmed either increased virulence or increased incidence. It is interesting to speculate that EBV may play a role in the development of Hodgkin's disease. The role of HPV in the development of anal cancers has been well described. We recommend treatment of these malignancies with standard therapy. However, close attention should be paid to the prophylaxis and treatment of opportunistic pathogens as well as to the appropriate use of antiretroviral agents.