HIV-related lymphomas: Treatment of systemic lymphoma

INTRODUCTION — Human immunodeficiency virus (HIV) infection results in impaired cellular immunity, a condition known to predispose persons to develop neoplasms. As the lifespan of people living with HIV (PLWH) has increased, malignancies cause substantial morbidity and mortality in this population. Certain non-Hodgkin lymphomas (NHLs; systemic high-grade B cell NHL, primary central nervous system [CNS] lymphoma), along with Kaposi sarcoma and invasive cervical carcinoma, constitute acquired immune deficiency syndrome (AIDS)-defining malignancies. Other types of lymphoma (eg, Hodgkin lymphoma, follicular lymphoma) and other cancers in PLWH are not considered AIDS-defining. HIV-related NHL can be divided into three general categories based on location:

●Systemic NHL

●Primary CNS lymphoma

●Primary effusion (or body cavity) lymphoma

Systemic NHL accounts for most HIV-related lymphomas. By contrast, primary CNS lymphoma, in which disease is limited to the CNS, accounts for approximately 10 percent of HIV-related lymphomas, while primary effusion lymphoma (peritoneal, pleural, and/or pericardial lymphoma without an identifiable tumor mass) is much less common.

Treatment of systemic NHL in PLWH infection may be complicated by the immunocompromised state and need to treat HIV concurrently.

Treatment of HIV-related systemic aggressive B cell lymphoma and Hodgkin lymphoma are reviewed in this topic.

Related topics include:

●(See "HIV-related lymphomas: Epidemiology, risk factors, and pathobiology".)

●(See "HIV-related lymphomas: Clinical manifestations and diagnosis".)

●(See "HIV-related lymphomas: Primary central nervous system lymphoma".)

●(See "Primary effusion lymphoma".)

●(See "HIV infection and malignancy: Epidemiology and pathogenesis".)

OVERVIEW — Approximately 70 to 90 percent of the lymphomas encountered in people living with HIV (PLWH) infection are clinically aggressive diffuse large B cell lymphomas (DLBCLs) or highly aggressive Burkitt-like lymphomas [1]. At least 80 percent of these patients with systemic lymphoma have stage IV disease at the time of presentation (table 1) [2], with the common extranodal sites of involvement being the gastrointestinal tract, liver, lung, bone marrow, and the central nervous system. (See "HIV-related lymphomas: Epidemiology, risk factors, and pathobiology" and "HIV-related lymphomas: Clinical manifestations and diagnosis".)

Introduction of antiretroviral therapy (ART) in the late 1990s was associated with an improved tolerance for chemotherapy in PLWH. Clinical outcomes in PLWH and lymphomas approach those in the general population.

There are factors associated with HIV infection that make the management of PLWH and lymphomas considerably more complex:

●Historically, doses of chemotherapy were restricted by the baseline cellular immunodeficiency and limited bone marrow reserve of patients with advanced HIV infection [3]. However, the standard concomitant administration of ART may improve bone marrow function by decreasing the adverse effects of the HIV infection on hematopoiesis, thereby allowing for full dosing of chemotherapy [4].

●Chemotherapy produces a significant and sustained reduction in CD4 cell counts and an increased risk (twofold in one study) of opportunistic infection [5,6]. The immunosuppression induced by chemotherapy exceeds the duration of exposure to these drugs by many months [6].

●ART is a key component of the treatment of lymphomas in PLWH. Opportunistic infections are less common in this setting now that the underlying HIV infection can be better controlled with ART. (See 'Incorporating antiretroviral therapy' below.)

●Medications taken to control the HIV infection may cause myelosuppression and other adverse effects that overlap with chemotherapy. (See "Selecting antiretroviral regimens for treatment-naïve persons with HIV-1: General approach".)

PRETREATMENT EVALUATION — To best treat patients with non-Hodgkin lymphoma (NHL), the initial evaluation must establish the precise histologic subtype, the extent and sites of disease, and the performance status of the patient. General approaches to the diagnostic work-up and staging of NHL are presented separately. (See "Clinical presentation and initial evaluation of non-Hodgkin lymphoma" and "Pretreatment evaluation and staging of non-Hodgkin lymphomas".)

The pretreatment evaluation both determines the extent of the disease and provides information about the individual's comorbidities that are likely to have an impact on treatment options. In addition to a history and physical examination, it is our practice to perform the following pretreatment studies in people with HIV-related systemic NHL:

●Laboratory studies include a complete blood count with differential, chemistries with liver and renal function and electrolytes, lactate dehydrogenase, uric acid, phosphate, CD4 count, viral load, hepatitis B and C screening, and pregnancy test (in women). (See "Hepatitis B virus reactivation associated with immunosuppressive therapy".)

●Unilateral bone marrow biopsy is recommended for all patients.

●Lumbar puncture is indicated in a subset of patients. Cerebrospinal fluid should be sent for both cytology and flow cytometry. (See 'Central nervous system-directed therapy in aggressive B cell lymphoma' below.)

●A study of cardiac ejection fraction (eg, echocardiogram or multigated acquisition) should be performed for most patients and is necessary prior to regimens that include highly cardiotoxic drugs (eg, anthracyclines).

●A positron emission tomography (PET) with computed tomography (CT) should be performed, which provides critical information on the measurement of disease prior to treatment and aids in staging [7]. When compared with the HIV-positive population, false-positive results may be more likely in a patient with a high viral load [8]. (See "Pretreatment evaluation and staging of non-Hodgkin lymphomas", section on 'Imaging'.)

INITIAL TREATMENT — Optimal initial therapy for HIV-related lymphomas is not defined. Antiretroviral therapy (ART) is started or modified (if already begun) to control the HIV infection and allow for the administration of chemotherapy and/or radiation therapy. As in the HIV-seronegative population, the choice of therapy is principally determined by the subtype of non-Hodgkin lymphoma (NHL) and the stage of disease. Modifications are made based on the degree of immunosuppression from HIV, as measured by the CD4 count.

As in the HIV-seronegative population, attention needs to be given to serum uric acid levels and other potential complications of the tumor lysis syndrome at the time of diagnosis and with the first cycle of cytotoxic chemotherapy [9]. (See "Tumor lysis syndrome: Pathogenesis, clinical manifestations, definition, etiology and risk factors".)

Incorporating antiretroviral therapy — The reduction in HIV viral load and improvement in immune function associated with ART is expected to result in better tolerance of chemotherapy, fewer opportunistic infections, and improvement in the overall treatment outcome. These expectations have largely been borne out in retrospective and prospective clinical data collected since the advent of ART. ART is an essential component to the treatment of people with HIV-related lymphoma. ART should be continued during chemotherapy, with an undetectable blood viral load being the targeted goal [10].

Since the incorporation of ART, there has been a decrease in the number of AIDS-defining opportunistic infections that develop during or shortly after completing chemotherapy. As an example, approximately 20 percent of patients treated with chemotherapy for HIV-related NHL in a large prospective trial conducted in the pre-ART era developed AIDS-defining opportunistic infections [11]. By contrast, opportunistic infections have been infrequently reported in later clinical trials in which ART has been administered with chemotherapy [12-17].

Concomitant administration of ART during chemotherapy for HIV-related lymphomas also appears to improve overall survival (OS) [12-23]. Improvements in immune function rendered by ART diminish the effects of HIV on survival, allowing the prognostic features of the lymphoma to become relatively more important to OS [24,25]. In studies conducted in the ART era, advanced disease, as determined by stage or International Prognostic Index score (IPI) (table 2), appears to have greater prognostic value than factors associated with HIV disease, such as immune function and prior opportunistic infection. Most clinical trials, however, continue to identify a CD4 count <100/microL as a negative prognostic finding [26,27].

●A pooled analysis of individual patient data from 1546 patients from 19 prospective clinical trials performed between 1989 and 2010 reported that the concurrent use of ART with chemotherapy was associated with superior complete remission rates (odds ratio 1.89; 95% CI 1.21-2.93) and a trend towards improved OS (hazard ratio [HR] 0.78; 95% CI 0.60-1.02, p = 0.07) in multivariate analysis [23].

●A retrospective study evaluated patients with HIV-related lymphoma treated either with one of two combinations [21]:

•Cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) chemotherapy plus ART (CHOP-ART, 24 patients)

•CHOP or a CHOP-like therapy with no ART (80 patients)

At a median follow-up of 7 to 8.5 months, the incidence of AIDS-defining opportunistic infection was significantly lower in the CHOP-ART group (18 versus 52 percent), and OS was significantly longer (median survival not reached versus seven months).

●A German study demonstrated treatment outcomes using CHOP and concurrent ART that were similar to those with diffuse large B cell lymphoma (DLBCL) in the general population, with a complete response (CR) rate of 79 percent and median survival not reached after a median of 47 months follow-up [12]. Although these results are comparable to those observed in nonimmunocompromised patients, it should be pointed out that this patient population had a higher median CD4 count (223/microL) and lower IPI score than other studies.

The response to ART has been shown to be an independent risk factor for survival in patients with HIV-related NHL [19,28]. In a study of 44 patients from two Italian treatment centers, virologic response to ART was the only variable independently associated with attainment of complete remission following chemotherapy [29]. One-year OS for the entire group was 49 percent, increasing to 78 percent for virologic responders and 84 percent among those who achieved immunologic restoration. Other studies support the importance of ART as a prognostic factor with respect to CR rate and event-free survival of those treated with CHOP chemotherapy [28]. (See 'Effect of antiretroviral therapy' below.)

With few exceptions, studies have shown that concomitant ART and antineoplastic chemotherapy have no clinically adverse effects on the metabolism of drugs in either the antiretroviral or conventional-dose antineoplastic regimens [10,30,31]. There are few data to support one antiretroviral regimen over another. However, the use of zidovudine is strongly discouraged due to the risk of overlapping myelotoxicity.

Studies that have evaluated interactions between protease inhibitors and chemotherapy have had mixed results. One study has shown no statistically significant difference between ART regimens containing protease inhibitors versus those not containing protease inhibitors with respect to response rates or survival [32]. However, protease inhibitor-containing regimens may potentiate the myelotoxicity of the chemotherapy regimen compared with antiretroviral regimens that do not contain a protease inhibitor. This is likely due to CYP3A inhibition by specific protease inhibitors. Even at lower doses, ritonavir is a potent CYP3A inhibitor that can be used to "boost" levels of other protease inhibitors. Although use of these agents concomitantly with conventional-dose chemotherapy is not contraindicated, alternatives might be considered, especially if more aggressive or high-dose chemotherapy regimens are to be utilized. The use of integrase inhibitor-based regimens (eg, raltegravir, dolutegravir) represents a good alternative as these agents do not have hepatic metabolism, and thus, the potential for drug interactions may be lessened [33]. These ART regimens are preferred at our institution for use in combination with most chemotherapy regimens, including high-dose chemotherapy.

The CD4 count may decline by at least 50 percent from the baseline during chemotherapy for HIV-related NHL [34], necessitating the use of prophylaxis for opportunistic infections in all patients. Although this effect on the CD4 counts may exceed the duration of the chemotherapy by several months, those patients receiving ART concurrent with chemotherapy reconstitute CD4 counts more rapidly [34]. In one study, the cohort of 20 patients with HIV-related NHL treated with chemotherapy and concomitant ART recovered CD4 and natural killer (NK) cell counts to baseline within one month after the completion of chemotherapy [34]. By contrast, 12 months were required for recovery following treatment with etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin (EPOCH) without concomitant ART [26]. The combination of ART with chemotherapy maintains the virologic response in the majority of patients, while the development of resistance mutations in HIV does not seem to be increased [35].

Prophylaxis for opportunistic infections — During treatment with combination chemotherapy, ART should be continued along with prophylaxis for Pneumocystis jirovecii pneumonia (previously Pneumocystis carinii pneumonia). Mycobacterium avium complex (MAC) prophylaxis may be appropriate for selected patients with severe immunocompromise (ie, CD4 <50/microL). Antibiotic prophylaxis for enteric organisms during neutropenia is strongly encouraged. Given the high incidence of recurrent herpes simplex, herpes zoster, and Candida infections in this population, many clinicians also advise instituting antiviral and antifungal prophylaxis. (See "Overview of prevention of opportunistic infections in patients with HIV" and "Prophylaxis of infection during chemotherapy-induced neutropenia in high-risk adults" and "Prophylaxis of invasive fungal infections in adults with hematologic malignancies".)

Standard chemotherapy — Historically, first-line chemotherapy regimens that are effective for lymphomas of similar histology in the general population have been associated with lower rates and durability of CR in people living with HIV (PLWH) [36]. This may be related, in part, to increased expression of the multidrug-resistance gene (MDR-1) in the lymphocytes of HIV-related lymphomas compared with the lymphocytes in the HIV-seronegative population [37]. This gene codes for the P-glycoprotein on the cell membrane, a transporter protein that promotes efflux from the cytoplasm of a wide variety of drugs and other molecules [38].

As mentioned above, the vast majority of patients with HIV-related NHL have advanced-stage DLBCL or Burkitt lymphoma/leukemia (BL) [39]. The following sections investigate these NHL subtypes in more detail.

Diffuse large B cell lymphoma — The treatment of DLBCL in the HIV-seronegative population is based on the stage of disease at diagnosis. Limited-stage disease is often treated with the combination of chemotherapy (eg, CHOP), immunotherapy (eg, directed at CD20), and involved-field radiation therapy, while advanced-stage disease is commonly treated with chemotherapy plus immunotherapy. Most PLWH with DLBCL present with advanced-stage disease. In this setting, the best choice of chemotherapy regimen and the role of immunotherapy are less clear. (See "Initial treatment of advanced stage diffuse large B cell lymphoma".)

Two prospective clinical trials demonstrated that outcomes of DLBCL in PLWH in the ART era do not differ from the HIV-negative counterparts. Moreover, the risk of relapse is associated with an unclassifiable histology, stage III or IV, and no concomitant ART during chemotherapy [40,41].

Our approach is to modify our choice of initial chemotherapy and the decision to use rituximab based on prognostic factors and the CD4 count (see 'Role of rituximab' below):

●For most patients with advanced-stage HIV-associated DLBCL, we suggest combination chemotherapy with CHOP.

●For patients with CD20-positive disease and CD4 ≥50/microL, we suggest the addition of immunotherapy with rituximab (eg, rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone [R-CHOP]). The decision whether to use rituximab in a patient with a CD20-positive disease and a CD4 count <50 cells/microL must be individualized. For patients with a CD4 count over 50/microL who also have high growth fraction disease (Ki67 >80 percent) or plasmablastic histology, we suggest dose-adjusted EPOCH rather than CHOP, each in combination with rituximab for CD20-positive disease.

●Patients with limited-stage DLBCL are those with disease that can be contained within one radiation field. Such patients may be considered for abbreviated chemotherapy or chemoimmunotherapy (as determined by CD4 count) followed by involved-field radiation therapy.

CHOP — Historically, the "standard" chemotherapy for most cases of HIV-related DLBCL has been cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP). CHOP and CHOP-like regimens, which were primarily used in the pre-ART era, were associated with CR rates of 30 to 50 percent and two-year OS <20 percent [5,10,11,36]. When administered in combination with ART, CR rates with CHOP alone in two studies were 47 and 62 percent, with corresponding OS rates at two years in both studies of approximately 50 percent [24,42]. (See "Initial treatment of advanced stage diffuse large B cell lymphoma".)

Despite initial reports suggesting a benefit to the incorporation of pegylated liposomal doxorubicin into the CHOP regimen over standard doxorubicin [38,43], follow-up studies in the HIV-seronegative populations failed to confirm this benefit [44,45]. In addition, liposomal doxorubicin substitution for doxorubicin in R-CHOP in 40 patients with HIV-related DLBCL was associated with a complete remission rate of 47.5 percent [42], comparable to that observed in patients treated with standard R-CHOP.

Dose-adjusted EPOCH — Dose-adjusted infusional etoposide, vincristine, and doxorubicin, prednisone, cyclophosphamide (daEPOCH) (table 3) is an alternative chemotherapy regimen for patients with HIV-associated DLBCL. In this approach, vincristine, doxorubicin and etoposide are administered as a 96-hour continuous infusion, followed by bolus administration of cyclophosphamide. Doses are adjusted based on nadir neutrophil counts in the preceding treatment cycle.

A prospective single-arm trial evaluated the daEPOCH regimen in 39 patients with newly diagnosed HIV-related lymphoma (80 percent DLBCL) [26]. ART was deferred until the completion of chemotherapy. A CR was achieved in 74 percent. At a median follow-up of 53 months, the estimated rates of progression-free survival (PFS) and OS at five years were 73 and 60 percent, respectively. Five-year OS rates for patients with CD4 counts >100/microL and ≤100/microL were 87 and 16 percent, respectively.

A multicenter trial to confirm these results and to examine the incorporation of rituximab into this regimen is discussed below [46]. (See 'Dose-adjusted EPOCH plus rituximab' below.)

Role of rituximab — The addition of the recombinant anti-CD20 antibody rituximab has revolutionized the treatment of CD20-positive lymphomas in the HIV-negative population. However, the role of rituximab in HIV-related lymphoma is not as clearly defined, given the profound cellular and humoral immunodeficiency that this agent can cause. The combination of rituximab with chemotherapy may cause a temporary decrease in the CD4, CD8, and CD19 (B cell) cell populations but does not appear to change the natural killer cell population [47,48]. In addition, panhypogammaglobulinemia of varying magnitude and duration typically follows the use of R-CHOP chemotherapy, leading to a decrease in humoral immunity [49]. The loss of humoral immunity in addition to the underlying destruction of the cellular immune system by HIV may increase the risk of infection above that which is normally seen with either HIV infection or rituximab therapy alone. (See "Initial treatment of advanced stage diffuse large B cell lymphoma".)

In general, our use of rituximab for patients with HIV-related CD20-positive lymphoma (eg, DLBCL, BL) depends on the extent to which the underlying HIV is controlled, as manifest by the CD4 count. For most patients with HIV-related CD20-positive lymphoma, we suggest the addition of rituximab to the treatment regimen if the CD4 count is over 50/microL. This preference places some emphasis on the known benefit of rituximab in the HIV-negative population despite recognizing the more profound cellular and humoral immunodeficiency that this agent can cause.

The decision whether to use rituximab in a patient with a CD20-positive disease and a CD4 count <50 cells/microL must be individualized. For patients who have not been treated with ART previously, it is reasonable to begin chemotherapy with rituximab and ART with the expectation that viral load will be controlled. Such patients should be followed closely and given enteric antibiotic prophylaxis. On the other hand, for a patient on ART who has end-stage AIDS and is unlikely to have an increase in the CD4 count during subsequent therapy, we favor chemotherapy alone.

This approach is supported by a pooled analysis of individual patient data from 1546 patients in 19 prospective clinical trials that were performed between 1989 and 2010 and reported that treatment with rituximab was associated with superior complete remission rates (odds ratio 2.89; 95% CI 1.64-5.08), PFS (HR 0.50; 95% CI 0.34-0.72), and OS (HR 0.51; 95% CI 0.38-0.71) [23]. The use of rituximab was associated with significantly improved outcomes in patients with CD4 ≥50 cells/microL but not in patients with CD4 <50 cells/microL, despite the fact that patients who received rituximab had a lower median CD4 count (179 versus 334 cells/microL) than those not receiving rituximab; however, the number of patients with a CD4 count <50 cells/microL overall was 14 percent.

Randomized trials are not available to guide the choice of chemotherapy along with rituximab. The studies that have used rituximab with chemotherapy in the HIV-positive and HIV-negative populations have generally supported its addition to chemotherapy regimens for aggressive B cell lymphoma. However, this decision must take into consideration the patient's CD4 count. We use R-CHOP (table 4) in average-risk patients with CD20-positive HIV-related DLBCL. By contrast, we prefer daEPOCH plus concurrent rituximab in patients with higher-risk DLBCL, such as those with high growth fraction (>80 percent) or nongerminal center DLBCL. For patients with Burkitt histology, we favor CODOX-M/IVAC (cyclophosphamide, vincristine, doxorubicin, high-dose methotrexate alternating with ifosfamide, cytarabine, etoposide, intrathecal methotrexate; also called the Magrath regimen) (table 5) with the addition of rituximab. EPOCH plus concurrent rituximab is a possible alternative (table 3) [46].

Rituximab plus CHOP or CDE — Prospective trials have investigated the use of rituximab in combination with cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) or infusional cyclophosphamide, doxorubicin, and etoposide (CDE) for the treatment of HIV-related lymphoma, with mixed results.

●In results pooled from three phase 2 trials using rituximab plus infusional CDE chemotherapy in 74 patients with HIV-related lymphoma, complete remission was achieved in 70 percent, with two-year failure-free survival and OS rates of 59 and 64 percent, respectively [50]. Not all patients received ART therapy. Serious infections, including opportunistic infections, were noted in 31 percent, with 8 percent of the patients dying of infection.

●A phase 2 nonrandomized trial evaluated R-CHOP in a highly selected group of 61 patients with high-grade lymphoma of B cell origin and no more than one of the following adverse characteristics: CD4 count <100 cells/microL, prior AIDS, or an Eastern Cooperative Oncology Group (ECOG) performance status <2 [51]. Approximately one-half of the patients on this trial received ART prior to starting chemotherapy. Treatment was well tolerated with an estimated two-year OS rate of 75 percent. Only two patients developed sepsis, and one death was possibly related to both treatment and infection in a patient with progressive lymphoma.

●A nonrandomized phase 2 trial administered six cycles of R-CHOP plus ART to 81 people with HIV-related DLBCL [52,53]. Patients were eligible if they had an ECOG performance status <3 and had no active opportunistic infections. Granulocyte colony-stimulating factor (G-CSF) was given if severe neutropenia had been observed on the previous cycle. Prophylaxis for P. jirovecii pneumonia was standard. Sixty-nine percent achieved a CR. Three-year disease-free survival and OS rates were 77 versus 56 percent, respectively. Six patients died of infections during treatment.

●A multicenter German trial studied the use of standard R-CHOP with ART in 72 patients with HIV-related lymphoma and zero or one of the following risk factors: CD4 <50/microL, prior opportunistic infection, and/or an ECOG performance score >3 [12]. The rates of CR and three-year OS were 72 and 60 percent, respectively, and are comparable to outcomes seen in the HIV-negative population.

By contrast to these initially favorable results from uncontrolled trials, benefits from rituximab could not be shown in a subsequent randomized multicenter phase 3 trial conducted by the AIDS Malignancy Consortium (AMC 010) [27]. In this trial, 150 patients with HIV-related aggressive or highly aggressive CD20-positive NHL were randomly assigned in a 2:1 ratio to R-CHOP or CHOP. All patients received G-CSF, ART, and P. jirovecii pneumonia prophylaxis. Antibiotic prophylaxis for enteric organisms was not required. The median follow-up was 137 weeks. When compared with CHOP alone, R-CHOP followed by rituximab maintenance resulted in:

●A nonsignificant trend towards higher overall complete remission (58 versus 47 percent)

●A nonsignificant reduction in disease progression (8.1 versus 21.6 percent)

●A similar median OS (2.6 versus 2.1 years)

●A greater incidence of grade 4 neutropenia (62 versus 48 percent)

●A significantly increased incidence of death due to infection (14 versus 2 percent)

Sixty percent of infection-related deaths occurred in patients with CD4 counts <50 cells/microL. In an analysis that excluded patients with CD4 counts <50 cells/microL, death rates due to infection were not significantly different between those patients who received or did not receive rituximab. It is noteworthy, however, that this study was not powered to detect the level of difference in outcome that was observed in the original GELA trial of CHOP versus R-CHOP in nonimmunocompromised patients [54].

A greater incidence of grade 4 neutropenia (62 versus 48 percent) and a significantly increased incidence of death due to infection in the R-CHOP arm of this study [27] as well as in patients treated with infusional R-CDE [50] raises serious concerns regarding the safety of adding rituximab to standard-dose chemotherapy in HIV-positive patients with poor performance status and/or low CD4 counts [27,51,55].

Dose-adjusted EPOCH plus rituximab — As described above, dose-adjusted infusional etoposide, vincristine, and doxorubicin, prednisone, cyclophosphamide (daEPOCH) (table 3) may be preferred for patients with HIV-related DLBCL and certain markers of poor prognosis [56]. The combination of daEPOCH plus rituximab has also been studied in patients with HIV, with mixed results [46,57,58]. (See 'Dose-adjusted EPOCH' above.)

●A randomized phase 2 AMC study explored daEPOCH plus rituximab in combination with antibiotic prophylaxis for enteric organisms in 101 patients with HIV-related B cell NHL (71 percent DLBCL, 29 percent Burkitt-like) [46]. All patients received daEPOCH and were randomly assigned to rituximab given either concurrently, 375 mg/m² prior to starting each 21-day cycle of EPOCH, or sequentially (weekly for six weeks following completion of chemotherapy). Compared with sequential rituximab, concurrent rituximab was associated with a similar two-year OS and PFS, higher rate of CR (73 versus 55 percent), and comparable toxicity (primarily cytopenia- and infection-related).

●A phase 2 study of 33 patients with HIV-related lymphoma treated with short-course EPOCH plus dose-dense rituximab for a minimum of three cycles reported 84 percent five-year OS and 68 percent five-year PFS [57]. Median CD4 cell count was 208/microL (42 percent with CD4 count <100/microL). There were no treatment-related deaths or new opportunistic infections during treatment. Outcomes were not associated with IPI (table 2), but germinal center B cell (GCB) subtype was associated with better PFS than non-GCB subtype.

In a pooled analysis of individual patient data for 1546 patients in 19 prospective clinical trials in patients with HIV-associated aggressive B cell lymphomas, the addition of rituximab to chemotherapy and the use of infusional EPOCH were associated with improved OS in patients with DLBCL [23]. These observations along with favorable results from the above phase 2 trials have resulted in rituximab, etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin (R-EPOCH) becoming the standard of care for most patients with HIV-associated DLBCL at many institutions in the United States. However, the level of evidence supporting the use of R-EPOCH (phase 2 trials and large retrospective review) are less than robust. As described earlier in this section, there are phase 2 trials using R-CHOP that have also been associated with excellent outcomes.

Our approach is to modify our choice of initial chemotherapy and the decision to use rituximab based on prognostic factors and the CD4 count:

●For many patients with advanced-stage HIV-associated DLBCL, without high-risk factors, we suggest combination chemotherapy with R-CHOP.

●For patients with CD20-positive disease and a CD4 count over 50/microL, we suggest the addition of immunotherapy with rituximab (eg, R-CHOP or R-EPOCH). The decision of whether to use rituximab in a patient with a CD20-positive disease and a CD4 count <50 cells/microL must be individualized. For patients with a CD4 count >50/microL who also have high growth fraction disease (Ki67 >80 percent), non-GCB cell of origin, double-hit cytogenetics, double protein positive immunohistochemistry, or high IPI score, we suggest dose-adjusted R-EPOCH rather than CHOP, despite limited clinical data.

●Patients with limited-stage non-bulky DLBCL are those with a disease that can be contained within one radiation field. Such patients may be considered for abbreviated chemotherapy or chemoimmunotherapy (as determined by CD4 count) followed by involved-field radiation therapy. However, in view of continuous late recurrence rate with this approach to treatment in the SWOG 8736 trial [59], we favor four cycles of chemoimmunotherapy plus two additional treatments with rituximab for patients with limited disease and no IPI risk factors (based on FLYER trial criteria) [60]. The Lymphoma Study Association (LYSA) Group randomized patients who had receive four or six cycles of RCHOP-14 to post-chemo radiotherapy or observation [61]. There was no statistical benefit to the addition of radiotherapy compared with chemotherapy alone.

●Treatment with two cycles of rituximab plus daEPOCH (daEPOCH-R) after achieving CT-documented CR was associated with similar outcomes, whether CR was achieved after two initial cycles or after four initial cycles (ie, patients received four total cycles or six total cycles, respectively) [62]. Such response-adapted treatment should be confirmed in prospective trials.

Burkitt lymphoma/leukemia — There is no standard treatment for BL in the HIV-seronegative population. These lymphomas are known to be highly aggressive tumors characterized by dysregulation of the MYC oncogene [63]. They are poorly responsive to moderate-dose chemotherapy (eg, CHOP) that is used for other types of lymphomas [64]. This is true not only in PLWH [25,65] but also in the HIV-seronegative population. HIV-seronegative patients with BL are typically treated with intensive, short-duration combination chemotherapy with central nervous system (CNS) prophylaxis. The most commonly used regimens in the HIV-seronegative population are CODOX-M/IVAC (Magrath) (table 5) and CALGB-9251 protocol [66-68]. daEPOCH-R may be considered for PLWH and BL, provided that there is no evidence of CNS involvement. (See 'Initial treatment' above.)

There are limited data regarding the use of these regimens in the setting of HIV infection. A retrospective study of the CODOX-M/IVAC (Magrath) regimen in 14 adults with HIV-related BL reported a CR rate of 63 percent [65]. The AMC reported outcomes in 34 patients with HIV and BL enrolled in a prospective study of a modified R-CODOX-M/IVAC regimen (AMC 048) [69]. The modifications intensified the treatment of leptomeningeal disease if present at diagnosis and included measures to reduce mucositis and hematologic and neurologic toxicity. Most patients were able to receive the intended treatment, and 12 percent discontinued therapy early due to toxicity. Severe (grade 3/4) neurotoxicity and hematologic toxicity occurred in 21 and 59 percent, respectively. One-quarter of patients developed neutropenic fever, and there was one treatment-related death. Estimated PFS at one year was 69 percent, and OS at one and two years was 72 and 69 percent, respectively. Similar rates were seen in the 19 patients whose BL was confirmed on central pathologic review. The addition of rituximab did not appear to increase toxicity. The results of this study compare favorably with results in nonimmunocompromised patients treated with the same regimen in previous trials.

A randomized phase 2 study of dose-adjusted infusional chemotherapy with the EPOCH regimen plus rituximab conducted by the AMC included 27 patients with BL [46]. These patients did well with this regimen in conjunction with concurrent rituximab. However, further study of this regimen is required before definitive recommendations regarding its use in BL can be made. One concern is the lack of CNS-penetrating systemic chemotherapy in the daEPOCH regimen given the high risk of leptomeningeal involvement in PLWH and BL. (See 'Dose-adjusted EPOCH plus rituximab' above.)

A phase 2 study reported that risk-stratified daEPOCH-R-based treatment was associated with similar outcomes whether patients had low-risk or high-risk BL [70]. The study included 113 previously untreated patients with BL, including 28 who were HIV positive. Low-risk patients (ie, normal lactate dehydrogenase, ECOG performance status 0 to 1, stage I or II disease, and a maximum tumor size <7 cm) received three cycles of daEPOCH-R (with no intrathecal [IT] prophylaxis, unless positron emission tomography [PET]/CT after two cycles was positive), while all other patients (ie, high risk) received six cycles of daEPOCH-R plus IT prophylaxis. With a median follow-up of nearly five years, there was no difference in OS or event-free survival based on high-risk versus low-risk disease, HIV positive versus HIV negative, and age >40 years versus ≤40 years. Because only 10 percent of patients had CNS disease at presentation, there are concerns about routine use of daEPOCH-R (which does not contain any CNS-penetrating systemic agents), despite its lower toxicity compared with R-CODOX-M/IVAC.

CNS prophylaxis is a key component to the treatment of patients with BL. In the absence of CNS prophylaxis, up to 30 percent of patients with BL will develop leptomeningeal involvement/recurrence regardless of HIV status [63]. The more successful treatment regimens, like CODOX-M/IVAC, include the use of systemic CNS-penetrating chemotherapeutic agents and IT chemotherapy administration as an integral part of treatment. (See 'Central nervous system-directed therapy in aggressive B cell lymphoma' below.)

The CODOX-M/IVAC and CALGB-9251 protocols are complex, and there are possible interactions between the antineoplastic agents and the components of the ART regimen. ART is generally interrupted during the administration of high-dose methotrexate, but this decision should be individualized based on the specific ART combination used. During treatment, ART should be continued if tolerated; however, if significant gastrointestinal toxicity is observed and results in the inconsistent use of antivirals, ART therapy may be discontinued for the duration of chemotherapy and restarted once chemotherapy is completed.

Plasmablastic lymphoma — Plasmablastic lymphoma (PBL) is an uncommon but highly aggressive lymphoma observed in the setting of HIV disease [71].

PBL often arises in the oral cavity of PLWH and typically demonstrates Epstein-Barr virus (EBV) and a high proliferation index [72,73]. The malignant cells have large plasmablastic morphology and express CD138, MUM1, and BLIMP1, while infrequently expressing PAX5 and CD20 [72]. Overexpression of MYC and mutations affecting PRDM1 and the JAK/STAT3 and/or RAS-MAPK signaling pathways are often found [74,75]. PBL is generally associated with a poor prognosis [76,77].

Optimal treatment is uncertain, but we favor intensive treatment (like those used for BL) plus CNS prophylaxis because CHOP alone is an inadequate therapy. We individualize decisions to use autologous hematopoietic cell transplantation (HCT) in first CR (CR1), although data are limited.

Studies that reported outcomes with PBL include:

●Retrospective analysis of 10 patients with previously untreated PBL, including six who received daEPOCH-based regimens, reported 67 percent one-year OS and 70 percent CR [78].

●A randomized phase 2 AMC study of daEPOCH-R with or without vorinostat for aggressive HIV-associated lymphomas included 15 patients with PBL among the 90 enrolled patients [79]. There was no difference in one-, two-, or three-year OS or event-free survival between study arms, and the CR rate was 67 percent.

●Addition of bortezomib to daEPOCH (V-EPOCH) was associated with 65 percent five-year OS and 92 percent CR among 13 patients (including six PLWH) [80]. In another study, V-EPOCH was associated with 50 percent two-year OS and PFS and 88 percent CR [81].

●Among five PLWH with PBL, CHOP induction therapy followed by autologous HCT in CR1 reported sustained CR in all patients for 13 to 83 months post-transplantation [82]. Retrospective review of 24 patients with PBL who underwent autologous HCT in CR1 also reported favorable outcomes [83].

Central nervous system-directed therapy in aggressive B cell lymphoma — Patients with HIV-related NHL have an increased risk of CNS involvement at diagnosis or later in the course of their disease. The incidence of CNS involvement varies with NHL subtype.

We perform a diagnostic lumbar puncture with cytology and flow cytometry of cerebrospinal fluid in all patients with BL and in selected higher-risk patients with DLBCL [84]. Up to one-third of patients with BL have leptomeningeal involvement/recurrence regardless of HIV status [63]. A study of 176 patients with systemic HIV-related NHL reported 10 percent meningeal involvement at presentation [85]. We apply criteria for higher-risk disease in PLWH and DLBCL like those for the HIV-negative population, as discussed separately. (See "Secondary central nervous system lymphoma: Clinical features and diagnosis", section on 'Lumbar puncture'.)

Patients who have leptomeningeal involvement with systemic lymphoma should receive either IT methotrexate or cytarabine and/or high-dose systemic antineoplastic agents that cross the blood-brain barrier.

●Regimens used for BL, such as CODOX-M/IVAC, include systemic CNS-penetrating chemotherapeutic agents and IT chemotherapy. (See "Treatment of Burkitt leukemia/lymphoma in adults", section on 'CNS involvement'.)

●The most effective CNS prophylaxis for high-risk DLBCL patients is not established, but four treatments with IT methotrexate or cytarabine are commonly used. We consider the addition of two to three doses of high-dose intravenous methotrexate upon the completion of anthracycline-based therapy for patients with testicular involvement by DLBCL or other patients with high risk for parenchymal brain recurrence.

Hodgkin lymphoma — The optimal treatment for HIV-related Hodgkin lymphoma (HL) has not been established in randomized trials. Prior to the introduction of potent ART, treatment of HL was associated with significant toxicity, and outcomes were poor [86,87]. Despite the difficulties of intensive treatment in PLWH, contemporary regimens have given promising results in several studies [88-91].

Prospective and retrospective results in patients with HIV-related HL support the role of standard chemotherapy approaches in combination with ART:

●The most extensive results come from a German multicenter study in which 108 patients with HIV-related HL were treated using a stage- and risk-adapted approach [90]. The 23 patients with early stage, favorable HL were treated with two to four cycles of ABVD chemotherapy followed by involved-field radiation therapy. The 14 patients with early-stage, unfavorable HL were managed with either the more intensive bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, prednisone (BEACOPP) regimen or with four cycles of doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) plus radiation therapy. The 71 patients with advanced-stage HL were managed with six to eight cycles of BEACOPP. Overall, 102 of the 108 patients (94 percent) were maintained on ART during treatment for HL.

Using this stratified treatment approach, complete remission rates for early-stage favorable HL, early-stage unfavorable HL, and advanced-stage HL were 96, 100, and 86 percent, respectively. For the entire group, the two-year PFS rate was 92 percent, and the two-year OS rate was 91 percent after a median follow-up of 26 months.

●Similar conclusions were drawn from a retrospective study of 224 HL patients who were consecutively treated with ABVD between 1997 and 2010 in a series that included 93 PLWH [91]. ART was given concurrently to 92 of the 93 (99 percent) PLWH during chemotherapy. The patients who were living with HIV had more extensive disease and more adverse prognostic features than those who were HIV negative. However, OS and PFS following treatment with ABVD were not adversely affected by positive HIV status.

The British HIV Association suggests that all PLWH with HL should receive potent ART and a standard chemotherapy regimen (eg, ABVD) as initial treatment of advanced disease [92]. Brentuximab vedotin plus AVD (doxorubicin, vinblastine, dacarbazine), which is approved for the treatment of newly diagnosed advanced-stage HL, was reported to be safe and efficacious in PLWH, but the study excluded patients who require ART that includes strong CYP3A4 inhibitors [93].

PATIENT FOLLOW-UP — After the completion of the initially planned treatment, patients should be evaluated to determine the disease response to treatment and should be followed longitudinally for relapse.

Response evaluation — One month following the completion of planned therapy (or sooner if the outcome is unfavorable), the response to treatment should be documented by history, physical examination, and laboratory studies (complete blood count, lactate dehydrogenase [LDH], and biochemical profile). The post-treatment imaging study of choice is a combined positron emission tomography (PET) with CT scan, which provides information on the size and activity of residual masses and allows for the distinction between active disease and fibrosis. PET/CT should be obtained six to eight weeks after the completion of chemotherapy and 12 weeks after the completion of radiation therapy [94]. PET/CT imaging obtained earlier than this is likely to demonstrate increased uptake due to an inflammatory reaction to treatment.

Treatment response is determined by PET/CT (table 6).

Patients who fail to obtain a complete response are treated as refractory disease. (See 'Treatment of relapsed disease' below.)

Surveillance for relapse — Following the completion of therapy, restaging, and documentation of complete remission, patients are seen at periodic intervals to monitor for treatment complications and assess for possible relapse. The frequency and extent of these visits depend on the comfort of both the patient and clinician. There have been no prospective, randomized trials comparing various schedules of follow-up.

Our approach to patient surveillance is to schedule patient visits every one to two months during the first year, every two to three months during the second year, and every four to six months starting two years after complete response. At these visits, we perform a history and physical examination, complete blood count, chemistries, and LDH.

There is no role for routine PET or PET/CT imaging in the longitudinal follow-up of asymptomatic patients after response assessment. For those patients who achieve complete remission, follow-up CT scanning should be considered at 6 and 12 months post-treatment. Beyond one-year risk of recurrence is low, and scanning should be performed only as clinically indicated. Clinical follow-up is recommended every three to six months for five years.

Relapsed disease can be suggested by changes on imaging studies but can only be confirmed by biopsy. As such, a biopsy should always be obtained to document relapsed disease before proceeding to salvage therapy.

TREATMENT OF RELAPSED DISEASE

Second-line chemotherapy — Although prospective trials have not evaluated clinical outcomes in patients treated with second-line chemotherapy, it is usual practice to administer platinum-based salvage regimens commonly employed in the nonimmunocompromised population, such as ICE (ifosfamide, carboplatin, etoposide), ESHAP (etoposide, methylprednisolone, cytosine arabinoside, and cisplatin), DHAP (dexamethasone, high-dose cytarabine, and cisplatin), or GDP (gemcitabine, dexamethasone and cisplatin) for patients with relapsed or refractory diffuse large B cell lymphoma (DLBCL) and Hodgkin lymphoma (HL). Many of these regimens have been utilized for salvage therapy in the autologous transplant studies [95-98]. Outcomes are generally poor without the use of autologous hematopoietic cell transplantation (HCT) in patients with chemotherapy-responsive disease. (See 'Hematopoietic cell transplantation' below.)

A retrospective study from the AIDS Malignancy Consortium (AMC) reported the outcomes of 88 patients with primary refractory (61 percent) or relapsed HIV-related lymphoma diagnosed from 1999 to 2008 and treated with curative intent [99]. The most common histologic variants were DLBCL (60 percent), Burkitt lymphoma and variants (14 percent), HL (11 percent), plasmablastic lymphoma (6 percent), and T cell lymphoma (6 percent). The overall response rate to second-line therapy was 37 percent (24 percent complete), and the median survival was 38 weeks. At a median follow-up of 88 weeks, the estimated one- and two-year overall survival (OS) rates were 42 and 35 percent, respectively. The best outcomes were seen in patients with HL, those with relapsed rather than refractory disease, those with a complete or partial response to second-line therapy, and those who underwent high-dose chemotherapy and autologous stem cell support.

Platinum-based regimens — A retrospective study evaluated the use of ESHAP in 13 patients with relapsed or refractory HIV-related non-Hodgkin lymphoma (NHL) after at least one prior chemotherapy regimen [100]. Complete and partial responses were seen in four (31 percent) and three (23 percent) patients, respectively. The median survival was 7.1 months. Hematologic toxicity was substantial. Similar results were seen in a retrospective study of 48 patients with relapsed/refractory HIV-related NHL treated with GDP [98]. The overall objective response rate was 54 percent (95% CI 40-68 percent), with 10 complete responses (21 percent) and 16 partial responses (33 percent).

Novel treatment approaches — Novel treatment approaches that are under investigation include the use of the protease inhibitor bortezomib, histone deacetylase inhibitors, and the use of risk stratification to determine therapy. As examples:

●Bortezomib has established antitumor activity in a variety of lymphoid malignancies and has been shown both in vivo and in vitro to dramatically reduce transcription from the HIV-1 LTR-promoter [101]. In addition, proteasome inhibitors, including bortezomib, markedly inhibit HIV-1 replication in human peripheral blood mononuclear cells [102] and have been shown in vitro to reverse rituximab resistance and sensitize rituximab-resistant HIV-associated B cell NHL cell lines to chemotherapy-induced apoptosis [103]. A phase 1/2 trial of bortezomib in combination with rituximab and ICE is currently being conducted through the AMC.

●Histone deacetylase inhibitors may also have a potential role in the management of HIV lymphomas. Activity has been observed in some lymphomas, but there is evidence that histone deacetylase inhibitors may activate latent HIV gene expression [104,105].

●As in aggressive B cell lymphomas in nonimmunocompromised patients, current initial treatment strategies are focusing on risk stratification using immunophenotyping or gene expression profiling as well as HIV risk factors to identify patients who may have better outcomes with different therapies. One study is currently evaluating the use of rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) for higher-risk HIV-related large cell lymphoma while using rituximab, etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin (R-EPOCH) for patients with better control of HIV disease and those with non-germinal center B cell lymphomas.

●There is limited experience with the anti-CD30 drug conjugate brentuximab vedotin in patients with HIV-related lymphoma. One report described the attainment of durable complete responses following brentuximab vedotin in two adults with relapsed ALK-negative anaplastic large cell lymphoma (ALCL) and relapsed classic HL, respectively [106]. Brentuximab vedotin is approved for treatment of relapsed or refractory classic HL and ALCL. Its use in this setting is appropriate despite the lack of experience in people living with HIV (PLWH). The AMC is evaluating this agent in a front-line multicenter prospective trial in combination with doxorubicin, vincristine, and dacarbazine (NCT0177107).

There has been anecdotal experience with lenalidomide and bortezomib for relapsed HIV-associated plasmablastic lymphoma [107].

Hematopoietic cell transplantation — Autologous HCT is a treatment option for selected patients with relapsed HIV-related lymphoma demonstrating sensitivity to second-line chemotherapy. In contrast, the role of allogeneic HCT in HIV-related lymphoma is unknown given the paucity of published data.

Autologous transplantation — High-dose chemotherapy with autologous stem cell support is an effective therapy in patients with HIV-associated lymphomas. Patients with relapsed or refractory disease (HL or NHL) should be considered potential candidates for this treatment approach, ideally in the setting of a clinical trial. To be eligible, these patients should have a demonstrated response to antiretroviral therapy (ART), adequate organ function, chemosensitive disease, and no active cytomegalovirus (CMV) infection or other opportunistic infections.

Autologous HCT has been successfully used for the treatment of patients with relapsed or refractory HIV-associated lymphoma receiving ART in both prospective and retrospective studies [95,96,99,108-114]. In general, the incidence of bacteremia and neutropenic fever has been similar to that seen in the HIV-negative transplant setting. Regimen-related toxicity is not significantly increased when ART is combined with the high-dose chemotherapy and radiation therapy required for HCT.

The following trials illustrate the feasibility and efficacy of autologous HCT in this population:

●In a multicenter prospective, single-arm trial, 40 patients with relapsed or refractory HIV-associated aggressive systemic lymphoma (including DLBCL, plasmablastic lymphoma, Burkitt or Burkitt-like lymphoma or classic HL) underwent autologous HCT after conditioning with BEAM (BCNU, etoposide, cytarabine, melphalan) [114]. ART was held from the beginning of the preparative regimen until at least seven days after the completion of BEAM or resolution of gastrointestinal toxicity. At a median follow-up of 25 months, the estimated rates of overall and progression-free survival at two years were 82 and 80 percent, respectively. Importantly, when compared with an HIV-negative matched control population, HIV infection did not appear to impact mortality, lymphoma relapse, or other negative outcomes.

●In another multicenter prospective, single-arm trial, 50 patients with relapsed or refractory HL or NHL demonstrating HIV response to ART therapy received high-dose chemotherapy followed by autologous HCT [97]. Twenty-seven patients had chemosensitive disease, underwent successful stem cell mobilization, and proceeded to HCT using BEAM as the preparative regimen. At a median follow-up of 44 months, 21 of 27 transplanted patients achieved a continuous complete response, and six patients had disease progression or relapse after complete response. No patient died from transplant-related complications. There were four cases of CMV reactivation and six cases of herpes zoster infection.

●In the AMC trial 020, 20 of 27 patients enrolled underwent autologous HCT with median times to neutrophil and platelet engraftment of 11 and 13 days, respectively [112]. There was only one treatment-related death and no unusual infectious complications. Fifty-three percent were in complete remission at 100 days post-transplantation.

In total, these studies support the use of autologous HCT for patients with control of their HIV who would otherwise be candidates for HCT.

Allogeneic transplantation — Initial interest in the investigational use of allogeneic HCT in HIV-associated lymphoma came after there were reports of successful solid organ transplantation in the HIV population. There is a paucity of published data on the use of allogeneic HCT in PLWH; only case reports and retrospective series are available [115-118].

The largest retrospective analysis evaluated 23 HIV-positive patients undergoing myeloablative (87 percent) or nonmyeloablative (13 percent) allogeneic HCT between 1987 and 2003 [116]. The median time to neutrophil recovery was 18 days. The cumulative incidence rates of acute and chronic graft-versus-host disease were 30 and 28 percent, respectively. At a median follow-up of 59 months, OS rates at one and two years were 30 percent, each. OS rates were higher for those transplanted after 1996. There were nine treatment-related deaths, most often due to pulmonary toxicity. Subsequent anecdotal reports of nonmyeloablative allogeneic transplant suggest safety of this approach [119]. Furthermore, the use of donor stem cells homozygous for the CCR5-delta32-mutation (which resist HIV infection) eliminated HIV in one patient along with the successful treatment of acute myeloid leukemia [120]. The patient was able to discontinue ART without viral rebound.

Results of a multicenter prospective pilot study of allogeneic transplant in 17 people with HIV-related hematologic malignancies, including lymphoma (BMTCTN 0903/AMC 080), demonstrated that at 100 days there was no nonrelapse mortality, 13 patients were in complete remission, 4 patients had relapsed/progressive disease, and 8 patients achieved complete chimerism. The cumulative incidence of grades 2 to 4 acute graft-versus-host disease was 41 percent. At six months, the OS was 82 percent; nine patients achieved complete chimerism. At one year, the OS was 59 percent; eight deaths were from relapsed/progressive disease (5), acute graft-versus-host disease (1), adult respiratory distress syndrome (1), and liver failure (1). There were no infectious deaths [121]. These observations suggest that allogeneic stem cell transplant is not more toxic in PLWH than in the general population and may represent effective therapy in selected patients.

Chimeric antigen receptor T cell therapy — Axicabtagene ciloleucel as it is the only approved chimeric antigen receptor (CAR)-T cell product that does not specifically exclude HIV patients.

There have been no production issues, unexpected toxicities, or failure of CAR-T expansion in any cases of CAR-T therapy in PLWH. One case report describes relapsed/refractory DLBCL in a PLWH who achieved complete response (CR) at three months with no cytokine release syndrome (CRS) or neurotoxicity, with an ongoing response after six months [122]. CR was maintained for >1 year and 28 days in two PLWH with refractory NHL treated with CAR-T cell therapy; one patient had grade 2 CRS and grade 3 neurotoxicity [123]. A third case report described a less favorable response to CAR-T cell therapy, but treatment was tolerated [124].

PROGNOSIS

General — Although few patients diagnosed early in the HIV epidemic were cured of their lymphoma, the introduction of antiretroviral therapy (ART) in the late 1990s resulted in a marked improvement in the ability of these patients to tolerate standard chemotherapy regimens and clinical outcomes that now approach those in the general population. As described above, outcomes with full-dose chemotherapy for aggressive B cell lymphomas appear to be similar in HIV-positive and HIV-negative populations treated with similar regimens [125-127]. Survival in patients with HIV-related lymphoma is related to factors associated with both HIV infection and the lymphoma. More than 50 percent of patients with HIV-related non-Hodgkin lymphoma (NHL) will have long-term survival with conventional-dose chemotherapy [12,26,27,46,50-52,57]. Of the patients who do not do well, almost one-half die from progression or recurrence of the lymphoma, and the remainder die from opportunistic infections or other AIDS-related complications [5,11,15,36,57,128-130].

A retrospective review of 23,050 patients with HIV at eight United States Centers for AIDS Research (CFAR) sites from 1996 to 2010 identified 476 (2.1 percent) cases of lymphoma with estimated overall survival (OS) rates at two and five years of 53 and 44 percent, respectively [131]. The following factors were associated with significantly higher mortality:

●Older age at diagnosis

●Lymphoma diagnosed while on ART

●Lower CD4 count at lymphoma diagnosis

●Higher HIV ribonucleic acid (RNA) at diagnosis

●Aggressive histologic subtype (eg, primary central nervous system [CNS] lymphoma; Burkitt lymphoma, diffuse large B cell lymphoma [DLBCL])

The impact of these factors is discussed in more detail in the sections that follow.

Risk factors prior to therapy

Clinical features — The response rate and survival of the HIV-related lymphomas appear to correlate with the number of adverse prognostic features. The largest pre-ART study, from the AIDS Clinical Trials Group, that addressed pretherapy prognostic features consisted of 192 patients treated with low- or standard-dose M-BACOD (methotrexate, bleomycin, doxorubicin, cyclophosphamide, vincristine, and dexamethasone) [129]. The following were identified as adverse factors:

●Age >35 years

●Intravenous drug use

●Stage III or IV disease

●CD4 cell count <100/microL

Patients with no or one adverse factor, compared with those with three or four adverse factors, had a higher median survival (46 versus 18 weeks) and 2.8-year survival rate (30 versus 0 percent).

Somewhat similar findings were noted in a series of 68 patients treated with CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone) [5]. Patients at a high risk of relapse were identified by at least two of the following three features: CD4 cell count <50/microL, Eastern Cooperative Oncology Group (ECOG) performance status of 3 or 4, and pre-existing AIDS-defining opportunistic infection. The median survival was 21 months in the good-risk group compared with four months in the high-risk group.

A prognostic index has been proposed based on the experience in 111 patients with HIV lymphoma identified from a large cohort of people living with HIV [16]. Regression modeling identified CD4 count and International Prognostic Index (IPI) score as significant prognostic factors, and four internally validated risk strata were identified with one-year OS rates of 82, 47, 20, and 15 percent.

Other studies have confirmed that a greater degree of immunosuppression, whether assessed by the CD4 cell count or a prior history of an AIDS-defining illness [3,6,128,132], advanced-stage of the lymphoma [3,133], age >35 to 40 years [128,132], and high-risk disease, as defined by the IPI score (table 2), [15,16,57,134-136] are important adverse risk factors. There are conflicting data on the importance of performance status [3,129,133], serum lactate dehydrogenase [128,137], and intravenous drug use [3,129].

Histology — As in the HIV-negative population, the prognosis of patients with HIV-related lymphoma depends at least partially on the histologic subtype of the lymphoma identified. Hodgkin lymphoma is associated with the best prognosis while primary CNS lymphoma is associated with the highest mortality rate.

As an example, a retrospective review of uncontrolled prospectively collected data on 23,050 patients with HIV at eight CFAR sites from 1996 to 2010 identified 476 (2.1 percent) cases of lymphoma with estimated OS rates at two and five years of 53 and 44 percent, respectively [131]. Corresponding rates according to histologic subtype were as follows:

●Hodgkin lymphoma – 72 and 62 percent

●Burkitt lymphoma – 53 and 50 percent

●Primary CNS lymphoma – 24 and 23 percent

●DLBCL – 56 and 44 percent

While some prospective therapeutic trials discussed elsewhere in this section have demonstrated better outcomes than these data would suggest, the patient population represented in this analysis is a heterogeneous one in which presentations, treatment strategies (if any), and comorbidities may vary greatly. It should serve only as a general overview of outcomes for these different histologic groups.

In HIV-negative patients with DLBCL, the immunophenotype has been shown to be of prognostic importance. As is the case for nonimmunocompromised patients with DLBCL, immunophenotypic analysis reveals a post-germinal center phenotype (non-GC) to be associated with significantly worse outcomes when compared with a GC phenotype [57,132,136]. (See "Prognosis of diffuse large B cell lymphoma", section on 'Overview'.)

Features associated with the plasmablastic and primary effusion subtypes of DLBCL, including negativity for CD20, negativity for CD10, and positivity for CD138/Syn-1, are associated with a worse prognosis [72,77].

Effect of antiretroviral therapy — Finally, as noted above, there is retrospective evidence to support a survival benefit from the addition of ART to chemotherapy. Treatment outcomes in prospective trials conducted since the introduction of ART show markedly improved survival times when compared with studies from the pre-ART era. (See 'Incorporating antiretroviral therapy' above.)

As an example, in a large French study in 485 patients treated with a risk-adapted approach to chemotherapy, ART use, immune status, and IPI score were independently associated with prognosis [15]. In another study, the two-year OS of HIV-positive patients with low, low-intermediate, or high-intermediate risk disease as defined by the IPI (table 2) who were treated with chemotherapy plus ART was not significantly different from that of HIV-negative patients [16]. These authors suggested that a weighted prognostic score combining the IPI with the CD4 cell count (<100 versus >100 cells/microL) provided further independent prognostic information for people with HIV-related systemic lymphoma treated in the ART era.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Management of Burkitt lymphoma".)

SUMMARY AND RECOMMENDATIONS

●Description – Patients with human immunodeficiency virus (HIV) infection have an increased risk for systemic non-Hodgkin lymphomas (NHLs). The treatment of systemic lymphoma in people living with HIV (PLWH) infection may be complicated by the immunocompromised state.

●Antiretroviral therapy – Antiretroviral therapy (ART) should be given to all PLWH but may require modification to enable the administration of chemotherapy and/or radiation therapy. (See 'Incorporating antiretroviral therapy' above.)

●Infection prophylaxis – Prophylaxis for Pneumocystis jirovecii pneumonia is generally provided to PLWH and systemic lymphoma. We individualize decisions about prophylaxis for Mycobacterium avium complex, viral infections, and fungal infections. If gastrointestinal toxicity causes inconsistent use of antiviral agents, ART therapy may be discontinued for the duration of chemotherapy and restarted once chemotherapy is completed. (See 'Prophylaxis for opportunistic infections' above.)

●Diffuse large B cell lymphoma – Treatment is stratified according to the presence of high-risk features: high proliferative index (Ki67 >80 percent), non-germinal center B cell origin, double-hit cytogenetics, double protein expression, high International Prognostic Index (IPI) score, or plasmablastic histology.

•High risk – We suggest daEPOCH (dose-adjusted infusional etoposide, vincristine, doxorubicin, prednisone, cyclophosphamide) (table 3) rather than CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone) (Grade 2C). (See 'Dose-adjusted EPOCH' above.)

For patients with CD4 count ≥50/microL, we add concurrent rituximab to daEPOCH (daEPOCH-R) rather than treating sequentially with chemotherapy followed by rituximab (Grade 2B).

•Standard risk – Informed by the CD4 count:

-≥50 CD4/microL – We suggest R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone) rather than CHOP alone (Grade 2B). (See 'Rituximab plus CHOP or CDE' above.)

-<50 CD4/microL – The use of rituximab is individualized, but we generally treat with CHOP alone rather than R-CHOP (Grade 2B). (See 'CHOP' above and 'Rituximab plus CHOP or CDE' above.)

●Burkitt lymphoma/leukemia – For HIV-related Burkitt lymphoma/leukemia, we suggest R-CODOX-M/IVAC (table 5) or daEPOCH-R rather than R-CHOP (Grade 2C).

daEPOCH-R is less toxic than R-CODOX-M/IVAC, but it has less of a track record in this setting, and it does not include high-dose methotrexate to control central nervous system disease. (See 'Burkitt lymphoma/leukemia' above.)

●Plasmablastic lymphoma – We treat plasmablastic lymphoma like Burkitt lymphoma/leukemia. (See 'Plasmablastic lymphoma' above.)

●Hodgkin lymphoma – As with Hodgkin lymphoma (HL) in HIV-negative patients, the treatment of HL in PLWH is guided by disease stage and prognostic factors. (See 'Hodgkin lymphoma' above.)

●Central nervous system involvement – We perform a diagnostic lumbar puncture for all patients with Burkitt lymphoma/leukemia and for patients with diffuse large B cell lymphoma who have a higher risk for central nervous system involvement. (See 'Central nervous system-directed therapy in aggressive B cell lymphoma' above.)

●Relapsed/refractory disease – Treatment is individualized according to prior treatments, medical fitness, and patient preference, as discussed above. (See 'Treatment of relapsed disease' above.)