IL-2-based immunotherapy after autologous transplantation for lymphoma and breast cancer induces immune activation and cytokine release: a phase I/II trial


We determined the safety, immune activating effects, and potential efficacy of i.v. infusion of ex vivo interleukin-2 (IL-2) activated natural killer (NK) cells (part I) or IL-2 boluses (part II) during daily s.c. IL-2 administration following hematopoietic recovery from autologous transplantation. In all, 57 patients with relapsed lymphoma (n=29) or metastatic breast cancer (n=28) were enrolled. In part I of the study, 34 patients were enrolled at three dose levels of ex vivo IL-2-activated NK cells. Lymphaphereses were performed on days 28 and 42 of s.c. IL-2 administration. Following overnight ex vivo IL-2 activation of the pheresis product, the cells were reinfused the following day. In part II, 23 patients were enrolled at three dose levels of supplemental i.v. IL-2 bolus infusions, given on days 28 and 35 during s.c. IL-2 administration. Toxicities were generally mild, and no patient required hospitalization. Lytic function was markedly enhanced for fresh peripheral blood mononuclear cells (PBMNCs) obtained 1 day postinfusion of either IL-2-activated cells or IL-2 boluses. IL-2 boluses transiently increased the levels of IL-6, IFN-γ, TNF-α and IL1-β, with increases in IL-6 and IFN-γ being dose dependent. A total of 37 patients (19 patients with lymphoma, 18 with breast cancer) treated with an optimum dose of post-transplant immunotherapy (defined as having received 1.75 × 106 IU/m2/day of s.c. IL-2 plus at least one of the planned ex vivo IL-2-activated cell infusions/IL-2 boluses) could be matched with controls from the Autologous Blood and Marrow Transplant Registry database. The matched-pairs analysis demonstrated no improvement in disease outcomes of survival and relapse. We conclude that IL-2-activated cells/IL-2 boluses can be safely administered, generate PBMNCs with enhanced cytotoxicity against NK-resistant targets, and increase cytokine levels. With this dose and schedule of administration of IL-2, no improvement in patient disease outcomes was noted. Alternative strategies will be needed to exploit the immunotherapeutic potential of IL-2-activated NK cells.


High-dose chemotherapy with autologous stem cell support has been widely employed in the treatment for relapsed lymphoma and metastatic breast cancer. In chemotherapy-sensitive lymphoma and metastatic breast cancer patients, autologous transplantation can result in high remission rates.1,2 However, many patients who achieve a complete remission following transplantation will relapse and die of their disease. Most patients relapse at sites of previous bulk disease, suggesting that one reason for treatment failure is inability to eradicate chemotherapy-resistant clonogenic tumor. An inadequately functioning immune system may also contribute to early relapse.

Human natural killer (NK) cells represent a population of large granular lymphocytes that express the CD56+/CD3− phenotype. Recombinant interleukin-2 (IL-2) stimulates both NK expansion and cytotoxic activity against tumor targets in vivo and in vitro, including lymphoma and breast cancer.3,4 Theoretically, IL-2-activated NK cells, functioning as short-term antitumor therapy, would be most efficacious in a minimal residual disease setting such as that induced by autologous transplantation. Following initial reports of the feasibility and immunologic potential of post-transplant IL-2 immunotherapy,5,6,7 many investigators explored IL-2 and/or IL-2-activated effectors for a variety of malignancies with encouraging but inconclusive results.8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27

We previously reported the results of a phase I trial of post-autologous transplant s.c. IL-2 (Amgen Inc., Thousand Oaks, CA, USA).4 In all, 12 patients (six lymphoma, six breast cancer) were enrolled on a dose-escalation study of s.c. IL-2 given for 84 days; the best-tolerated dose was 0.25 × 106 U/m2/day. Patients receiving at least 28 days of s.c. IL-2 exhibited a greater than 10-fold increment in circulating NK cells. In addition, lytic function was increased against the NK-resistant targets Raji (lymphoma) and MCF-7 (breast cancer). In vivo IL-2-primed NK cells obtained by lymphapheresis were activated in large-scale ex vivo incubation in high-dose IL-2 (1000 U/ml), with further enhancement of NK lytic function against Raji and MCF-7 targets. We concluded that low-dose s.c. IL-2, although safe and easily administered in the outpatient setting, resulted in submaximal NK lytic function against tumor targets, and hypothesized that ex vivo incubation in IL-2 could be used to generate NK cells with potent antitumor effects. In the current study, we sought to extend these observations by performing a phase I/II trial of infusion of ex vivo IL-2-activated NK cells to determine if this strategy would generate NK cells with sustained and potent antitumor effects. We subsequently evaluated whether i.v. boluses of IL-2 could be substituted for the ex vivo IL-2-activated cell infusions, while maintaining the same degree of immune activation.

Patients and methods

Patient characteristics and eligibility criteria

In all, 34 patients between 18 and 65 years of age who had undergone an autologous stem cell transplant for non-Hodgkin's lymphoma, Hodgkin's disease, or metastatic breast cancer with no evidence of disease progression were enrolled on the ex vivo IL-2 activation part of the study (Table 1). A total of 23 additional patients were enrolled on the IL-2 bolus part of the study. Patients with T-cell immunophenotype non-Hodgkin's disease were not eligible. Conditioning regimens were as previously described.1,28 Patients had to be between 30 and 180 days from transplant, transfusion independent, without need for growth factor support, and have an unsupported platelet count 80 000 × 109/l, hemoglobin 9 gm/dl, and an ANC 1.0 × 109/l. The protocol received institutional review board approval and all patients exercised written informed consent when eligible for IL-2 therapy. Of the 28 patients with breast cancer, 18 (64%) also received s.c. IL-2 as part of pretransplant stem cell mobilization as previously described.28

Table 1 Patient characteristics

s.c. IL-2 therapy

Amgen IL-2 had been used in our previous phase I trial of s.c. IL-2.4 As Amgen IL-2 was no longer available, IL-2 was provided and manufactured by Chiron (Emeryville, CA, USA) for the current study. Hence, it was necessary to first establish the maximum tolerated dose (MTD) of Chiron-IL-2. Acetaminophen 650 mg, ibuprofen 400 mg, and diphenhydramine 50 mg were self-administered as oral premedications to IL-2, and repeated 4 h following each self-injection. In part I of the study, IL-2 doses of 0.25, 0.75, 1.25, 1.75, and 2.25 × 106 IU/m2/day for 56 days were tested, within a range of Chiron IL-2 previously reported to expand NK cells in vivo without significant toxicity.29 Patients were enrolled in cohorts of 3. Patients were not enrolled to the next dose level until a minimum of two patients had completed therapy at an IL-2 dose level with no dose limiting toxicity (DLT) attributable to IL-2 administration. DLT was defined by the Cancer and Leukemia Group B (CALGB) expanded common toxicity criteria as Grade 3 toxicity. Toxicity was first formally assessed at day 14 of IL-2 therapy, then every 14 days until 7–14 days after IL-2 was stopped. If no DLT had occurred by day 14, the patient could escalate to the next dose level for the duration of therapy. IL-2 was discontinued in patients experiencing DLT, then restarted at the next lower dose level once toxicity had resolved. The MTD of IL-2 was defined as the dose below that in which 33% of patients experienced DLT. The MTD was used as the daily s.c. dose of IL-2, administered for 42 days, in part II of the study.

Lymphapheresis and ex vivo IL-2 activation (part I)

Minimal laboratory values required prior to lymphapheresis included platelets >50 000 × 109/l, hemoglobin 9 gm/dl, ANC 1.0 × 109/l, and normal renal and liver functions. Prior to the first lymphapheresis (day 28 of IL-2 therapy), patients must have received IL-2 for at least 21 of the first 28 days and for five of the 7 days immediately prior to the planned apheresis. Prior to the second lymphapheresis (day 42), patients must have received IL-2 for at least 28 of the first 42 days and five of the 7 days immediately prior to apheresis. Patients not meeting the criteria for apheresis continued on s.c. IL-2 alone. A volume of 10–12 l leukaphereses were performed using a CS3000 cell sorter (Baxter, Deerfield, IL, USA). The lymphapheresis product was cultured in AIM-V serum-free medium (Gibco) supplemented with 1000 IU/ml IL-2 in polyolefin gas permeable bags (Lifecell Tissue Culture Flask, Fenwal-Baxter) for 16 h, a time period previously shown to result in sustained induction of cytolytic activity against MCF-7 and Raji cell targets.4 All patients received premedications 1 h prior to cell infusion and again 3–4 h later. IL-2-activated lymphocytes were infused over 1–2 h through a 170 μm blood administration filter the day following apheresis (days 29 and 43) in the outpatient clinic. Cell doses were based on mononuclear cells with the total cell count determined just prior to infusion. Patients were enrolled in cohorts of 3 patients each with dose escalation to a maximum of a full lymphapheresis product. Cohort 1 received 4.0 × 107 cells/kg and cohort 2 received 8.0 × 107 cells/kg. If the planned cell dose was greater than the actual number of cells collected, then the dose infused was the full lymphapheresis product. Cohort 3 received a full lymphapheresis product. The median cell dose in a full lymphapheresis product was 7.8 × 107 cells/kg, with a range of 0.33–2.09 × 108 cells/kg. Until the s.c. IL-2 dose escalation was completed and the MTD of daily IL-2 determined to be 1.75 × 106 IU/m2/day, all patients received the lowest cell dose of 4.0 × 107 cells/kg. Cell dose escalation then proceeded only after all patients at a given dose were evaluated through day 56 of therapy (the last day of s.c. IL-2) with no DLT attributable to administration of IL-2 or IL-2-activated lymphocytes. Patients received the infusions on an outpatient basis and were observed for toxicity for a minimum of 4 h following infusion. Patients were formally assessed for toxicity the day prior to apheresis, during and following infusion of activated cells, and the day after cell infusion. DLT was defined as Grade 2 or greater toxicity by CALGB criteria except for the following where Grade 3 toxicity was dose limiting: liver, neurologic, hematologic, infection, pulmonary, dermatologic, allergy, performance status. A platelet count of <30 000 × 109/l was considered dose-limiting. After the MTD of IL-2 in combination with a full lymphapheresis was shown to be safe, accrual was extended to both ascertain safety and until part II of the study utilizing bolus IL-2 infusions could be initiated.

i.v. bolus IL-2 (part II)

The eligibility requirements for part II of the study were identical to part I of the study. All patients initiated s.c. IL-2 at the MTD established in part I of the study of 1.75 × 106 IU/m2/day. If no DLT had occurred after 14 days, the patient continued at that dose for a total of 42 days of therapy. Patients who developed intolerable constitutional symptoms that did not constitute DLT decreased their IL-2 dose to 1.25 × 106 IU/m2/day for the remainder of the study. Patients received the same pre- and post- medications for IV boluses of IL-2 as for the infusions of IL-2-activated cells. The IL-2 boluses were given as an outpatient over 2 h on days 28 and 35, with dose escalation per cohort of three patients each of 2.0, 4.0, and 6.0 × 106 IU/m2 per infusion. The dose of bolus IL-2 was not escalated until all patients in a given cohort had completed the entire course of IL-2 therapy. Hence, more than three patients were ultimately enrolled in the second and third cohorts. Prior to receiving bolus IL-2, patients must have received IL-2 for at least 21 of the first 28 days and for five of 7 days immediately prior to the planned infusion. Patients not meeting these criteria continued on daily s.c. IL-2 alone. The IL-2 was infused over 2 h on an outpatient basis and patients were monitored for a minimum of 4 h for complications. Patients were formally assessed for toxicity the day prior to the IL-2 bolus infusion, during and following the infusion and the day after infusion. The definition of DLT for bolus IL-2 was the same as for ex vivo IL-2-activated cell infusions.

Cytolytic activity

Heparinized peripheral blood, serum, and plasma were obtained at each clinic visit. Peripheral blood mononuclear cells (PBMNCS) were prepared from peripheral blood by Ficoll–Hypaque (sp. grav. 1.077) (Sigma, St Louis, MO, USA) density gradient centrifugation as described.4 Cell surface antigens were determined by direct staining of cells with mouse monoclonal antibodies. Fluorescein isothiocyanate (FITC)- or phycoerythrin (PE)-coupled antibodies (Becton Dickinson, Mountain View, CA, USA) were directed at CD3, CD25, CD56, and HLA-DR. FITC- and PE-coupled isotype-matched immunoglobulins were used as controls. All analyses were performed with FACSCalibur (Becton Dickenson) and CELLQuest software (Becton Dickenson). Cytotoxicity assays were performed in triplicate using fresh PBMNC or 16 h IL-2-activated PBMNC against the NK-resistant B-cell lymphoma line Raji (American Tissue Culture Collection (ATCC), Rockville) and the breast cancer cell line MCF-7 (ATCC) in a 4-hour 51Cr release assay.4 To control for target variability, frozen targets (expanded from the same batch) were thawed every 4–6 weeks.

Cytokine levels

Serum or plasma was stored at −70°C and later assayed for cytokine levels using ELISA kits for interleukin-1β (IL-1β), interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α), and interferon-γ (IFN-γ). Assays were performed using the manufacturer's recommendations (R&D Systems, Minneapolis, MN, USA). Normal cytokine values for >95% of healthy individuals are considered by the University of Minnesota Cytokine Reference Laboratory to be as follows: IL-1β (<0.9 pg/ml), IL-6 (<3.0 pg/ml), TNF-α (<3.0 pg/ml), and IFN-γ (<3.0 pg/ml).


Results of experimental points were reported as mean± standard error of the mean (s.e.m.). Statistical comparisons of experimental points between independent groups were completed with the two-sided Student's t-test. Comparisons of values at different time points but within the same population were made with a two-sided paired-comparison t-test.30

Patients enrolled at the MTD of s.c. IL-2 and who received at least one of the planned ex vivo IL-2-activated cell infusions or IL-2 boluses (20 patients with lymphoma, 20 with breast cancer) were eligible to be matched to a maximum of three controls from the Autologous Blood and Marrow Transplant Registry (ABMTR) database. Controls were selected from 1726 potentially eligible patients from the database who had received an autologous transplant for lymphoma or breast cancer, had not received any post-transplant immunotherapy, were not transplanted at one of the study institutions, and had similar follow-up time. IL-2-treated patients were matched to controls for the following variables: disease (NHL, HD, metastatic breast cancer), histologic subtype at diagnosis or those patients with NHL or HD (Working Formulation subtype for NHL, nodular sclerosis/mixed cellularity/unknown for HD), disease status at time of transplant (primary induction failure with chemosensitive disease, clinical remission, or chemosensitive/resistant relapse for lymphoma; clinical remission, clinical remission except for bone abnormalities, or chemosensitive relapse for breast cancer), year of transplant (within 5 years), age at transplant (within 10 years), and time from diagnosis to transplant (within1.5 years). Patients with breast cancer were also matched to controls for the presence of visceral metastases and hormone receptor status. The case cohort included only patients who were in remission long enough to receive the IL-2 therapy after autologous transplantation, whereas the control cohort included all similar patients who were not treated with IL-2 therapy after autologous transplantation from the ABMTR. The IL-2-treated case cohort therefore excluded patients who died early or had an early relapse. This could introduce a bias favoring IL-2-treated cases. To adjust this potential waiting time bias, the controls in each match pair were selected from among the patients with a complete remission after transplantation at least as long as the interval between the transplantation and initial IL-2 therapy for the IL-2 recipient. If more than three controls were eligible for matching with an IL-2-treated patient, then the three controls closest in age to the case patient were chosen. Cox's regression model stratified on each matched pair was used to analyze the effect of IL-2 therapy on the end points of survival and relapse. The proportionality assumption of the Cox model was tested by adding a time-dependent covariate.


Clinical tolerability of subcutaneous IL-2 therapy

Of nine patients enrolled at dose levels of 0.25 to 1.25 × 106 IU/m2/day, all but two were able to have their dose of IL-2 increased by one dose level at day 14 of IL-2 treatment. One patient enrolled at 0.75 × 106 IU/m2/day did not dose escalate secondary to thrombocytopenia; one patient enrolled at 1.25 × 106 IU/m2/day did not dose escalate because of fatigue and nausea. Toxicities were otherwise mild, including Grade I fatigue, nausea, rash, cough, fever, myalgias, and sweats. The majority of patients developed transient induration and erythema around the s.c. injection sites.

A total of 25 patients began IL-2 at 1.75 × 106 IU/m2/day. In all, 15 (60%) of the 25 patients were able to dose escalate to 2.25 × 106 IU/m2/day on day 14. Of these, 13 were able to remain on the higher dose for the duration of treatment, whereas the other two patients required a subsequent decrease back to the 1.75 × 106 IU/m2/day dose level because of either anemia or thrombocytopenia. In all, 10 (40%) of the patients, however, were unable to have their dose escalated at day 14 because of side effects, and seven of these patients ultimately were removed from study. Toxicities included: severe swelling of the tongue and face (n=1); thrombocytopenia (n=2); nausea and vomiting (n=1); intolerable fatigue (n=5); edema (n=1); and bacterial infection (n=1). In all, eighteen (72%) of the 25 patients who began IL-2 at 1.75 × 106 IU/m2/day received 95% of the total number of planned injections. No patient required hospitalization; and all toxicity resolved within 1 week of discontinuation of IL-2. The MTD of IL-2 was thus established at 1.75 × 106 IU/m2/day; therefore, no patients were enrolled at a starting dose of 2.25 × 106 IU/m2/day.

Clinical tolerability of ex vivo IL-2-activated cell infusions

All patients who received s.c. IL-2 at less than the MTD (0.25–1.25 × 106 IU/m2/day) received a cell dose of 4.0 × 107 cells/kg. Grade I toxicities observed infrequently at this dose level included fever and chills at time of the infusion. At the MTD of s.c. IL-2, 1.75 × 106 IU/m2/day, four patients received a cell dose of 4.0 × 107 cells/kg (one patient underwent one apheresis, the remaining patients two aphereses); six patients received 8.0 × 107 cells/kg (two patients underwent one apheresis, the remaining four patients two aphereses); 10 patients underwent two lymphaphereses with reinfusion of the full product. Grade I–II toxicities included wheezing (n=1), mild fever (37.1–38.0°C) (n=4), chills (n=4), transient decrease in oxygen saturation to 88% during chills (n=1), and transient decrease of systolic blood pressure to 80–85 mmHg (n=1). No patient required hospitalization. Two patients were removed from study after the first apheresis (one for facial swelling, one for a bacterial infection). A third patient declined the second apheresis secondary to difficulties with venous access.

Clinical tolerability of IL-2 boluses

All 23 patients received s.c. IL-2 at the MTD of 1.75 × 106 IU/m2/day. Three patients made up the first cohort, and all three received 42 days of s.c. IL-2 injections and both planned i.v. IL-2 boluses of 2 × 106 IU/m2. Six (50%) of the 12 patients enrolled at the 4 × 106 IU/m2 dose level received both i.v. boluses. Boluses were not infused because of thrombocytopenia (n=2), Clostridium difficile infection (n=1), rash (n=1), insomnia (n=1), and neutropenia (n=1). Of the 12 patients, eight received all 41–42 doses of s.c. IL-2; four patients received 23 days of s.c. IL-2. Grade I toxicities occurring at the IL-2 bolus dose of 4 × 106 IU/m2 included fever (n=2), chills (n=3), and shortness of breath (n=1). Orthostatic hypotension occurred in two patients–one patient responded to intravenous normal saline, the other had spontaneous resolution of symptoms. All eight patients enrolled at the highest dose of bolus IL-2, 6 × 106 IU/m2, received all 42 days of s.c. IL-2 injections and both infusions. Grade II toxicities of fever and chills were noted in all patients during the bolus infusions, and thus this dose was felt to be the MTD of IL-2 boluses. Three patients experienced a 10 mmHg decline in systolic blood pressure and required infusion of normal saline for blood pressure support.

Immune activation after IL-2 therapy

Laboratory tests were performed every 2 weeks to monitor immune activation. There was a significant increase in the total white blood cell count, peaking at day 14 and after each activated cell infusion or IL-2 bolus infusion, analogous to our previously reported findings with post transplant subcutaneous IL-2.4 The increase in total WBC was primarily because of an increase in the number of circulating lymphocytes and eosinophils; the absolute number of circulating monocytes did not change with IL-2 therapy. The lymphocyte increase included a greater than 10-fold increase in circulating CD56+ bright/CD3− NK cells, which was sustained throughout the IL-2 treatment period. The majority of the NK cells were CD56+ bright/CD16+/CD3−. Less than 2% of CD56+ cells expressed CD3 at any time point. In contrast to the increase in NK cells, the absolute number of T cells remained fairly stable throughout IL-2 administration, consistent with our previous finding.4 The phenotype of the mononuclear cells was unaltered by ex vivo incubation in IL-2. After 28 days of s.c. IL-2, 44±4.6% (n=20) of PBMNC were CD56+/CD3− NK cells, which did not change after overnight incubation in IL-2 (43±4.4%), again in agreement with our previous findings.4

We tested the lytic function of PBMNC obtained at study entry prior to any s.c. IL-2 administration, pre- and post infusion of IL-2-activated cells, and pre- and post infusion of IL-2 boluses, as well as that of the IL-2-activated NK-cell infusion product against Raji and MCF-7 cells (Figure 1). Prior to s.c. IL-2, fresh nonactivated PBMNC had no lytic activity (<5%) against either target. In vitro cytotoxicity was modestly enhanced following 28 days of s.c. IL-2, prior to infusion of either IL-2-activated cells or i.v. IL-2 boluses (about 10% at effector:target ratio of 20:1). In contrast, cytotoxicity was markedly enhanced for fresh PBMNC obtained the day postinfusion of IL-2-activated cells or i.v. IL-2 boluses (25–45% at effector:target ratio of 20:1), as well as for the first IL-2-activated NK-cell infusion product (40–60% at effector:target ratio of 20:1). The enhancement in cytotoxicity following the second IL-2-activated cell infusion or IL-2 bolus was as potent as the first (data not shown).

Figure 1

Ex vivo IL-2-activated cell infusions or i.v. IL-2 boluses enhance NK-cell lytic activity against Raji (lymphoma) and MCF-7 (breast cancer) targets. Peripheral blood mononuclear cells (PBMNC) from patients receiving 1.75 × 106 IU/m2/day of s.c. IL-2 were studied before and after starting post-transplant immunotherapy in a 4-h chromium release cytotoxicity assay against Raji (left panel) and MCF-7 (right panel) targets. Patient PBMNC were tested prior to study entry (x, n=16 for Raji, n=10 for MCF-7), after 28 days of s.c. IL-2 alone prior to either cell () or bolus infusions (□) (n=18 for Raji, n=18 for MCF-7) and 24 h after ex vivo IL-2-activated NK cell infusions (•, n=11 for Raji, n=15 for MCF-7) or IV IL-2 (▪, n=9 for both targets). The activity of the first infusion product is also shown (•–, n=13 for Raji, n=15 for MCF-7).

Cytokines induced by IL-2

We previously reported that s.c. administration of IL-2 in the post-transplant setting induced the release of soluble IL-2Rα and IFN-γ.4 In contrast, neither IL-12, TNF-α, IL1-β nor IL-6 was significantly augmented compared to baseline levels. In the current study, we determined the circulating cytokine levels of IL-6, IFN-γ, TNF-α, and IL1-β in patients at day 0, day 14, and throughout their course of bolus IL-2 infusions. As shown in Figure 2, IL-2 boluses transiently increased the levels of all four cytokines, although the increase in the level of IL-1β was minimal. Peak levels of IL-6, IFN-γ, and TNF-α were noted 2 h after each IL-2 infusion. The peak level for IL1-β occurred 24 h after infusion; a second peak was not noted after the second IL-2 bolus. There was a dose-dependent increase in IL-6 and IFN-γ. Patients who received 4–6 × 106 IU/m2 had a statistically significant greater increase in peak cytokine levels of IL-6 and IFN-γ compared to patients who received 2 × 106 IU/m2 of IL-2 (data not shown).

Figure 2

IL-2 infusions induce a secondary cytokine cascade. Serum was obtained from patients prior to IL-2, after 14 days of s.c. IL-2 (1.75 × 106 IU/m2/day) alone, and before and after IL-2 infusions (4 – 6 IU/m2/day) on days 28 and 35 as indicated by the arrows on the X-axis. IFN-γ, TNF-α, IL-6, and IL-1β were analyzed by ELISA from serum samples collected at the time points listed. The arrows on the Y-axis indicate the cytokine level found for >95% of healthy individuals. Each point represents the mean±s.e.m. of paired serum samples from five treated patients.

Clinical outcome following IL-2 immunotherapy: matched-pairs analysis

Patients enrolled at the MTD of s.c. IL-2 and who received at least one of the planned ex vivo IL-2-activated cell infusions/IL-2 boluses (n=40, 20 lymphoma and 20 breast cancer) were eligible to be matched to controls from the ABMTR database. In all, 19 of the 20 patients with lymphoma and 18 of the 20 patients with metastatic breast cancer could be matched with control patients. For the 19 patients with lymphoma, one patient was matched with one control, one with two controls, and 17 with three controls from the registry. For the 18 patients with breast cancer, one patient was matched with one control and 17 patients with three controls from the registry. Patient characteristics are shown in Table 2. There were either very minimal or no differences between case patients and control patients for all matching characteristics. Disappointingly, improvement in survival or relapse between case and control patients was not detected (Table 3). The power to detect a difference in this analysis was low because of the limited patient numbers, but no suggested advantage for the IL-2-based therapy was apparent. The risk of relapse for breast cancer patients treated with IL-2 was greater than for control patients; unmeasured clinical variables not accounted for in the matching criteria may have contributed to the outcome of this analysis.

Table 2 Patient characteristics in matched-pairs analysisa
Table 3 Relative risk of death or relapse of patients in matched-pairs analysis


Our previous study suggested that low-dose s.c. IL-2, although safe and easily administered in the outpatient setting, results in submaximal NK lytic function against tumor targets postautologous transplant. Furthermore, based on correlative laboratory studies we hypothesized that additional ex vivo incubation in IL-2 may be used to generate NK cells with enhanced NK lytic function. The goal of the current study was to test this hypothesis by performing a cell dose escalation of IL-2-activated cell therapy with concomitant determination of NK-cell cytotoxicity. We subsequently sought to minimize patient inconvenience as well as the cost associated with ex vivo incubation and reinfusion of activated cells by substitution with bolus IL-2 infusions. After first establishing the MTD of Chiron IL-2, we demonstrated that cytotoxicity was markedly enhanced for cells incubated overnight with IL-2, and from fresh PBMNC obtained the day postinfusion of the IL-2-activated cells against both lymphoma and breast cancer targets. We subsequently confirmed that IL-2 boluses could substitute for ex vivo-activated cells with maintenance of both patient safety as well as the immune-enhancing effects of IL-2.

Toxicities reported in a number of studies with IL-2 and lymphokine-activated killer (LAK) cells have included fever, chills, rash, nausea, dyspnea, thrombocytopenia, and hypotension,7,12,13,15,20,21,23,26,29,31 similar in spectrum to the toxicities noted in our study. Although direct comparison of the severity of toxicities between studies is problematic, low-dose IL-2 therapies with or without LAK cells appear to be less toxic as high doses of IL-2 and LAK cells have been reported to cause vascular leak syndrome, cardiac arrhythmia, myocardial infarction, and even death.31

Administration of IL-2 following autologous transplantation may improve immunologic function by inducing cellular changes and indirectly by stimulating the release of cytokines.7,10 Normally following autologous transplantation, endogenous NK cells are absent from the blood and appear no earlier than the third week.7 Both in our prior study and this study, we noted a greater than 10-fold increase in circulating CD56+ bright/CD3− NK cells by day 28 of s.c. IL-2, which was then sustained throughout the IL-2 treatment period. Ex vivo incubation for 16 h in IL-2 did not alter the phenotype of the cells, but greatly enhanced their cytolytic function to a degree analogous to i.v. IL-2 boluses. Other investigators, including Caligiuri et al8,9 and Soiffer et al,10 have noted a similar selective expansion of NK cells in the peripheral blood of patients treated with prolonged uninterrupted continuous infusions of low-dose IL-2, accompanied by enhancement of cytolytic activity. Caligiuri et al9 treated patients with advanced cancer for 90 days, and demonstrated a gradual, progressive expansion of NK cells with no evidence of plateau effect during the 3 months of therapy. Soiffer et al10 treated 13 marrow recipients with continuous infusion IL-2 for 90 days. All 13 patients experienced a gradual, incremental (five to 40-fold) increase in NK cell number during IL-2 treatment, and the majority of NK cells were CD56 bright+CD16+CD3−. A minor increase in T-cell number was noted in only four of 13 patients. The differences in the time course and degree of NK-cell expansion in these studies may be related to the differences in IL-2 source, dose, route of delivery, type of disease, or treatment setting (transplant or nontransplant).

Other investigators have explored the administration of IL2 with IL-2-activated NK cells or LAK cells in the postautologous.12,13,16 Lister et al13 reported 11 patients with lymphoma and metastatic breast cancer who, 2 days following autologous transplantation, received an infusion of IL-2-activated NK cells, followed by additional IL-2 at varying doses for over 90 days. The cells were generated pretransplant after 6 days of low-dose IL-2 priming with 3 × 106/IU/m2 IL-2 (Chiron) by continuous infusion, leukapheresis for 4 days, followed by IL-2 ex vivo culture for 14–18 days. In the Benyunes et al12 study of 16 patients following autologous transplantation for malignant lymphoma, patients were initially treated with IL-2 (Roche) at 3.0 × 106 U/m2/day by continuous infusion for 5 days. Patients then underwent three consecutive lymphaphereses, for LAK cell generation. Following 5 days of culture with IL-2, a median of 65, 61, and 43% of the infused LAK cells expressed CD3, CD8 or CD56, respectively. The LAK cells lysed the Daudi target in vitro with a median lysis of 54% at an effector to target ratio of 25:1, not dissimilar to the results in the current study.

NK cells constitutively express receptors for monocyte-derived cytokines (monokines) and produce critical cytokines (including IFN-γ, TNF-α, and granulocyte–macrophage colony-stimulating factor (GM-CSF)) in response to monokine stimulation.32 In addition to their presumed role in the constitutional symptoms observed with IL-2 administration,33 theoretically cytokines may be either important mediators or suppressors of IL-2 cell activation.34,35,36 Several studies have attempted to define the nature of the complex cytokine cascade following IL-2 administration. IL-6, granulocyte colony-stimulating factor (G-CSF), and GM-CSF are induced in the blood after administration of IL-2.37 Bonig et al38 demonstrated a consistent and early rise of IL-10, IL-5, and GM-CSF during IL-2 therapy. TNF-α, IFN-γ, and IL-6 are produced in significant amounts in supernatants of marrow mononuclear cells cultured in IL-2.39 In the current study, IL-2 boluses transiently increased the levels of IL-6, IFN-γ, TNF-α and IL1-β, with a dose-dependent increase in IL-6 and IFN-γ.

Several trials with differing eligibility and design have been performed with the aim of evaluating the safety and efficacy of immunotherapy with IL-2 in lymphoma. Benyunes et al12 treated 17 patients with IL-2 with or without LAK cells after autologous transplantation. The clinical outcomes compared favorably with institutional controls. Nagler et al18 conducted a clinical trial involving 56 lymphoma patients with minimal residual disease following autologous stem cell transplantation utilizing a combination of s.c. IL-2 and INF-α in an outpatient setting and compared the results with 61 matched historical controls. The overall survival, disease-free survival and relapse rate of treated patients were significantly improved compared with historical controls. In a phase I/II study, Robinson et al15 treated 19 patients with lymphoma with escalating doses of ‘induction’ IL-2; following a 4-day rest period, maintenance IL-2 was given by continuous i.v. infusion for 10 days. Encouragingly, 58% of the non-Hodgkin's lymphoma patients in the phase II trial remained in clinical remission with a minimum follow-up of 1 year. Margolin et al22 reported 24 patients with lymphoma who received bone marrow and/or G-CSF-mobilized autologous peripheral blood stem cells that had been exposed to IL-2 for 24 h ex vivo. Patients subsequently received IL-2 by low-dose continuous i.v. infusion until hematologic reconstitution and then intermediate-dose continuous i.v. infusion for six maintenance cycles. Among the 24 lymphoma patients, nine were in continuous CR from 18 to 48 months, and 15 had died (12 due to relapse and three to therapy-related toxicities).

Other investigators have attempted to determine if IL-2 therapy is efficacious in patients with metastatic breast cancer. Gravis et al25 noted no clinical beneficial effect for 21 advanced heavily pretreated breast cancer patients treated with either i.v. high-dose IL-2 or s.c. low-dose IL-2 following autologous transplantation. Toh et al26 reported a prospective phase II trial of 33 patients with chemosensitive metastatic breast cancer, who underwent transplantation with autologous peripheral blood stem cells cultured in IL-2, followed by post-transplant low-dose IL-2. Compared to historical controls group, the Kaplan–Meier-estimated 2-year progression-free survival was 35% for IL-2-treated patients, compared with 17% in the control arm (P=0.73) and the estimated 2-year survival was 78%, compared with 61% in the control arm (P=0.22).

As no prior study has reported a matched-pairs analysis of postautologous transplantation IL-2 therapy alone, we performed one to assess the possible efficacy of the immune activation we observed. Although there was no trend towards an advantage to IL-2-based immunotherapy, the power to detect a difference in this matched-pairs analysis was low because of limited patient numbers. In order to achieve narrow enough confidence intervals to show statistical significance for the observed estimates of relative risk of death, the study would require approximately 35 lymphoma cases and 25 breast cancer cases with a corresponding increase in the number of controls. Patient heterogeneity may also have obscured a clinical benefit. Thus, a larger and more uniform, prospective randomized study at the MTD of s.c. IL-2- and ex vivo-activated cells or i.v. bolus IL-2 would be needed to clarify the potential for clinical benefit. Enhancement of the potency of IL-2-activated NK cytotoxicity may require the use of more infusions of ex vivo-activated cells or IL-2 boluses, utilization of highly purified NK cells or NK-cell subsets, a combination of IL-2 with tumor-reactive monoclonal antibodies to induce effective antibody-dependent cellular cytotoxicity,40 combinations of NK-cell immunotherapy with chemotherapy or other cytokines such as IL-12,41 or augmentation by NK-cell inhibitory receptor blockade.42 Indeed, as malignant cells may have multiple mechanisms by which they escape from immune control, more than one immunotherapeutic intervention may be necessary to achieve clinical benefit. Further study is clearly needed to take advantage of the immunotherapeutic potential of IL-2-activated NK cells.


  1. 1

    Weisdorf DJ, Verfaillie CM, Miller WJ et al. Autologous bone marrow versus non-mobilized peripheral blood stem cell transplantation for lymphoid malignancies: a prospective, comparative trial. Am J Hematol 1997; 54: 202–208.

  2. 2

    Antman KH, Rowlings PA, Vaughan WP et al. High-dose chemotherapy with autologous hematopoietic stem-cell support for breast cancer in North America. J Clin Oncol 1992; 15: 1870–1879.

  3. 3

    Verma UN, Areman E, Dickerson SA et al. Interleukin-2 activation of chemotherapy and growth factor-mobilized peripheral blood stem cells for generation of cytotoxic effectors. Bone Marrow Transplant 1995; 15: 199–206.

  4. 4

    Miller JS, Tessmer-Tuck J, Pierson BA et al. Low dose subcutaneous interleukin-2 after autologous transplantation generates sustained in vivo natural killer cell activity. Biol Blood Marrow Transplant 1997; 3: 34–44.

  5. 5

    Gottlieb DJ, Brenner MK, Heslop HE et al. A phase I clinical trial of recombinant interleukin 2 following high dose chemo-radiotherapy for haematological malignancy: applicability to the elimination of minimal residual disease. Br J Cancer 1989; 60: 610.

  6. 6

    Blaise D, Olive D, Stoppa AM et al. Hematologic and immunologic effects of the systemic administration of recombinant interleukin-2 after autologous bone marrow transplantation. Blood 1990; 76: 1092–1097.

  7. 7

    Higuchi CM, Thompson JA, Petersen FB et al. Toxicity and immunomodulatory effects of interleukin-2 after autologous bone marrow transplantation for hematologic malignancies. Blood 1991; 77: 2561–2568.

  8. 8

    Caligiuri MA, Murray C, Soiffer RJ et al. Extended continuous infusion low-dose recombinant interleukin-2 in advanced cancer: prolonged immunomodulation without significant toxicity. J Clin Oncol 1991; 9: 2110–2119.

  9. 9

    Caligiuri MA, Murray C, Robertson MJ et al. Selective modulation of human natural killer cells in vivo after prolonged infusion of low dose recombinant interleukin 2. J Clin Invest 1993; 91: 123–132.

  10. 10

    Soiffer RJ, Murray C, Cochran K et al. Clinical and immunologic effects of prolonged infusion of low dose recombinant interleukin-2 after autologous and T-cell depleted allogeneic bone marrow transplantation. Blood 1992; 79: 517–526.

  11. 11

    Weisdorf DJ, Anderson PM, Blazar BR et al. Interleukin-2 immediately after autologous bone marrow transplantation for acute lymphoblastic leukemia: a phase I study. Transplantation 1993; 55: 61–66.

  12. 12

    Benyunes MC, Higuchi C, York A et al. Immunotherapy with interleukin-2 with or without lymphokine-activated killer cells after autologous bone marrow transplantation for malignant lymphoma: a feasibility trial. Bone Marrow Transplant 1995; 16: 283–288.

  13. 13

    Lister J, Rybka WB, Donnenberg AD et al. Autologous peripheral blood stem cell transplantation and adoptive immunotherapy with activated natural killer cells in the immediate posttransplant period. Clin Cancer Res 1995; 6: 607–614.

  14. 14

    Mazumder A . Experimental evidence of interleukin-2 activity in bone marrow transplantation. Cancer J Sci Am 1997; 3 (Suppl. 1): S37–S42.

  15. 15

    Robinson N, Benyunes MC, Thompson JA et al. Interleukin-2 after autologous stem cell transplantation for hematologic malignancy: a phase I/II study. Bone Marrow Transplant 1997; 19: 435–442.

  16. 16

    Benyunes MC, Massumoto C, York A et al. Interleukin-2 with or without lymphokine-activated killer cells as consolidative immunotherapy after autologous bone marrow transplantation for acute myelogenous leukemia. Bone Marrow Transplant 1993; 12: 159–163.

  17. 17

    Meloni G, Vignetti M, Pogliani E et al. Interleukin-2 therapy in relapsed acute myelogenous leukemia. Cancer J Sci Am 1997; 3 (Suppl. 1): S43–S47.

  18. 18

    Nagler A, Ackerstein A, Or R et al. Immunotherapy with recombinant human interleukin-2 and recombinant interferon-alpha in lymphoma patients post-autologous marrow or stem cell transplantation. Blood 1997; 89: 3951–3959.

  19. 19

    Blaise D, Attal M, Pico JL et al. The use of a sequential high dose recombinant interleukin 2 regimen after autologous bone marrow transplantation does not improve the disease free survival of patients with acute leukemia transplanted in first complete remission. Leukemia Lymphoma 1997; 25: 469–478.

  20. 20

    Van Besien K, Margolin K, Champlin R, Forman S . Activity of interleukin-2 in non-Hodgkin's lymphoma following transplantation of interleukin-2-activated autologous bone marrow or stem cells. Cancer J Sci Am 1997; 3 (Suppl. 1): S54–S58.

  21. 21

    Meehan KR, Arun B, Gehan EA et al. Immunotherapy with interleukin-2 and alpha-interferon after IL-2-activated hematopoietic stem cell transplantation for breast cancer. Bone Marrow Transplant 1999; 23: 667–673.

  22. 22

    Margolin KA, Van Besien K, Wright C et al. Interleukin-2-activated autologous bone marrow and peripheral blood stem cells in the treatment of acute leukemia and lymphoma. Biol Blood Marrow Transplant 1999; 5: 36–45.

  23. 23

    DeMagalhaes-Silverman M, Donnenberg A, Lembersky B et al. Posttransplant adoptive immunotherapy with activated natural killer cells in patients with metastatic breast cancer. J Immunother 1999; 23: 154–160.

  24. 24

    Blaise D, Attal M, Reiffers J et al. Randomized study of recombinant interleukin-2 after autologous bone marrow transplantation for acute leukemia in first complete remission. Eur Cytokine Net 2000; 11: 91–98.

  25. 25

    Gravis G, Viens P, Vey N et al. Pilot study of immunotherapy with interleukin-2 after autologous stem cell transplantation in advanced breast cancers. Anticancer Res 2000; 20: 3987–3991.

  26. 26

    Toh HC, McAfee SL, Sackstein R et al. High-dose cyclophosphamide + carboplatin and interleukin-2 (IL-2) activated autologous stem cell transplantation followed by maintenance IL-2 therapy in metastatic breast carcinoma – a phase II study. Bone Marrow Transplant 2000; 25: 19–24.

  27. 27

    Toren A, Nagler A, Rozenfeld-Granot G et al. Amplification of immunological functions by subcutaneous injection of intermediate-high dose interleukin-2 for 2 years after autologous stem cell transplantation in children with stage IV neuroblastoma. Transplantation 2000; 70: 1100–1104.

  28. 28

    Burns LJ, Weisdorf DJ, DeFor TE et al. Enhancement of the antitumor activity of a peripheral blood progenitor cell graft by mobilization with interleukin 2 plus granulocyte colony-stimulating factor in patients with advanced breast cancer. Exp Hematol 2000; 28: 96–103.

  29. 29

    Meropol NJ, Porter M, Blumenson LE et al. Daily subcutaneous injection of low-dose interleukin 2 expands natural killer cells in vivo without significant toxicity. Clin Cancer Res 1996; 2: 669–677.

  30. 30

    Snedecor G, Cocharn W . Statistical Methods. Iowa State University Press: Iowa, 1989.

  31. 31

    Rosenberg SA, Lotze MT, Muul LM et al. A progress report on the treatment of 157 patients with advanced cancer using lymphokine-activated killer cells and interleukin-2 or high-dose interleukin-2 alone. N Engl J Med 1987; 316: 889–897.

  32. 32

    Cooper MA, Fehniger TA, Turner SC et al. Human natural killer cells: a unique innate immunoregulatory role for the CD56bright subset. Blood 2001; 97: 3146–3151.

  33. 33

    Gemlo BT, Palladino MA, Jaffe HS et al. Circulating cytokines in patients with metastatic cancer treated with recombinant interleukin 2 and lymphokine-activated killer cells. Cancer Res 1988; 48: 5864–5867.

  34. 34

    Holter W, Goldman CK, Casabo L et al. Expression of functional IL-2 receptors by lipopolysaccharide and interferon-gamma stimulated human monocytes. J Immunol 1987; 138: 2917–2922.

  35. 35

    Greenbaum LA, Horowitz JB, Woods A et al. Autocrine growth of CD4+ T cells. Differential effects of IL-1 on helper and inflammatory T cells. J Immunol 1988; 140: 1555–1560.

  36. 36

    Mills CD . Molecular basis of ‘suppressor’ macrophages: arginine metabolism via the nitric oxide synthesis pathway. J Immunol 1991; 46: 2719–2723.

  37. 37

    Tritarelli E, Rocca E, Testa U et al. Adoptive immunotherapy with high-dose interleukin-2: kinetics of circulating progenitors correlate with interleukin-6, granulocyte colony-stimulating factor level. Blood 1991; 77: 741–749.

  38. 38

    Bonig H, Laws HJ, Wundes A et al. In vivo cytokine responses to interleukin-2 immunotherapy after autologous stem cell transplantation in children with solid tumors. Bone Marrow Transplant 2000; 26: 91–96.

  39. 39

    Klingemann HG, Neerunjun J, Schwulera U, Ziltener HJ . Culture of normal and leukemic bone marrow in interleukin-2: analysis of cell activation, cell proliferation, and cytokine production. Leukemia 1993; 7: 1389–1393.

  40. 40

    Sondel PM, Hank JA . Combination therapy with interleukin-2 and antitumor monoclonal antibodies. Cancer J Sci Am 1997; 3 (Suppl. 1): S121–S127.

  41. 41

    Wigginton JM, Komschlies KL, Back TC et al. Administration of IL-12 with pulse interleukinn 2 and thhe rapid and complete eradication of murine renal carcinoma. J Natl Cancer Inst 1996; 88: 38–43.

  42. 42

    Koh CY, Blazar BR, George T et al. Augmentation of antitumor effects by NK cell inhibitory receptor blockade in vitro and in vivo. Blood 2001; 97: 3132–3137.

Download references


We thank RN Juliette Gay and RN Cindy Steele for providing assistance in patient care, RN Roby Nicklow, RN Eva Gallagher, Victoria Johnson, and Melissa Walker for data management, MD Fred Sanchez for patient care, and Maurice Wolin and Priscilla Ayers (Chiron). LJB is the recipient of a Department of Defense grant, and is also supported by the Janie Lymphoma Research Fund and the Minnesota Medical Foundation. This work was supported in part by National Institute of Health Grant PO1-CA-65493 (JSM), R21CA78904 (DJW) and Grant M01-RR00400 from the National Center for Research Resources. We also acknowledge the support of the Bone Marrow Transplant Research Fund of the University of Minnesota and the Masonic Order of the Eastern Star.

Author information

Correspondence to L J Burns.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Burns, L., Weisdorf, D., DeFor, T. et al. IL-2-based immunotherapy after autologous transplantation for lymphoma and breast cancer induces immune activation and cytokine release: a phase I/II trial. Bone Marrow Transplant 32, 177–186 (2003).

Download citation


  • immunotherapy
  • NK cells
  • IL-2

Further reading