Phase 1 dose-escalation trial of tremelimumab plus sunitinib in patients with metastatic renal cell carcinoma

Authors


Abstract

BACKGROUND:

On the basis of potential additive or synergistic immunostimulatory antitumor effects, in this phase 1 study, the authors evaluated the combination of sunitinib and tremelimumab (CP-675206; an antibody against cytotoxic T-lymphocyte–associated antigen 4 [CTLA4]) in patients with metastatic renal cell carcinoma (mRCC) was evaluated.

METHODS:

Adult patients with mRCC who had received ≤1 previous systemic treatment received tremelimumab (6 mg/kg, 10 mg/kg, or 15 mg/kg) intravenously once every 12 weeks and oral sunitinib (50 mg daily for 4 weeks then 2 weeks off or 37.5 mg daily as a continuous dose). The primary objective was to determine the maximum tolerated dose (MTD). Secondary objectives were to assess antitumor activity, safety, and pharmacokinetics.

RESULTS:

Twenty-eight patients were enrolled. Two of 5 patients who received 50 mg sunitinib plus tremelimumab 6 mg/kg experienced dose-limiting toxicities (DLTs), and no further enrollment to the combination with sunitinib 50 mg dosing was pursued. Among patients who received continuous sunitinib 37.5 mg daily, 1 of 7 patients who received tremelimumab 10 mg/kg plus sunitinib suffered a sudden death, and 3 of 6 patients who received tremelimumab 15 mg/kg plus sunitinib experienced DLTs. An expansion cohort (n = 7) was enrolled at tremelimumab 10 mg/kg plus sunitinib 37.5 mg daily; 3 of those patients experienced DLTs. Overall, rapid-onset renal failure was the most common DLT. Nine of 21 patients who were evaluable for response achieved partial responses (43%; 95% confidence interval, 22%-66%), and 4 of those responses were ongoing at the time of the current report.

CONCLUSIONS:

In this study of tremelimumab plus sunitinib, rapid-onset acute renal failure was observed unexpectedly, and further investigation of tremelimumab doses >6 mg/kg plus sunitinib 37.5 mg daily is not recommended. Preclinical investigation may be warranted to understand the mechanism of renal toxicity. Cancer 2011. © 2010 American Cancer Society.

Management options for metastatic renal cell carcinoma (mRCC) previously were limited to surgical resection or administration of cytokines, such as interleukin-2 or interferon-alfa (IFN-α). However, both of those cytokines are toxic when administered at high doses and are associated with clinical benefit in only a limited subset of patients.1 Sunitinib, an oral, multitargeted tyrosine kinase inhibitor of signaling through vascular endothelial growth factor receptor and platelet-derived growth factor receptor, has antitumor activity in mRCC. Compared with IFN-α, sunitinib significantly prolongs progression-free survival (PFS) (11 months vs 5 months; P < .001) and significantly increases objective response rates (31% vs 6%; P < .001).2 In addition to inhibiting angiogenesis, preclinical evidence suggests that sunitinib also may have immunostimulatory effects.3 Because 1 mechanism through which tumors escape recognition and destruction by the immune system is believed to be inadequate antigen presentation, treatment with sunitinib may enhance response to immunotherapy.3 One study in patients with mRCC who received sunitinib at a dose of 50 mg daily demonstrated a significant improvement in the immunostimulatory, T-helper type 1 response, as indicated by the increased percentage of T cells that IFN-γ measured in the peripheral blood (P = .001) on Day 28 compared with baseline.4 There also was a reduction in the immunosuppressive T-helper type 2 response, as indicated by diminished interleukin-4 production (P = .05) and regulatory T cells.4 In addition, myeloid-derived suppressor cells were reduced, and the reduction was correlated with type 1 T-cell augmentation.5 Collectively, these data indicated that sunitinib may modulate antitumor immunity and provided a rationale for the combination of sunitinib and immunotherapy for mRCC.

Historically, the known immunogenicity of RCC supported the use of immunotherapy in mRCC.6 Until recently,2 immunotherapeutic regimens with recombinant human IFN-α2b, used alone or in combination, were applied.6 The use of cytokines, however, has been limited by toxicity and generally poor overall response rates6; and as noted above, PFS was longer and response rates were higher in patients who received sunitinib than in patients who received IFN-α.2 However, the rationale for an immunotherapeutic approach to mRCC treatment remains.

Tremelimumab (CP-675206) is a fully human monoclonal antibody that blocks signaling through cytotoxic T-lymphocyte–associated antigen 4 (CTLA4), thus enhancing T-cell activation and proliferation. Immunotherapy with tremelimumab has been associated with durable objective responses in patients with metastatic melanoma.7-10

On the basis of the potential for additive or synergistic immunostimulatory antitumor effects, we hypothesized that the combination of tremelimumab plus sunitinib might have greater antitumor activity than either agent alone. Therefore, we investigated the safety and tolerability of escalating doses of tremelimumab in combination with 2 dosing schedules of sunitinib in this phase 1 study of mRCC.

MATERIALS AND METHODS

Patients

Adult patients with histologically documented mRCC who had received no more than 1 previous systemic therapy regimen were eligible. All patients had good Eastern Cooperative Oncology Group (ECOG) performance status (0 or 1). Any treatment-related toxicities from a previous therapy or surgical procedure must have resolved to grade 0 or 1 according to version 3.0 of the National Cancer Institute Common Terminology Criteria for Adverse Events.11 Patients were required to have evidence of adequate organ function determined within 21 days of enrollment (serum aspartate aminotransferase and serum alanine aminotransferase levels ≤2.5 times the upper limit of normal [ULN] or ≤5 times the ULN with liver metastases, total serum bilirubin ≤1.5 times the ULN, absolute neutrophil count ≥1.5 × 109 cells/L, platelets ≥100 × 109 cells/L, hemoglobin ≥8 g/dL without the support of growth factors, serum calcium ≤12.0 mg/dL, serum creatinine ≤1.5 times the ULN, and left ventricular ejection fraction ≥the lower limit of normal as assessed by multigated acquisition scan). Any major surgery, radiation therapy, or systemic therapy (including chemotherapy, dendritic cell-based immunotherapy, vaccines, or other biologic therapy) must have been completed ≥4 weeks before screening. Prior treatment with sunitinib or temsirolimus was not allowed. Patients with brain metastases or leptomeningeal disease were excluded. Patients also were excluded if they fell under the following categories: received treatment on another clinical trial within 21 days of enrollment; any grade 3 hemorrhage within 4 weeks of enrollment; concurrent malignancy that might require treatment during study; uncontrolled hypertension; history of chronic autoimmune disease; treatment with warfarin within 5 days of enrollment; ongoing cardiac dysrhythmia grade ≥2; atrial fibrillation; a QTc interval >450 msec for men or >470 msec for women; myocardial infarction, severe/unstable angina, coronary/peripheral artery bypass graft, congestive heart failure, cerebrovascular accident, transient ischemic attack, or pulmonary embolism within 12 months before enrollment; concomitant treatment with a cytochrome P450 3A4 inhibitor or inducer or with any drug that has proarrhythmic potential; a history of chronic inflammatory bowel disease, celiac disease, acute colitis, diverticulitis, or any chronic gastrointestinal condition associated with diarrhea; received treatment with immunosuppressants within 4 weeks of enrollment; known acquired immunodeficiency syndrome-related illness; or currently pregnant or lactating.

This study was conducted in compliance with the Declaration of Helsinki and its amendments and relevant International Conference on Harmonization of Good Clinical Practice guidelines and in agreement with the local institutional review board, independent ethics committee, or research ethics board. All study participants provided written informed consent before participating in the trial. This trial is registered as National Clinical Trial NCT00372853 (www.clinicaltrials.gov).

Study Design

In this phase 1, open-label, multicenter, dose-escalation study, we evaluated the safety and tolerability of escalating doses of tremelimumab in combination with sunitinib in patients with mRCC. Patients received oral sunitinib at a dose of either 50 mg daily for 4 weeks on an every-6-week cycle or 37.5 mg daily without any interruptions in combination with intravenous tremelimumab (6 mg/kg, 10 mg/kg, or 15 mg/kg) on Day 1 of every 90-day cycle (every 12 weeks) (Fig. 1). The first cohort of patients received sunitinib 37.5 mg daily plus 6 mg/kg tremelimumab. Once that cohort was filled, the next cohort of patients received 50 mg/kg sunitinib plus 6 mg/kg tremelimumab. Subsequent cohorts were assigned to the sunitinib dosing regimen based on the safety observed in the previous group of patients that received the same sunitinib regimen, whereas the tremelimumab dose was escalated to the next defined level. Patients were not enrolled in subsequent dose levels until all current dose level patients (≥3 patients) were observed for at least 6 weeks for possible dose-limiting toxicities (DLTs). Patients who did not experience disease progression or intolerable toxicity received a maximum of 6 tremelimumab infusions. Sunitinib monotherapy could continue after this time point at the discretion of the treating physician.

Figure 1.

The study schema is shown. RCC indicates renal cell carcinoma; IV, intravenous; Q12W, every 12 weeks; Q6W, every 6 weeks; MTD, maximum tolerated dose.

In this study, DLTs were defined as any of the following: any grade 4 nonhematologic treatment-related toxicity, any grade ≥3 adverse event (AE) despite medical therapy (except skin rash that did not require immunosuppressive therapy, hypophosphatemia, or grade 4 uric acid without clinical symptoms), any grade ≥3 AE that was potentially immune mediated and involved critical organs, severe hematologic toxicity (grade 4 decrease in neutrophils for ≥7 days, febrile neutropenia sustained over a 24-hour period, neutropenic infection, or grade 4 decrease in platelets for ≥7 days), or the inability to receive at least 75% of the planned doses of sunitinib because of treatment-related toxicity. Patients who experienced DLTs during the first 6 weeks after the first dose of tremelimumab completed treatment at that level to determine the next dose level; dose escalation did not occur until all patients at a given level had been observed for at least 6 weeks.

In this study, we used a standard 3 + 3 design. If no DLTs were observed among 3 patients in the first 6 weeks, then the dose was to be escalated to the next higher level. If 1 DLT was observed among 3 patients in the first 6 weeks, then 3 additional patients were to receive treatment at the same dose level. If 1 of those 6 patients experienced a DLT in the first 6 weeks, then dose escalation was to continue. If ≥2 of those 6 patients (33%) experienced a DLT in the first 6 weeks, then the next lower dose level was to be considered the maximum tolerated dose (MTD) provided at least 6 patients had been treated and followed for 6 weeks at that level, and dose escalation would be discontinued. If ≥2 of 3 to 6 patients (33%) had DLTs during the first 6 weeks at the starting dose level, then further investigation of tremelimumab in combination with the relevant dose and schedule of sunitinib was to be halted.

The dose and schedule of sunitinib was based on a regimen that was approved for use in mRCC (50 mg daily for 4 weeks followed by a 2-week rest period) and an alternative, continuous daily dosing regimen that was used in a phase 2 study in patients with RCC.12 The doses of tremelimumab that we tested in the current study were based on doses that were used on phase 1 and 2 studies in patients with melanoma.7, 10

To provide pharmacokinetic (PK) data with which to evaluate the effect of tremelimumab administration on the PK parameters of sunitinib, a cohort of 12 patients was to be enrolled in an expansion cohort at the MTD. In this cohort, samples were taken on Day 28 of Cycle 1 to evaluate the levels of sunitinib and its metabolite. In addition, samples were taken to determine the PK parameters of tremelimumab on a schedule that would permit the determination of single-dose parameters. These procedures were planned to evaluate the effect of sunitinib administration on the PK parameters of tremelimumab. The expansion cohort also was used to collect further safety data at the MTD.

The primary objective was to determine the MTD of tremelimumab when it was administered every 12 weeks in combination with 2 different sunitinib dose regimens. Secondary objectives included assessments of any preliminary clinical evidence of antitumor activity, safety, and PK parameters.

Assessments

Baseline screening was performed within 21 days of study entry and included complete medical history, physical examination, body weight, ECOG performance status, ophthalmologic assessment, 12-lead resting electrocardiogram, hematology, blood chemistry, urinalysis, clinical evaluation of autoimmunities, and pregnancy test. Patients were assessed for AEs at baseline and on Days 1, 14, 28, 42, and 56 during Cycle 1 and on Days 1, 28, and 42 of all subsequent cycles. Computed tomography (CT) or magnetic resonance imaging (MRI) studies of the chest, abdomen, and pelvis were obtained to assess tumors at screening, on Day 42 of Cycle 1, and within 10 days before each dose of tremelimumab. Scans also were obtained to confirm any response within 4 weeks of initial documentation and at the end of treatment or withdrawal from study. PK parameters were to be assessed in 12 patients from the expansion cohort who received sunitinib 37.5 mg daily and tremelimumab 10 mg/kg every 12 weeks to determine the sunitinib steady state and the 10 mg/kg single-dose tremelimumab PK parameters.13

Statistical Analyses

The 3 + 3 dose-escalation design was selected to determine the MTD. The sample size for this phase 1 study was estimated for a typical 3 + 3 dose-escalation design. Listings and summary statistics were calculated for continuous variables (eg, means, medians) and for discrete variables (eg, frequency, percentages). Confidence intervals (CIs) for the median time-to-event endpoints (eg, duration of response, PFS, and overall survival) were calculated using the method of Brookmeyer and Crowley.

RESULTS

Patients

In total, 28 patients (21 patients in the main study and 7 patients in an expansion cohort) were enrolled at 4 sites between December 2006 and January 2009. Most patients were men (22 of 28 patients; 79%), and were aged <65 years (18 of 28 patients; 64%). The majority of patients (26 of 28 patients; 93%) had clear cell histology (Table 1).13 All patients had stage IV RCC at study entry, and most patients (24 of 28 patients; 86%) had multiple involved disease sites. Three patients (11%) had received prior systemic treatment for their disease.

Table 1. Baseline Demographics and Disease Characteristics (N=28)a
Demographics/CharacteristicsNo. of Patients (%)
  • ECOG indicates Eastern Cooperative Oncology Group.

  • a

    Adapted from Gordon MS, Stein M, Shannon P, et al. Phase I dose escalation trial of tremelimumab plus sunitinib in patients (pts) with metastatic renal cell carcinoma (mRCC) [poster presentation]. Presented at: 45th Annual Meeting of the American Society of Clinical Oncology (ASCO); May 29 to June 2, 2009; Orlando, Fla; Abstract 5115.13

Mean age [range], y60 [39-77]
Sex
 Men22 (79)
 Women6 (21)
 Mean disease duration [range], y2 [0-13]
ECOG performance status
 018 (64)
 110 (36)
Histologic classification
 Predominantly clear cell26 (93)
 Mixed clear cell with papillary features1 (4)
 High-grade sarcomatoid1 (4)
Patients who received previous cancer therapy
 Bevacizumab3 (11)
 Interleukin-21 (4)
 Interleukin-211 (4)
 Sorafenib2 (7)
 Patients who underwent previous nephrectomy24 (86)
 Patients who received previous radiation1 (4)
Involved disease sites
 Lung21 (75)
 Unknown/other12 (43)
 Lymph node, distant8 (29)
 Lymph node, regional7 (25)
 Adrenal4 (14)
 Kidney4 (14)
 Liver4 (14)
 Mediastinum2 (7)
 Brain1 (4)

Safety

Across treatment groups (n = 28), the median number of cycles received was 2 (range, 1-6 cycles received). Five patients (18%) received ≥5 treatment cycles. All patients developed treatment-related AEs. Ten patients (36%) had serious AEs, which were defined as any untoward medical occurrences that, at any dose, resulted in death; were life-threatening, required inpatient hospitalization or prolongation of existing hospitalization, resulted in persistent or significant disability/incapacity, or resulted in intrauterine exposure. Seventeen patients (61%) had grade 3 or 4 AEs. In the initial cohort, (sunitinib 50 mg daily plus tremelimumab 6 mg/kg), 2 of 5 patients had DLTs (Table 2).13 These DLTs included grade 3 fatigue/mucositis (n = 1) and grade 5 perforated diverticula resulting in death (n = 1). This exceeded the defined MTD criteria for this regimen, and further exploration of this dosing regimen was halted. The study then enrolled patients at sunitinib 37.5 mg daily (continuous dosing) plus tremelimumab. None of the 3 patients who received sunitinib plus 6 mg/kg tremelimumab experienced DLT, and the tremelimumab dose was escalated to 10 mg/kg. At this dose level, 1 patient had a DLT (sudden death), and the tremelimumab dose was escalated further. At 15 mg/kg tremelimumab plus sunitinib, 3 of 6 patients (50%) had a defined DLT. These DLTs included 2 patients with grade 3 renal events (renal failure and serum creatinine [1 patient each]) and grade 3 dyspnea in a patient who had dyspnea at baseline. Given the frequency and severity of the toxicities observed, an expansion cohort of 7 patients received sunitinib 37.5 mg daily plus tremelimumab 10 mg/kg, which was the protocol-defined MTD based on the initial cohort, to provide additional PK parameters and safety information. Among all 14 patients who received 10 mg/kg tremelimumab every 12 weeks plus sunitinib 37.5 mg daily in the initial and expansion cohorts, there were 4 DLTs: renal failure (n = 2), grade 2 colitis (n = 1), and sudden death (n = 1). The frequency of DLTs, including renal failures, at this dose level indicated that it was not the MTD based on the prespecified definition in the study design.

Table 2. Dose-Limiting Toxicitiesa
DLTNo. of Patients (%)
Sunitinib 37.5 mg/d PlusSunitinib 50.0 mg/d Plus Tremelimumab 6 mg/kg, n=5
Tremelimumab 6 mg/kg, n=3Tremelimumab 10 mg/kg, n=14bTremelimumab 15 mg/kg, n=6
  • DLT indicates dose-limiting toxicity.

  • a

    Adapted from Gordon MS, Stein M, Shannon P, et al. Phase I dose escalation trial of tremelimumab plus sunitinib in patients (pts) with metastatic renal cell carcinoma (mRCC) [poster presentation]. Presented at: 45th Annual Meeting of the American Society of Clinical Oncology (ASCO); May 29 to June 2, 2009; Orlando, Fla. Abstract 5115.13

  • b

    Seven of these patients were part of an expansion cohort at this dose level.

Any DLT04 (29)3 (50)2 (40)
Renal failure 21 
Elevated creatinine  1 
Colitis 1  
Sudden death 1  
Dyspnea  1 
Fatigue/mucositis   1
Perforated diverticula   1

Acute renal failure

Acute renal failure was the most common DLT. The 4 patients who had renal failure had received tremelimumab at the 2 highest dose levels tested: 10 mg/kg and 15 mg/kg every 12 weeks (Table 3).13 The time to onset for all 4 renal failure events was approximately 2 weeks (median, 14.5 days; range, 11-16 days) after the start of treatment. All patients required extensive medical management.

Table 3. Summary of Renal-Associated Dose-Limiting Toxicitiesa
Sunitinib 37.5 mg/d PlusSexAge, yHeight, cm/ Weight, kgEventOnsetDuration, dAccompanying AEsKey ManagementCurrent Status
  • DLTs indicates dose-limiting toxicities; G, grade; AEs, adverse events; AST, aspartate aminotransferase.

  • a

    Adapted from Gordon MS, Stein M, Shannon P, et al. Phase I dose escalation trial of tremelimumab plus sunitinib in patients (pts) with metastatic renal cell carcinoma (mRCC) [poster presentation]. Presented at: 45th Annual Meeting of the American Society of Clinical Oncology (ASCO); May 29 to June 2, 2009; Orlando, Fla. Abstract 5115.13

  • b

    The patient withdrew consent, and additional follow-up is not available.

Tremelimumab 10 mg/kg (n=14)
 Patient 10011062Man56178/95Acute renal failure G4Study D 14; Cycle 1, D 1029bFever/chills, diarrhea G1/anorexia G2, dehydration G3, AST G1, rash, thrombocytopenia G2, fatigueSteroids, dialysisStable on dialysis
 Patient 10061062Man71188/107Acute renal failure G3Study D 16; Cycle 1, D 1414Fever/flu-like symptoms, dehydration G2, orthostatic vitals, tubulointerstitial nephritisDialysis, plasmapheresis, steroidsResolved with sequelae of baseline elevated creatinine
Tremelimumab 15 mg/kg (n=6)
 Patient 10011013Man58185/108Acute renal failure G3Study D 15; Cycle 1, D 11228Diarrhea G3, fever, infection (diverticulitis), edemaAntibioticsResolved with sequelae of baseline elevated creatinine
 Patient 1002100Man61183/102G3 elevated creatinineStudy D 11; Cycle 1, D 96Fever, diarrhea G2, elevated liver function test resultsSteroidsA rechallenge with sunitinib 4 mo later led to elevated creatinine

The first patient who developed acute renal failure presented with fever and diarrhea 12 days after treatment with sunitinib 37.5 mg sunitinib and 15 mg/kg tremelimumab. At baseline, his creatinine level was 1.4 mg/dL, and this increased to 2.1 mg/dL on Day 12. On study Day 16, CT scans of the abdomen diagnosed infectious diverticulitis. The patient's creatinine level peaked at 7.1 mg/dL on Day 18. Treatment for renal failure was intravenous fluids and bicarbonate supplementation. The patient recovered completely from infection. Seven months after going off study, the patient had a creatinine level of 1.60 mg/dL and was retreated with 50 mg sunitinib (4 weeks on, 2 weeks off 4/2 schedule). After 1 month of treatment with sunitinib alone, his creatinine level was 1.52 mg/dL. This patient remains on treatment with single-agent sunitinib, and his creatinine levels have ranged from 1.29 mg/dL to 1.67 mg/dL.

The second patient who developed renal failure had received sunitinib 37.5 mg daily and 15 mg/kg tremelimumab. On study Day 11, he was admitted to hospital for grade 2 diarrhea and dehydration. Blood and stool cultures were negative. The patient's creatinine level became elevated; on study Day 11, his creatinine level was 2.7 mg/dL, and it rose to a peak of 4.7 mg/dL on study Day 13. Study drugs were discontinued permanently; and, with hydration and supportive care, creatinine levels steadily improved, and the patient recovered to stable condition with creatinine remaining between 1.6 mg/dL and 2.0 mg/dL. Approximately 4 months after the last dose of tremelimumab, this patient was rechallenged with sunitinib alone outside of this protocol, because evidence of tumor shrinkage was noted during the combination treatment. At that time, the patient's baseline creatinine had recovered to 1.6 mg/dL. Within 2 weeks of restarting sunitinib, the patient's creatinine began to increase. Treatment with sunitinib was discontinued. When creatinine increased during this rechallenge, an off-study, unplanned sample was collected for tremelimumab PK analyses. The sample, which was collected 161 days after treatment with tremelimumab, had 0.88 μg/mL tremelimumab, which was within the range expected (median, 1.26 μg/mL; range, 0-2.52 μg/mL) based on previous studies with single-agent tremelimumab. Therefore, it was the investigator's opinion that the initial creatinine elevation and renal failure were secondary to sunitinib treatment.

The third patient who developed renal failure was admitted to the emergency room 11 days after treatment with sunitinib 37.5 mg daily and 10 mg/kg tremelimumab. He had moderate-to-severe chills, fever, and a full body rash. Upon admission, he had an elevated creatinine level (6.31 mg/dL), thrombocytopenia (51 × 109/L), and elevated aspartate aminotransferase and bilirubin levels. No tests were performed to determine whether thrombocytopenia was immune-mediated. His creatinine level rose to a peak of 9.58 mg/dL during hospitalization. Intravenous fluids and steroids were administered, and dialysis was started. The patient developed an allergic reaction to the dialysis membrane, and this resulted in atrial fibrillation. Conservative management with rate control and heparin were used, but the patient went into respiratory distress and flash pulmonary edema. Intubation relieved the symptoms, but the patient had aphasia and weakness on the right side after extubation. An intracerebral hemorrhage was identified. Treatment with anticoagulants ceased, and the platelet count was maintained at ≥100,000/μL. Additional treatment with dexamethasone and levetiracetam was initiated, and the patient was continued on intermittent dialysis for elevated creatinine. The patient discontinued the study, withdrew consent, and still was recovering from the acute renal failure, central nervous system hemorrhage, and thrombocytopenia at the time of this report. At the time the patient withdrew consent, the his creatinine level was 3.94 mg/dL, and his platelet count was 122 × 109/L.

The fourth who developed renal failure presented with fever and dehydration 14 days after treatment with sunitinib 37.5 mg daily and 10 mg/kg tremelimumab in the first cycle. The patient's serum creatinine rose to 3.2 mg/dL. Blood urea nitrogen was 37 mg/dL, and calcium was 7.2 mg/dL. The patient was hospitalized and underwent a right kidney biopsy, which revealed acute, tubular, interstitial nephritis with eosinophilia consistent with a hypersensitivity reaction (Fig. 2). The patient received amlodipine for hypertension, cefotaxime and cephalexin for a urinary tract infection, prednisone for interstitial nephritis, and epoietin alfa for anemia. The patient underwent plasmapheresis and hemodialysis for 6 days. Although the patient still was taking oral medications for hypertension and interstitial nephritis, he was discharged, and his serum creatinine recovered to 1.72 mg/dL approximately 6 weeks later. Within 3 months of discharge, his creatinine level and platelet count had recovered to normal levels of 1.4 mg/dL and >200,000/μL, respectively. This patient was retreated later with sunitinib 37.5 mg daily alone and continued for 2 full cycles (4 weeks on, 2 weeks off) without developing recurrent toxicity of any clinical significance.

Figure 2.

This image is from a kidney biopsy in the fourth patient who developed acute renal failure and hypersensitivity reaction. The patient received sunitinib 37.5 mg daily and tremelimumab 10 mg/kg. Fourteen days after the first treatment cycle, this patient presented with fever and dehydration. The patient's serum creatinine rose to 3.2 mg/dL, blood urea nitrogen was 37 mg/dL, and calcium was 7.2 mg/dL. The image shown is from a biopsy of the right kidney, which revealed acute, tubular, interstitial nephritis and eosinophilia consistent with a hypersensitivity reaction. This patient recovered with clinical sequelae and later was retreated with sunitinib alone, as described in the text.

Ten patients had estimated baseline creatinine clearance rates <60 mL per minute, 1 patient had a rate of 60 mL to 65 mL per minute, and 17 patients had rates >65 mL per minute. Estimated creatinine clearance rates for the patients who developed renal failure were as follows: first patient, 109.56 mL per minute; second patient, 68.88 mL per minute; third patient, 81.05 mL per minute; and fourth patient, 58.40 mL per minute.

Sudden death

A woman aged 67 years with RCC received 1 cycle of tremelimumab (10 mg/kg; cumulative dose, 1055 mg) and sunitinib (37.5 mg daily; cumulative dose, 450 mg). This patient also received concomitant medications for several comorbid conditions, including asthma, hypertension, diabetes, and upper respiratory infection. On Day 11, the patient reported cough with sputum production and sinus congestion, which was treated with oral levofloxacin and homatropine/hydrocodone syrup. On Day 13, the patient was discovered not breathing and was unresponsive; she was transported to hospital, where she was pronounced dead. The investigator considered it possible that the patient's sudden death was related to tremelimumab and sunitinib therapy. However, the patient had several risk factors for cardiovascular events and was on multiple relevant therapies. Levofloxacin is a quinolone known to cause moderate QT prolongation and Torsade de Pointes (TdP). The combination of levofloxacin and a diuretic may have caused a reduction in potassium levels, TdP, and sudden death.

Colitis

A man aged 59 years who received a single dose of tremelimumab 10 mg/kg and intermittent sunitinib 37.5 mg daily experienced colitis. On Day 28, the patient developed grade 2 diarrhea, grade 2 fatigue, grade 1 abdominal distention and pain, grade 1 intermittent rectal bleeding, and grade 1 anorexia. A colonoscopy revealed diffuse colitis with inflammation; dysplastic polyps, adenomas, and chronic active colitis were evident on biopsy. In addition to loperamide for diarrhea, the patient received midazolam and fentanyl. Colitis resolved by Day 40. The investigator considered colitis to be a DLT for tremelimumab. The patient permanently was discontinued from receiving additional tremelimumab but resumed sunitinib therapy after his colitis resolved. A partial response that was documented before Cycle 2 was sustained at study Day 90.

Pharmacokinetics

Because the expansion cohort was terminated early, only the maximum concentration (Cmax) values of tremelimumab after the first dose were determined. Trough concentrations of sunitinib and its active metabolite, SU012662, that were achieved 28 days after the first dose were analyzed and summarized. The median (range) Cmax value of tremelimumab after the first dose was 193 μg/mL (range, 168-438 μg/mL) for patients who received 6 mg tremelimumab (n = 4), 230 μg/mL (range, 60-336 μg/mL) for patients who received 10 mg tremelimumab (n = 12), and 397 μg/mL (range, 264-428 μg/mL) for patients who received 15 mg tremelimumab (n = 6). The median (range) trough concentration of sunitinib and its active metabolite, SU012662, (combined) on Day 28 was 47.5 ng/mL (range, 4.8-69.2 ng/mL) for patients who received sunitinib 37.5 mg daily (n = 13) and 49.5 ng/mL (range 11.1-173.2 ng/mL) for patients who received sunitinib 50 mg daily for 4 weeks followed by 2 weeks of rest (n = 4). In 3 of the 4 patients who developed serious renal events, the Cmax values for tremelimumab were 251 μg/mL (10 mg/kg tremelimumab), 420 μg/mL (15 mg/kg tremelimumab), and 264 μg/mL (15 mg/kg tremelimumab dose), respectively. No sample for PK analysis was collected from the fourth patient. Although the data are limited, the serious renal events did not appear to be associated with high tremelimumab concentrations. Trough concentrations of sunitinib plus SU012662 were not available in patients who had serious renal events, because those events occurred before Day 28, which was the scheduled sample collection time for sunitinib PK analyses. Thus, the relation between serious renal events and sunitinib concentrations could not be assessed.

Efficacy

Nine of 21 patients (43%; 95% CI, 22%-66%) who were evaluable for response had a partial response to treatment with tremelimumab plus sunitinib.13 Six of 9 responses occurred in patients who received 10 mg/kg tremelimumab plus sunitinib 37.5 mg daily (continuous dosing). The duration of the 6 partial responses ranged from ≥2.8 months to ≥18.2 months (4 responses were ongoing at the time of this report; the median response duration was not achieved). Seven patients (33%) had a best overall response of stable disease, and 2 of those patients maintained stable disease for >6 months. Overall, 16 of 21 patients (76%) who were evaluable for response who had data available had tumor shrinkage in target lesions compared with baseline in at least 1 assessment (Fig. 3).

Figure 3.

A waterfall plot is shown. The greatest percentage change in target tumor lesion size is presented for each patient who had tumor measurements available (n = 21). In patients who had multiple target lesions, the sum of lesion measurements was used. PR indicates partial response; SD, stable disease; PD, progressive disease.

DISCUSSION

In the current study, tremelimumab was investigated in combination with 2 different schedules of sunitinib in patients with mRCC. In patients with mRCC who received sunitinib 50 mg daily day for 4 weeks followed by 2 weeks off, the MTD for tremelimumab every 12 weeks was <6 mg/kg was not defined. In patients with mRCC who received sunitinib 37.5 mg daily, the MTD for tremelimumab every 12 weeks was 10 mg/kg; however, based on the unacceptable toxicity experienced in the expansion cohort, the MTD is not recommended for phase 2 investigation.

The most common DLT was acute renal failure, which occurred approximately 2 weeks after starting treatment in both of the highest tremelimumab dosing cohorts. Currently, there is limited insight into the mechanism of the renal toxicity we observed in this study, particularly because acute renal failure is not a common toxicity for either of the study drugs when administered alone. In studies of single-agent tremelimumab in nearly 1000 patients, there have been 11 serious AEs of renal failure or similar events (Pfizer; unpublished results). Five of those events were considered tremelimumab treatment-related, and all of them were prerenal failures associated with diarrhea/hypovolemia and were not a direct drug effect on the kidney. Five other events were considered unrelated to treatment with tremelimumab, and an additional event remains under review. Treatment-related renal failure (grade ≥3) has been reported in 0.4% of patients (n = 34) and acute renal failure has been reported in 0.1% of patients (n = 12) who were treated on previous studies of sunitinib (N = 8709; Pfizer; unpublished results). The onset time for these renal toxicities coincided with the expected time to steady-state concentration for sunitinib; however, the steady-state levels of sunitinib that were achieved in combination with tremelimumab were within the expected range for single-agent sunitinib. Fever accompanied all acute renal failure events, and this may have been indicative of an immune-related mechanism. This limited evidence could support an immune-mediated interaction between sunitinib and tremelimumab.

It was possible to determine only the tremelimumab Cmax values and the sunitinib trough values from the limited samples that were available in this study. Trough concentrations of sunitinib plus SU012662 and Cmax values of tremelimumab were variable, but they were within the expected range as determined in previous studies of sunitinib14 and tremelimumab (Pfizer; unpublished results). Serious renal events did not appear to be associated with high tremelimumab concentrations. It was not possible to determine from available samples if the concentration of sunitinib was associated with renal toxicities. However, among the patients who did not have serious renal events, only 1 patient had a trough concentration of sunitinib plus SU012662 >100 ng/mL, which is the lower limit of concentrations often associated with dose-limiting toxicities,14 and this patient did not have a serious AE. Therefore, the kinetic parameters measured for sunitinib and SU012662 in that study did not appear to be related to the renal toxicities reported in our study.

Partial responses, some of which were durable, were observed in 9 of 21 patients (43%) who were evaluable for response. This response rate was within the normal range of historic data for treatment with single-agent sunitinib and was somewhat higher than that observed in a phase 3 trial of sunitinib alone in 750 patients with previously untreated mRCC (31%).2 Although clinical activity was a secondary objective and was not fully assessable in the current small phase 1 study, the objective response rate was not increased obviously with the addition of tremelimumab in this preliminary analysis of efficacy. Several factors may have contributed to this observation. The dose and schedule of the 2 agents that we used in this study may not have been optimal for clinical synergy. Because the potential immunomodulatory mechanism of sunitinib remains to be elucidated, sunitinib in combination with immunotherapeutic agents remains a viable investigative endeavor.

Safety data suggest that the addition of tremelimumab increased the risk of grade ≥3 AEs. In the phase 3 study of sunitinib alone, the proportion of grade 3 or 4 AEs was 1% to 13%, and no renal toxicities were reported2; in the current study of sunitinib plus tremelimumab, 17 of 28 patients (61%) had grade 3 or 4 AEs, and 4 patients (14%) had renal toxicities. Although further investigation of doses >6 mg/kg tremelimumab plus sunitinib 37.5 mg daily is not recommended, further preclinical investigation of this combination may be warranted to fully elucidate the mechanism of renal toxicity.

Acknowledgements

We thank Isan Chen, PhD; Poe-Hirr Hsyu, PhD; and Nicola Wallis, MD of Pfizer Inc. for their roles in study conduct, data reviews, and article development. Medical editorial assistance was provided by Tamara Fink, PhD of ProEd Communications, Inc. and was funded by Pfizer Inc.

CONFLICT OF INTEREST DISCLOSURES

This study was sponsored by Pfizer Inc. Dr. Rini has served as a consultant for and received research grant funding from Pfizer Inc. Ms. Catlett, Ms. Healey, and Dr. Huang are employed by Pfizer Inc. Ms. Healey owns stock in Pfizer Inc. Dr. Gordon has received research grant funding from Pfizer Inc, and served in a consultant/advisory role (unrelated to article development) for Pfizer Inc. Dr. Stein has received honoraria for a Speakers' Bureau (unrelated to article development) from Pfizer Inc.

Ancillary