• lymphoma;
  • chemotherapy;
  • dose escalation;
  • hematology;
  • acquired immunodeficiency syndrome;
  • human immunodeficiency virus–associated non-Hodgkin lymphoma


  1. Top of page
  2. Abstract


Optimal treatment of human immunodeficiency virus (HIV)-associated non-Hodgkin lymphoma (NHL) has yet to be defined, because chemotherapy could exacerbate immunodeficiency, with subsequent adverse effects for patients.


The authors investigated the feasibility of an intensive chemotherapy regimen for HIV-associated NHL. Thirty-eight patients were treated with a first course of cyclophosphamide (Cy), vincristine, and prednisone; followed by 3 courses of high-dose Cy (2000 mg/m2), doxorubicin (Doxo; 50 mg/m2), vincristine, and prednisone (modified high-dose CHOP); 1 course of high-dose methotrexate (MTX; 8000 mg/m2); and 1 course of high-dose cytarabine (8000 mg/m2). Radiotherapy was added to the treatment regimen for patients with bulky disease or residual tumor. Chemotherapy was administered in conjunction with granulocyte–colony-stimulating factor and antiretroviral therapy.


Patients received 91.5%, 93%, 66%, and 63% of the scheduled doses of Cy, Doxo, MTX, and cytarabine, respectively. The complete response rate was 60.5%, with a total response rate of 79%. The 40-month overall survival rate was 43%, the disease-free survival rate was 65%, and the recurrence-free survival rate was 39%. Both an International Prognostic Index score of 0 or 1 and Burkitt-type histology had positive effects on survival, whereas CD4-positive lymphocyte counts, viral burden, and previous highly active antiretroviral therapy did not. CD4-positive T lymphocyte levels decreased from 0.197 ± 0.156 ×109/L before treatment to 0.152 ± 0.1 ×109/L at 6 months after the end of treatment. A decrease in viral load, from 380,000 ± 785,000 copies/mL before treatment to 25,000 ± 43,000 copies/mL at 6 months after the end of treatment, also was observed.


The results of the current study indicate that intensive chemotherapy is effective and tolerable for patients with HIV-associated NHL. Cancer 2004;100:667–76. © 2004 American Cancer Society.

Lymphomas occur with increased frequency in patients with congenital or acquired immunodeficiency.1 The use of highly active antiretroviral therapy (HAART) to treat human immunodeficiency virus (HIV) infection, together with the improved management and prophylaxis of opportunistic infections (OI), has been found to delay progression of HIV infection to acquired immunodeficiency syndrome (AIDS).2

Since the introduction of HAART, the incidence of systemic AIDS-associated non-Hodgkin lymphoma (NHL) has decreased,1, 3–9 although this decrease has been less dramatic than the corresponding decreases in the incidence rates of Kaposi sarcoma and primary central nervous system (CNS) NHL. AIDS-associated NHL is characterized by disseminated anatomic stages with visceral, bone marrow, and CNS involvement.10 Despite significant rates of response to therapy, the prognosis for patients with AIDS-associated NHL is poor, due to 1) advanced disease stage; 2) aggressive tumor behavior; 3) the increased hematologic toxicity associated with chemotherapy; and 4) immunosuppression, which leads to increased risk of severe OI.10 The evolution of the prognosis for patients with systemic NHL in the HAART era remains under debate; overall improvement has been observed by some investigators,8 but not by others.11

Many efficient chemotherapy regimens have been tested,12–23 but there remains a general lack of agreement regarding the optimal treatment for patients with AIDS-related NHL. The aggressive presentation of AIDS-related NHL suggests the need for intensive treatment regimens, which result in high complete response (CR) and survival rates for patients with non-HIV-associated NHL.24, 25 Nonetheless, the poor tolerability associated with chemotherapy has prompted investigators to test reduced-dose regimens. Dose reduction led to a decrease in CR rates and did not limit treatment-related toxicity in the study conducted by Tirelli et al.,14 whereas Kaplan et al.18 observed unchanged CR rates and decreased hematologic toxicity when reduced-dose chemotherapy was used. Ratner et al.26 recently demonstrated the efficiency and tolerability of both modified-dose mCHOP (cyclophosphamide [Cy] 375 mg/m2, doxorubicin [Doxo] 25 mg/m2, vincristine, and prednisone) and full-dose CHOP (Cy 750 mg/m2, Doxo 50 mg/m2, vincristine, and prednisone) chemotherapy. The observed CR rates were 30% for patients who received mCHOP and 48% for patients who received standard CHOP. The median duration of CR was 9 months for patients treated with mCHOP and had not yet been reached for patients treated with standard CHOP. Toxicity profiles did not differ between the two regimens. Compared with mCHOP administered without granulocyte–colony-stimulating factor (G-CSF), standard-dose CHOP administered with G-CSF reduced the incidence of neutropenia. Although the study conducted by Rattner et al.26 was a Phase II trial without randomization, the observed rates and durations of CR suggest that standard-dose CHOP holds some advantage over lower-dose regimens.

The four-drug CHOP regimen is as effective as newer regimens and is one of the standard treatment methods for patients with high-grade NHL.27, 28 We considered the intensive chemotherapy regimens used at our institution (Institut Paoli-Calmettes) to treat NHL in immunocompetent hosts,24, 25 together with the findings of Shipp et al.,29 in designing a ‘high-dose’ CHOP regimen for treating HIV-related NHL. In addition to intrathecal chemotherapy, high-dose single-drug chemotherapy (methotrexate [MTX] 8000 mg/m2 and high-dose cytarabine [ARA-C] 8000 mg/m2) was used to pass the blood-brain barrier and treat or prevent CNS localization, which is common in AIDS-related NHL.


  1. Top of page
  2. Abstract


From February 1997 to September 2001, 38 consecutive patients age > 18 years and < 65 years who had untreated, biopsy-confirmed NHL along with HIV infection were treated (after providing informed consent) with an intensive chemotherapy regimen. The study was approved by the local committee on human investigations and was in accordance with an assurance filed and approved by the local Department of Health and Human Services. All high-grade histologic subtypes except for leukemic forms of Burkitt lymphoma (defined by bone marrow involvement > 30%) were included in the current study. Among the exclusion criteria were the presence of evolutive OI at the time of diagnosis, heart failure (as evidenced by clinical signs, rather than by decreased ventricular ejection fraction only), renal insufficiency (creatinine > 2.5 times the normal level), and elevated bilirubin (> 2.5 times the normal level). We did not use the unfavorable prognostic factors described in the literature (poor performance status and CD4-positive T lymphocyte counts < 0.1 × 109 cells/L) to exclude patients.

Treatment Regimen


Six treatment cycles were delivered (Fig. 1). Treatment began with 1 cycle of COP (Cy 500 mg/m2 [30-minute infusion] on Day 1, vincristine [Oncovin; Eli Lilly, Indianapolis, IN] 2 mg bolus on Day 1, and prednisone 1 mg/kg on Days 1 and 2). One week after the start of the COP cycle, the first of 3 high-dose CHOP cycles (Cy 2000 mg/m2 [1-hour infusion], corresponding to a 167% increase relative to the 750 mg/m2 Cy dose used in conventional CHOP, on Day 1; vincristine 2 mg bolus on Day 1; Doxo 50 mg/m2 bolus on Day 1; and prednisone 1 mg/kg on Days 1 and 2) was administered; the next 2 CHOP cycles were initiated 3 weeks after the start of the preceding CHOP cycle. Three weeks after the start of the final CHOP cycle, a high-dose MTX cycle (8000 mg/m2 [23-hour infusion] on Day 1, with folinic acid rescue [50 mg every 6 hours for at least 72 hours and until serum MTX concentration was < 5 × 10−8 M] initiated 4 hours after the end of MTX) was administered. Next, 2 weeks after the start of the high-dose MTX cycle, a high-dose ARA-C cycle (2000 mg/m2 [1-hour perfusion] twice daily on Days 1 and 2) was initiated. Finally, radiotherapy (when performed) was administered 4 weeks after the start of the high-dose ARA-C cycle. Intrathecal chemotherapy, in the absence of cerebrospinal fluid involvement, consisted of MTX 15 mg and injectable methylprednisolone acetate (Depo-Medrol; Pfizer, New York, NY) 40 mg administered simultaneously before each chemotherapy course (6 intrathecal injections total). In the presence of cerebrospinal fluid involvement, intrathecal chemotherapy (the same regimen used in the absence of cerebrospinal fluid involvement, along with ARA-C 40 mg administered simultaneously) was administered 3 times weekly until the disappearance of NHL cells and then before each chemotherapy course (9–12 intrathecal injections total). This intrathecal regimen represents the standard treatment used at our institution for both involved and non-involved cerebrospinal fluid..24, 25, 30

thumbnail image

Figure 1. Schematic diagram of the intensive chemotherapy regimen used to treat human immunodeficiency virus–associated non-Hodgkin lymphoma. W: Week; COP: cyclophosphamide, vincristine, and prednisone; CHOP: cyclophosphamide, doxorubicin, vincristine, and prednisone; MTX: methotrexate; ARA-C: cytarabine; RadioTh: radiotherapy.

Download figure to PowerPoint

Patient management

Blood cell counts were determined three times weekly. The CD4-positive cell count was assessed before chemotherapy. Grade 4 neutropenia was defined by a neutrophil count < 0.5 × 109 cells/L. Grade 4 thrombocytopenia was defined by a platelet count < 20 × 109 cells/L. Packed red blood cells (RBCs) were transfused when hemoglobin levels decreased to < 80 g/L. Platelet concentrates were transfused when platelet counts decreased to < 10 × 109 cells/L, when a hemorrhagic syndrome was present, or when an invasive procedure was required. Blood products were irradiated at a dose of 25 grays (Gy). Conditions for the administration of chemotherapy included a platelet count > 90 × 109 cells/L, a polymorphonuclear neutrophil count > 1 × 109 cells/L, and resolution of extrahematologic toxicity. Chemotherapy was stopped for patients who did not meet these conditions on Day 35 after high-dose CHOP or on Day 28 after high-dose MTX. Patients who experienced partial correction of platelet counts (to > 50 × 109 cells/L) or late correction (between Days 28 and 35 after high-dose CHOP or between Days 21 and 28 after high-dose MTX) received chemotherapy that was reduced by 25%.

The antiretroviral regimens that patients were receiving at diagnosis usually were not modified during chemotherapy. For patients who were not receiving antiretroviral therapy at diagnosis, a three-agent regimen, typically involving stavudine, lamivudine, and indinavir, was initiated. Therefore, all patients, either before or at the start of chemotherapy, were receiving HAART. Regardless of the initial CD4-positive cell count, prophylaxis for Pneumocystis carinii was performed either with monthly pentamidine aerosols (Pentacarinat; Aventis Pharma, Strasbourg, France) or with cotrimoxazole, dapsone, or pyrimethamine. Prophylaxis for Mycobacterium avium complex infection was performed with either rifabutin or azithromycin if CD4-positive cell counts were < 0.1 × 109 cells/L. Prophylaxis for cytomegalovirus (CMV) infection usually was not performed, but an early determination of CMV infection status by systematic viremia detection and ocular examination (both done once before each chemotherapy cycle and then monthly for at least 6 months) was made if CD4-positive cell counts were < 0.1 × 109 cells/L. During aplasia or severe neutropenia, systematic oral decontamination with amphotericin B and sodium bicarbonate was initiated.

Hematopoietic growth factor support

G-CSF (filgrastim; 300 μg) was administered systematically on Day 6 after each chemotherapy cycle (except for the initial COP cycle) until neutrophil counts increased to > 0.5 × 109 cells/L for 2 consecutive days. G-CSF typically was administered after the first COP cycle only for patients with neutrophil counts < 0.5 × 109 cells/L.


Radiotherapy was used for patients with tumor size > 4 cm and for patients with residual tumor after the completion of chemotherapy. Radiotherapy was delivered to the involved areas with a linear accelerator (6–15 megavolts) at 1–2 months after the completion of the last chemotherapy cycle. Involved-field radiotherapy at a dose of 35–40 Gy (1.8 Gy daily for 5 days each week) represents the optimal schedule for radiotherapy when used exclusively.31 To minimize toxicity, we adopted a reduced-dose (20 Gy) schedule, in accordance with our institutional experience.24

Assessment of response

CR was assessed 1 month after the end of radiotherapy or, when no radiotherapy was performed, 1 month after the last chemotherapy cycle. CR was defined as the disappearance of all clinical evidence of disease along with normalization of laboratory values and radiologic results. The presence of a radiologic residual mass < 2 cm in size that remained stable for more than 2 months also qualified as a CR. Partial response (PR) was defined as a decrease of ≥ 50% in the size of all initial masses; no response (NR) was defined as a decrease of < 50% in any of these masses. Patients in the current study were examined at least once monthly throughout the follow-up period. To detect possible recurrences, computed tomography scanning was performed at 3-month intervals during the first year, at 6-month intervals during the second year, and once yearly afterward.

Immunologic and virologic assessment

Flow cytometric cell analysis was performed on an Epics XL flow cytometer (Beckman Coulter, Roissy, France) using fluoroisothiocyanate-labeled anti-CD4 and phycoerythrin-labeled anti-CD8 (Becton-Dickinson, San Jose, CA). The analyzer was gated on viable cells as identified based on forward versus side scatter; 5000 events were collected. In all cases, results were expressed after subtraction of the background, which was represented by isotype-matched fluorescent control probes (Beckman Coulter). Plasma HIV RNA titers were measured using the Amplicor Monitor HIV quantitative polymerase chain reaction kit32 (Roche Diagnostic Systems, Neuilly-sur-Seine, France; detection limit, 20 copies/mL).

Statistical Analysis

Overall survival (OS) and disease-free survival (DFS) were estimated using the Kaplan–Meier method. OS was defined as the interval between the first course of treatment and death or the most recent follow-up evaluation. For patients who experienced CR, DFS was defined as the interval between the last course of treatment and recurrence, death, or the most recent follow-up evaluation.


  1. Top of page
  2. Abstract

Patient Characteristics

Thirty-two men and 6 women were included in the current study (median age, 40 years; range, 29–63 years; number of patients age ≥ 50 years, 7). All referred patients were treated using our intensive regimen. HIV infection was caused by heterosexual contact (39.5%), intravenous drug use (29%), homosexual contact (18.5%), or blood transfusion (8%); in the remaining 5% of cases, the cause of infection was unknown. The mean CD4-positive lymphocyte count at lymphoma diagnosis was 0.197 × 109 ± 0.157 × 109 cells/L (median, 0.158 × 109 cells/L; range, 0.006–0.639 × 109 cells/L); 12 patients had CD4-positive cell counts < 0.1 × 109 cells/L (data not shown). The mean viral load at diagnosis was 380,000 ± 785,000 copies/mL. All patients had a viral load of > 500 copies/mL at diagnosis, with 15 patients having a viral load of > 100,000 copies/mL (data not shown). Twenty-two patients had Centers for Disease Control Stage A HIV infection, 8 patients had Stage B disease, and 8 patients had Stage C disease. Ten patients had a history of OI. At diagnosis, the majority of patients were being treated with HAART (58%), although some patients were being treated with non-HAART antiretroviral therapy (13%) or had received no therapy (29%).

Three main histologic types of NHL (Table 1) were represented in the study population: Burkitt and Burkitt-like lymphoma (31%), diffuse large cell lymphoma (31%), and immunoblastic lymphoma (24%). We investigated lymphoma-linked prognostic factors such as age-adjusted International Prognostic Index (IPI) score,33 performance status ≥ 2, lactate dehydrogenase (LDH) levels greater than normal, and Ann Arbor Stage III or IV disease. Eleven patients had IPI scores of 0 or 1, and 27 patients had IPI scores of 2 or 3. Three patients had meningeal involvement at diagnosis (data not shown).

Table 1. Lymphoma: Baseline and Follow-Up Characteristics
CharacteristicNo. of cases (%)
  1. N: normal maximum value; IPI: International Prognostic Index.

 Burkitt/Burkitt-like 13 (34)
 Diffuse large cell lymphoma 12 (31)
 Immunoblastic  9 (24)
 Diffuse, mixed cellularity  2 (5.5)
 Unclassified  2 (5.5)
Anatomic stage 
 I  3 (8)
 II  4 (10.5)
 III  3 (8)
 IV 28 (73.5)
Lactate dehydrogenase levels 
 ≤ 1N 13 (34)
 > 1N 25 (66)
Performance status 
 0/1 19 (50)
 2 15 (40)
 3/4  4 (10)
Bilirubin levels 
 > 2.5N  0
Creatinine levels 
 > 2.5N  0
IPI score 
 Low (0 risk factors)  4 (10.5)
 Low-intermediate (1 risk factor)  7 (18.5)
 High-intermediate (2 risk factors) 16 (42)
 High (3 risk factors) 11 (29)
 Complete 23 (60.5)
 Partial  7 (18.5)
 None  4 (10.5)
 Not evaluated  4 (10.5)
Median duration of follow-up (wks)166 (range, 24–266)
Disease-free survival rate (40 mos) 65%
Recurrence-free survival rate (40 mos) 39%

Regimen-Related Toxicities

Hematologic or hematology-related toxicities and transfusional support

Most patients (31 of 38) experienced at least 1 day of severe neutropenia (< 0.5 × 109 cells/L), and many (27 of 38) also experienced at least 1 day of fever. Fewer than half of all patients (15 of 38) required transfusion of platelet concentrates, whereas the majority of patients (27 of 38) required at least 1 transfusion of red blood cell concentrates.

Infectious complications

Documented bacterial infections included 7 cases of septicemia (Streptococcus α-hemolyticus [n = 1], Escherichia coli [n = 2], Salmonella enteritidis [n = 1], Enterobacter cloacae [n = 1], Staphylococcus haemolyticus [n = 1], or Pseudomonas aeruginosa [n = 1]), 1 case of suppuration at the site of adenopathy biopsy (P. aeruginosa), and 1 fatal pneumopathy (P. aeruginosa). Three patients had severe documented viral infections: two of these patients experienced herpesvirus infection (herpes zoster and herpetic meningitis in one patient each), and the third had fatal cytomegaloviral septicemia. Esophageal candidiasis was observed in one patient.

Extrahematologic complications

The most commonly observed complication was mucitis (Grade 1 [n = 3], Grade 2 [n = 13], Grade 3 [n = 2], or Grade 4 [n = 2]). The most serious complications observed were reversible acute renal failure after MTX administration (n = 1; Grade 3); severe bleeding in association with thrombopenia (n = 1; Grade 4); iatrogenic, MTX-related encephalitis (n = 1; Grade 2); and Cy-related hemorrhagic cystitis (n = 2; Grade 3). Other reported toxicities included vincristine-related peripheral neuropathy (n = 4; Grade 1 [n = 2] or Grade 2 [n = 2]), erythroderma following MTX administration (n = 1; Grade 1), severe constipation and/or abdominal pain (n = 3; Grade 2 [n = 2] or Grade 1 [n = 1]), hypokalemia (n = 1), and hypertension (n = 1).

Administration of Chemotherapy and Radiotherapy

Two hundred twenty-eight chemotherapy courses (38 patients × 6 courses per patient) were planned, and 200 courses (87.7%) were administered. Three patients withdrew from treatment due to adverse events (hematologic [n = 2] or neurologic [n = 1] toxicity), and 5 patients died during the chemotherapy period (3 due to lymphoma progression). Only six patients had dose modifications. We evaluated the administration of chemotherapeutic agents, because there is a dose-response relation between Cy or Doxo administration and NHL prognosis.34–37 The median cumulative doses received were as follows: Cy, 5947 ± 1369 mg/m2 over the 10 weeks of Cy-containing chemotherapy; Doxo, 139 ± 26 mg/m2 over the 9 weeks of Doxo-containing chemotherapy; MTX, 5263 ± 3584 mg/m2; and ARA-C, 5002 ± 3910 mg/m2. On an intent-to-treat basis, considering both dose modifications and discontinuation of therapy (due to withdrawal from the current therapeutic protocol or death), the administered doses represented 91.5% of the planned Cy doses, 93% of the planned Doxo doses, 66% of the planned MTX doses, and 63% of the planned ARA-C doses. The mean duration of chemotherapy was 15.6 ± 4.6 weeks, corresponding to dose intensities (from the first day of chemotherapy until its end) of 381 mg/m2 per week for Cy (planned dose, 542 mg/m2 per week) and 8.9 mg/m2 per week for Doxo (planned dose, 12.5 mg/m2 per week).

Twenty-two patients underwent radiotherapy—10 for residual adenopathy after chemotherapy and 12 for initial bulky disease. Most (18 of 22) received the planned dose of 20 Gy in 20 fractions, but 4 patients received additional doses (25, 40, 36, and 36 Gy, respectively) due to the persistence of adenopathy. For 1 patient, persistent hematologic toxicity delayed radiotherapy until > 2 months after the end of chemotherapy.

Response and Outcome

Six months after the end of treatment, the mean CD4-positive T lymphocyte count was 0.152 × 109 ± 0.1 × 109 cells/L, and the mean viral load was 25,000 ± 43,500 copies/mL. We observed 8 deaths during treatment (due to NHL progression [n = 3], sepsis [n = 2], or unknown causes [n = 3]). On an intent-to-treat basis, 23 patients (60.5%) achieved CR and 7 (18.5%) achieved PR, whereas 4 patients (10.5%) did not have responses; 4 patients (10.5%) were not evaluable. For 3 patients who had systemic symptoms putatively linked to lymphoma and residual lymph nodes < 2 cm in diameter, we performed gallium scintigraphy; findings were negative in all cases (data not shown). The mean follow-up period was 166 weeks (range, 24–266 weeks); 17 patients (45%) currently are alive, and all 17 are in the midst of a continuous first CR. Only 2 recurrences were observed; one was noted in the third month after the end of chemotherapy, and the other was noted 18 months after the end of therapy. Both patients with recurrent disease had non-Burkitt lymphoma. The recurrences did not respond to salvage therapy (dexamethasane, high dose Ara-C and cisplatin [DHAP] regimen), and these patients died shortly thereafter. All patients who did not experience responses died shortly after treatment, despite the initiation of salvage therapy (using the DHAP regimen for patients with non-Burkitt lymphoma and the cyclophosphamide, vincristine, prednisone, doxorubicin and methotrexate [COPADM] regimen for patients with Burkitt or Burkitt-like lymphoma). Patients who achieved PR also received salvage therapy, according to the same guidelines, when possible; no responses were observed, and all patients survived less than 6 months. Of the 23 patients who achieved CR, 6 died, 2 due to recurrent disease and 4 due to causes that were not related to lymphoma (HIV-related encephalopathy [n = 2], P. carinii infection [n = 1], or unknown causes [n = 1]).

The 40-month OS rate, calculated using the Kaplan–Meier technique, was 43%, and the 40-month DFS rate (i.e., the survival rate for patients who achieved CR) was 65% (Fig. 2). The recurrence-free survival (RFS) rate was 39%. We also analyzed the survival of patient subpopulations (Table 2). The 40-month survival rate for patients with IPI score 0 or 1 was 68%, significantly higher (two-sided, two-sample Mann–Whitney rank sum test; P < 0.05) than the rate of 27% observed among patients with IPI score 2 or 3. The survival rate for patients with Burkitt or Burkitt-like lymphoma (60% at 40 months) was significantly better (P < 0.05) than the rate for patients with non-Burkitt lymphoma (30% at 40 months). The 40-month survival rate for patients with CD4-positive T cell counts < 0.1 × 109 cells/L was less than the rate for patients with CD4-positive T cell counts > 0.1 × 109 cells/L (32% vs. 47%), but the difference was not statistically significant (P > 0.05). The survival rate for patients with a high viral load (≥ 100,000 copies/mL) at diagnosis (45% at 40 months) did not differ (P > 0.05) from the rate for patients with a lower viral load (42% at 40 months). Finally, no difference in survival (P > 0.05) was observed between patients who had received HAART before diagnosis of lymphoma (36% at 40 months) and patients who had not previously received HAART (44% at 40 months).

thumbnail image

Figure 2. Kaplan–Meier curves for overall and disease-free survival among patients who experienced complete responses to treatment.

Download figure to PowerPoint

Table 2. Analysis of Prognostic Factors
Prognostic factorSurvival rate at 40 mos (%)
  1. IPI: International Prognostic Index; HAART: highly active antiretroviral therapy; NHL: non-Hodgkin lymphoma.

IPI score 
 Burkitt type60
 Non-Burkitt type30
CD4-positive T lymphocytes (× 109 cells/L) 
 < 0.132
 > 0.147
Viral load at diagnosis (copies/mL) 
 > 100,00045
 ≤ 100,00042
HAART before NHL diagnosis 


  1. Top of page
  2. Abstract

Thirty-eight consecutive patients were treated using the current intensive chemotherapy regimen. The CR rate was 60.5%, with a total response rate of 79%. The 40-month OS rate was 43%, the DFS rate was 65%, and the RFS rate was 39%. All patients who did not have responses died shortly after treatment, despite the administration of salvage therapy. Patients who achieved PR did not experience responses to the change in chemotherapy, and all of these patients survived less than 6 months. Of the 23 patients who achieved CR, 2 died of recurrent disease, 2 died of HIV-related encephalopathy, 1 died of P. carinii infection, and 1 died of unknown causes. We observed a moderate decrease in CD4-positive T cell counts at the end of treatment, while a noteworthy decrease in HIV viral load also was observed.

In searching for prognostic factors, we first tested IPI score, which initially was developed for non-HIV-related NHL and is based on anatomic staging, performance status, and LDH levels. As in HIV-negative patients, the use of IPI score yielded three prognostic classes. While survival among patients with IPI score 0 or 1 could be considered satisfactory, survival among patients with IPI score 2 was markedly lower. In both of these groups, however, we observed a survival plateau, with some patients probably cured of their lymphoma. In sharp contrast, all patients with IPI score 3 died after a relatively short time, demonstrating the need for another treatment strategy for this group. Comparison of histologic types indicated a more favorable outcome for patients with Burkitt or Burkitt-like lymphoma; this finding is consistent with the preserved immune status observed in these patients. Nonetheless, the impressive results reported by Cortes et al.38 in patients with Burkitt lymphoma or leukemia suggest an alternative strategy for treating this subtype of lymphoma. It is noteworthy that the other three putative ‘immunologic’ prognostic factors that we tested (CD4-positive T cell counts, viral burden, and previous HAART) failed to delineate populations with different outcomes. This finding suggests that efficient prevention of OI at least transiently overcomes the deleterious effects of uncontrolled HIV infection. Regarding HAART, the current study suggests that its use before diagnosis of NHL is not a prognostic factor. In contrast, the use of HAART from the start of chemotherapy is pivotal; Cortes et al.38 reported that most patients who did not receive HAART during chemotherapy died, whereas most patients who did receive HAART were alive and experiencing CR at a median follow-up time of 29 months.

The current report demonstrates the feasibility of an outpatient sequential high-dose chemotherapy regimen for patients with systemic HIV-related NHL, as the actual relative dose intensity delivered was > 90% of the planned dose, with 87.5% of patients receiving > 80% of the projected dose intensity. The median CD4-positive T cell count in the study population indicates that some patients had well preserved immunity, which also is correlated with functional bone marrow. It was possible that patients with less well preserved immune function would not be able to tolerate the intensive regimen that was used; however, analysis of patients with very low CD4-positive T cell counts (< 1 × 109 cells/L) indicated that their outcome was similar to that of patients with higher CD4-positive cell counts (> 1 × 109 cells/L).

The finding that the treatment regimen was well observed can be attributed to the moderate toxicity of the regimen, particularly with respect to hematologic parameters. Although some patients had severe thrombocytopenia, which required platelet transfusion, no life-threatening hemorrhagic events occurred. Nonetheless, platelet counts were the primary cause of delays in chemotherapy administration. Red blood cell support was not necessary for all patients; the total required support was < 1 unit per chemotherapy cycle per patient. Recombinant erythropoietin could be useful in eliminating the need for RBC transfusions and increasing quality of life. An important issue in the current study was the usefulness of hematopoietic colony-stimulating factors (CSFs) in association with chemotherapy. As reviewed by Errante et al.,39 the hematologic toxicity of ‘standard’ or even ‘low-dose’ chemotherapy (using the World Health Organization criteria) without CSF support is high in AIDS-related NHL, with Grade 4 myelosuppression and febrile neutropenia observed in > 40% of patients. The majority of patients treated with the current intensive regimen experienced febrile neutropenia, and we documented seven cases of septicemia, with two deaths directly related to sepsis. Although the number of patients examined does not allow statistical comparisons, the incidence of infectious events and sepsis-related deaths in the current study compares favorably with the incidence data associated with less intensive regimens.39 Nonetheless, to reduce the number of infectious events and sepsis-related deaths, stem cell support could be considered, as recent publications have demonstrated the feasibility of such an approach.22, 40 Another noteworthy direction for further research involves attempting to improve antitumor activity without increasing toxicity, via the use of chemotherapy in conjunction with anti-CD20 monoclonal antibodies.41, 42

Another significant issue in the treatment of AIDS-related malignancies is the effect of chemotherapy and/or associated treatment on immune status, because 1) T lymphocytes are relatively sensitive to chemotherapy43; and 2) the use of CSF (either G-CSF20 or granulocyte-macrophage–CSF44) has been reported to contribute to the enhancement of viral replication. All patients in the current study received effective antiretroviral treatment during chemotherapy. We did not observe clinically significant short-term immunosuppression. We observed a moderate decrease in CD4-positive T cell counts, to 77% of pretherapy levels at 6 months after the completion of therapy, whereas viral load decreased by a factor of 15 during the same period. No conclusions can be drawn regarding the direct effect of chemotherapy on immune status, because other parameters (specifically, antiretroviral agents and observance of treatment) were modified during lymphoma treatment. Nonetheless, we cannot exclude the possibility that chemotherapy itself—at least when used in conjunction with effective antiretroviral therapy—had favorable effects in depleting infected cells, as suggested by recent data on cytostatic and antiretroviral drug combinations used in HIV treatment.45 Detailed flow cytometric analysis of a subset of patients in the current study suggests that our therapeutic procedure does not have a deleterious effect on the immune system.46

On an intent-to-treat basis, we observed OR (79%) and CR (60.5%) rates that compare favorably with the rates observed among HIV-negative patients at our institution who were treated with intensive chemotherapy regimens supported by blood stem cell infusion.24, 25 In their study, Stoppa et al.25 addressed the treatment of patients with high-grade NHL who had two or three adverse prognostic criteria according to the IPI. These patients were treated with three courses of modified CHOP chemotherapy followed by three additional cycles of the same regimen plus etoposide/cisplatin and stem cell reinjection. The observed CR rate was 65% (PR rate, 20%), compared with the 60.5% CR rate (PR rate, 18.5%) that we observed. Bouabdallah et al.24 reported a CR rate of 69%, with no PRs. It is noteworthy that in the studies conducted by Stoppa et al.25 and Bouabdallah et al.,24 the 3-year DFS rate (86% in the former study and 79% in the latter) was greater than it was in the current study (65%). This difference is not due to lymphoma recurrence, but rather to the underlying immune deficiency. Thus, the results of the current study are similar to those of other studies involving patients with non-AIDS-related NHL and comparable adverse prognostic factors, although the results of the current study are based on an intent-to-treat analysis and represent an essentially unselected patient population.

Some studies have compared the use of standard and reduced-dose chemotherapy for HIV-related NHL. In the randomized study conducted by Kaplan et al.,18 there was no difference in terms of CR rates between standard (CR rate, 42%) and reduced-dose (CR rate, 41%) methotrexate, bleomycin, doxorubicin, cyclophosphamide, vincristine, and dexamethasone (m-BACOD) therapy, with less toxicity in the reduced-dose group. In the nonrandomized study conducted by Ratner et al.,26 the CR rate was 30% in the reduced-dose CHOP group, compared with 48% in the full-dose group, with comparable toxicity in both groups and a relatively short duration of CR in the reduced-dose group; these findings did not support the use of reduced-dose CHOP.

Recently, Cortes et al.38 examined the use of the rather intensive hyper-CVAD regimen (hyperfractionated Cy, vincristine, Doxo, and dexamethasone) in the treatment of HIV-related Burkitt lymphoma and leukemia. Cortes et al. demonstrated the feasibility of hyper-CVAD, which exhibited a toxicity profile similar to the one observed in HIV-negative patients and which yielded impressive response rates (CR rate, 92%; PR rate, 8%) and long-term survival data. These findings support 1) the use of hyper-CVAD to treat all patients with HIV-related Burkitt lymphoma; and 2) the possible further intensification of chemotherapy regimens used to treat HIV-related non-Burkitt lymphoma.


  1. Top of page
  2. Abstract
  • 1
    Rabkin CS. Epidemiology of B-cell lymphomas. Curr Top Microbiol Immunol. 1999; 246: 235239.
  • 2
    Sepkowitz KA. AIDS—the first 20 years. N Engl J Med. 2001; 344: 17641772.
  • 3
    Grulich AE. AIDS-associated non-Hodgkin's lymphoma in the era of highly active antiretroviral therapy. J Acquir Immune Defic Syndr. 1999; 21 Suppl 1: 2730.
  • 4
    Buchbinder SP, Holmberg SD, Scheer S, Colfax G, O'Malley P, Vittinghoff E. Combination antiretroviral therapy and incidence of AIDS-related malignancies. J Acquir Immune Defic Syndr. 1999; 21 Suppl 1: 2326.
  • 5
    Sparano JA, Anand K, Desai J, Mitnick RJ, Kalkut GE, Hanau LH. Effect of highly active antiretroviral therapy on the incidence of HIV-associated malignancies at an urban medical center. J Acquir Immune Defic Syndr. 1999; 21 Suppl 1: 1822.
  • 6
    Jones JL, Hanson DL, Dworkin MS, Ward JW, Jaffe HW. Effect of antiretroviral therapy on recent trends in selected cancers among HIV-infected persons. Adult/Adolescent Spectrum of HIV Disease Project Group. J Acquir Immune Defic Syndr. 1999; 21 Suppl 1: 1117.
  • 7
    Chassagne-Clement C, Blay JY, Treilleux I, et al. [ Epidemiology of non-Hodgkin's lymphoma: recent data]. Bull Cancer. 1999; 86: 529536.
  • 8
    Besson C, Goubar A, Gabarre J, et al. Changes in AIDS-related lymphoma since the era of highly active antiretroviral therapy. Blood. 2001; 98: 23392344.
  • 9
    Jacobson LP, Yamashita TE, Detels R, et al. Impact of potent antiretroviral therapy on the incidence of Kaposi's sarcoma and non-Hodgkin's lymphomas among HIV-1-infected individuals. Multicenter AIDS Cohort Study. J Acquir Immune Defic Syndr. 1999; 21 Suppl 1: 3441.
  • 10
    Tirelli U, Errante D, Carbone A, Gloghini A, Vaccher E. Malignant lymphomas in patients with HIV infection. Leuk Lymphoma. 1996; 22: 245257.
  • 11
    Matthews GW, Bower M, Mandalia S, Powles T, Nelson MR, Gazzard BG. Changes in acquired immunodeficiency syndrome-related lymphoma since the introduction of highly active antiretroviral therapy. Blood. 2000; 96: 27302734.
  • 12
    Kaplan LD, Abrams DI, Geigal EU. AIDS-associated non-Hodgkin's lymphoma in San Francisco. JAMA. 1989; 261: 719724.
  • 13
    Kaplan L, Kahn J, Northfelt D, Abrams D, Volberding P. Novel combination chemotherapy for Hodgkin's disease (HD) in HIV-infected individuals [abstract]. Proc Am Soc Clin Oncol. 1991; 10: A7.
  • 14
    Tirelli U, Errante D, Oksenhendler E, et al. Prospective study with combined low-dose chemotherapy and zidovudine in 37 patients with poor-prognostic AIDS-related non-Hodgkin's lymphoma. French-Italian Cooperative Study Group. Ann Oncol. 1992; 3: 843847.
  • 15
    Gisselbrecht C, Oksenhendler E, Tirelli U, et al. Human immunodeficiency virus-related lymphoma treatment with intensive combination chemotherapy. French-Italian Cooperative Group. Am J Med. 1993; 95: 188196.
  • 16
    Remick SC, McSharry JJ, Wolf BC, et al. Novel oral combination chemotherapy in the treatment of intermediate-grade and high-grade AIDS-related non-Hodgkin's lymphoma. J Clin Oncol. 1993; 11: 16911702.
  • 17
    Aviles A, Nambo MJ, Halabe J. Treatment of acquired immunodeficiency syndrome-related lymphoma with a standard chemotherapy regimen. Ann Hematol. 1999; 10: 452455.
  • 18
    Kaplan LD, Straus DJ, Testa MA, et al. Low-dose compared with standard-dose m-BACOD chemotherapy for non-Hodgkin's lymphoma associated with human immunodeficiency virus infection. National Institute of Allergy and Infectious Diseases AIDS Clinical Trials Group. N Engl J Med. 1997; 336: 16411648.
  • 19
    Gill PS, Levine AM, Krailo M, et al. AIDS-related malignant lymphoma: results of prospective treatment trials. J Clin Oncol. 1987; 5: 13221328.
  • 20
    Sparano JA, Wiernik PH, Hu X, et al. Pilot trial of infusional cyclophosphamide, doxorubicin, and etoposide plus didanosine and filgrastim in patients with human immunodeficiency virus-associated non-Hodgkin's lymphoma. J Clin Oncol. 1996; 14: 30263035.
  • 21
    Campbell P, Iland H, Gibson J, Joshua D. Syngeneic stem cell transplantation for HIV-related lymphoma. Br J Haematol. 1999; 105: 795798.
  • 22
    Gabarre J, Leblond V, Sutton L, et al. Autologous bone marrow transplantation in relapsed HIV-related non-Hodgkin's lymphoma. Bone Marrow Transplant. 1996; 18: 11951197.
  • 23
    Holland HK, Saral R, Rossi JJ, et al. Allogeneic bone marrow transplantation, zidovudine, and human immunodeficiency virus type 1 (HIV-1) infection. Studies in a patient with non-Hodgkin lymphoma. Ann Intern Med. 1989; 111: 973981.
  • 24
    Bouabdallah R, Stoppa AM, Rossi JF, et al. Intensive sequential chemotherapy (ISC 95) with growth factors and blood stem cell support in high-intermediate and high-risk (IPI2 and IPI 3) aggressive non-Hodgkin's lymphoma: an oligocentric report on 42 patients. Leukemia. 1999; 13: 950956.
  • 25
    Stoppa AM, Bouabdallah R, Chabannon C, et al. Intensive sequential chemotherapy with repeated blood stem-cell support for untreated poor-prognostic non-Hodgkin's lymphoma. J Clin Oncol. 1997; 15: 17221729.
  • 26
    Ratner L, Lee J, Tang SC, et al. Chemotherapy for human immunodeficiency virus-associated non-Hodgkin's lymphoma in combination with highly active antiretroviral therapy. J Clin Oncol. 2001; 19: 21712178.
  • 27
    Gordon LI, Harrington D, Anderson J. Comparison of second generation combination chemotherapeutic regimen (m-BACOD) with a standard regimen (CHOP) for advanced diffuse non-Hodgkin's lymphoma. N Engl J Med. 1992; 327: 13421349.
  • 28
    Fisher RI, Gaynor ER, Dahlberg S. Comparison of a standard regimen (CHOP) with three intensive chemotherapy regimens for advanced non-Hodgkin's lymphoma. N Engl J Med. 1993; 328: 10021006.
  • 29
    Shipp MA, Neuberg D, Janicek M, Canellos GP, Shulman LN. High-dose CHOP as initial therapy for patients with poor-prognostic aggressive non-Hodgkin's lymphoma: a dose-finding pilot study. J Clin Oncol. 1999; 13: 29162923.
  • 30
    Bouabdallah R, Coso D, Costello R, et al. Role of high-dose therapy and initial response in survival of poor-risk patients with aggressive non-Hodgkin's lymphoma: a retrospective series on 126 patients from a single center. Bone Marrow Transplant. 2000; 25: 3540.
  • 31
    Bernstein JI, Coleman CN, Strickler JG, Dorfman RF, Rosenberg SA. Combined modality therapy for adults with small noncleaved cell lymphoma (Burkitt's and non-Burkitt's types). J Clin Oncol. 1986; 4: 847858.
  • 32
    Mulder J, McKinney N, Christopherson C, Sninsky J, Greenfield L, Kwok S. Rapid and simple PCR assay for quantitation of human immunodeficiency virus type 1 RNA in plasma: application to acute retroviral infection. J Clin Microbiol. 1994; 32: 292300.
  • 33
    The International Non-Hodgkin's Lymphoma Prognostic Factors Project. A predictive model for aggressive non-Hodgkin's lymphoma. N Engl J Med. 1993; 239: 987992.
  • 34
    Coiffier B, Gisselbrecht C, Herbrecht R, Tilly H, Bosly A, Brousse N. LNH-84 regimen: a multicenter study of intensive chemotherapy in 737 patients with aggressive malignant lymphoma. J Clin Oncol. 1989; 7: 10181026.
  • 35
    Kwak LW, Halpern J, Olshen RA. Prognostic significance of actual dose intensity in diffuse large cell lymphoma: results of a tree-structured survival analysis. J Clin Oncol. 1990; 8: 963977.
  • 36
    Collins C, Mortimer J, Livingston BB. High-dose cyclophosphamide in the treatment of refractory lymphoma and solid tumor malignancies. Cancer. 1989; 63: 228232.
  • 37
    Takvorian T, Canellos GP, Ritz J, et al. Prolonged disease-free survival after autologous bone marrow transplantation in patients with non-Hodgkin's lymphoma with a poor prognosis. N Engl J Med. 1987; 316: 14991505.
  • 38
    Cortes J, Thomas D, Rios A, et al. Hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone and highly active antiretroviral therapy for patients with acquired immunodeficiency syndrome-related Burkitt lymphoma/leukemia. Cancer. 2002; 94: 14921499.
  • 39
    Errante D, Vaccher E, Tirelli U. Are hematopoietic colony-stimulating factors useful in association with chemotherapy in the treatment of HIV-related non- Hodgkin's lymphomas? Ann Oncol. 1996; 7: 233237.
  • 40
    Gabarre J, Azar N, Autran B, Katlama C, Leblond V. High-dose therapy and autologous haematopoietic stem-cell transplantation for HIV-1-associated lymphoma. Lancet. 2000; 355: 10711072.
  • 41
    Spina M, Sparano JA, Jaeger U, Rossi G, Tirelli U. Rituximab and chemotherapy is highly effective in patients with CD20-positive non-Hodgkin's lymphoma and HIV infection. AIDS. 2003; 17: 137138.
  • 42
    Coiffier B, Haioun C, Ketterer N, et al. Rituximab (anti-CD20 monoclonal antibody) for the treatment of patients with relapsing or refractory aggressive lymphoma: a multicenter Phase II study. Blood. 1998; 92: 19271932.
  • 43
    Sparano JA, Wiernik PH, Strack M, et al. Infusional cyclophosphamide, doxorubicin and etoposide in HIV-related non-Hodgkin's lymphoma: a follow-up report of a highly active regimen. Leuk Lymphoma. 1994; 14: 263271.
  • 44
    Kaplan LD, Kahn JO, Crowe S, et al. Clinical and virologic effects of recombinant human granulocyte-macrophage colony-stimulating factor in patients receiving chemotherapy for human immunodeficiency virus-associated non-Hodgkin's lymphoma: results of a randomized trial. J Clin Oncol. 1991; 9: 929940.
  • 45
    Lori F, Jessen H, Lieberman J, Clerici M, Tinelli C, Lisziewicz J. Immune restoration by combination of a cytostatic drug (hydroxyurea) and anti-HIV drugs (didanosine and indinavir). AIDS Res Hum Retroviruses. 1999; 15: 619624.
  • 46
    Costello RT, Brunet C, Dignat-George F, Sampol J, Olive D, Gastaut JA. Immune reconstitution during intensive chemotherapy in patients with human immunodeficiency virus related non-Hodgkin lymphoma. Hematol J. 2002; 3: 216218.