SEARCH

SEARCH BY CITATION

Keywords:

  • sorafenib;
  • expanded access program;
  • renal cell carcinoma;
  • progression-free survival;
  • bevacizumab;
  • brain metastases;
  • first-line;
  • nonclear cell

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES

BACKGROUND:

The Advanced Renal Cell Carcinoma Sorafenib (ARCCS) program made sorafenib available to patients with advanced renal cell carcinoma (RCC) before regulatory approval.

METHODS:

In this nonrandomized, open-label expanded access program, 2504 patients from the United States and Canada were treated with oral sorafenib 400 mg twice daily. Safety and efficacy were explored overall and in subgroups of patients including those with no prior therapy, nonclear cell (nonclear cell) RCC, brain metastases, prior bevacizumab treatment, and elderly patients. Sorafenib was approved for RCC 6 months after study initiation, at which time patients with no prior therapy or with nonclear cell RCC could enroll in an extension protocol for continued assessment for a period of 6 months.

RESULTS:

The most common grade ≥2 drug-related adverse events were hand-foot skin reaction (18%), rash (14%), hypertension (12%), and fatigue (11%). In the 1891 patients evaluable for response, complete response was observed in 1 patient, partial response in 67 patients (4%), and stable disease for at least 8 weeks in 1511 patients (80%). Median progression-free survival in the extension population was 36 weeks (95% confidence interval [CI], 33-45 weeks; censorship rate, 56%); median overall survival in the entire population was 50 weeks (95% CI, 46-52 weeks; censorship rate, 63%). The efficacy and safety results were similar across the subgroups.

CONCLUSIONS:

Sorafenib 400 mg twice daily demonstrated activity and a clinically acceptable toxicity profile in all patient subsets enrolled in the ARCCS expanded access program (clinicaltrials.gov identifier: NCT00111020). Cancer 2010. © 2010 American Cancer Society.

In 2008, 54,390 new cases of cancer of the kidney or renal pelvis were estimated to be diagnosed in the United States, with approximately 90% classified as renal cell carcinoma (RCC), and 13,010 patients were estimated to die of the disease.1, 2 The traditional 5-year survival rate for patients with metastatic RCC is estimated to be <10%.2, 3 At the time of diagnosis, 20% to 30% of RCC patients have metastatic disease, which is generally resistant to classic chemotherapy and radiation therapy.4 Before the introduction of vascular endothelial growth factor (VEGF) pathway- and mammalian target of rapamycin (mTOR)-targeted therapy, recombinant human interleukin-2 and recombinant human interferon alpha2b (IFN-α2b) either alone or in combination were the standard treatments.5 However, these agents have limited efficacy and are associated with significant toxicity.6-12

Recognition of the essential role of VEGF in RCC pathogenesis led to the testing of angiogenesis inhibitors in this disease.13 Sorafenib is a potent multikinase inhibitor of receptor tyrosine kinases VEGF receptors 1, 2, and 3 and platelet-derived growth factor receptors α and β, as well as the Raf/MEK/ERK pathway at the level of Raf kinase.14 In xenograft models, sorafenib administered as a single agent had potent antiangiogenic activity and was found to inhibit the growth of RCC tumors.15, 16

Sorafenib was investigated as monotherapy in 4 phase 1 trials17-20 and had an acceptable toxicity profile. In a phase 2 randomized discontinuation trial, sorafenib prolonged progression-free survival (PFS) compared with placebo in patients with metastatic RCC in whom previous treatment was ineffective.21 These positive results were subsequently confirmed in a large phase 3, randomized, placebo-controlled trial (Treatment Approaches in Renal Cancer Global Evaluation Trial [TARGET]) of sorafenib in 903 previously treated cytokine-refractory patients with metastatic RCC.22 Sorafenib therapy reduced the risk of disease progression by 56%; the median PFS with sorafenib (5.5 months) was twice that with placebo (2.8 months; P < .001).

The positive results in the TARGET trial substantiated the need to assess the use of sorafenib in patients who might benefit while it was under review by the US Food and Drug Administration (FDA) and not commercially available. The Advanced Renal Cell Carcinoma Sorafenib (ARCCS) expanded access program was therefore instituted in North America to make sorafenib available to patients with advanced RCC who had no access to or did not qualify for other sorafenib clinical trials. We report here the safety and efficacy of sorafenib in this large group of patients representative of those with advanced RCC in the community setting.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES

Patients

Eligible patients were ≥15 years old, had histologically confirmed advanced RCC, and an Eastern Cooperative Oncology Group performance status of 0 to 2, with waivers granted to selected patients with performance status 3 or 4. Unlike previous controlled clinical trials, this expanded access program included patients with nonclear cell pathology, patients with active or progressive metastatic brain disease despite previous treatment, those who had not received prior drug therapy for advanced RCC, and patients who had received therapy with bevacizumab at least 28 days before enrolling in this study (provided they had no grade ≥3 hemorrhagic episode during therapy, and no history of grade ≥2 hemorrhagic event within 6 months of bevacizumab therapy). Patients were excluded if they had treatment with other investigational drugs in the 4 weeks before enrollment, cardiac arrhythmias, active coronary artery disease or ischemia, uncontrolled hypertension, renal impairment requiring dialysis, recent or active bleeding diathesis, or life expectancy <2 months.

All patients provided written informed consent. The study was approved by the institutional review board at each center and complied with the provisions of the Declaration of Helsinki, Good Clinical Practice guidelines, and local laws and regulations.

Study Design

This nonrandomized, open-label expanded access program was conducted at 319 centers in the United States and at 8 centers in Canada (clinicaltrials.gov identifier: NCT00111020). Enrollment in the United States started in June 2005 and ended in December 2005 when the FDA approved the use of sorafenib for the treatment of advanced RCC, and it became commercially available. In Canada, enrollment took place between June 2005 and July 2006. Patients received oral sorafenib 400 mg twice daily on a continuous basis. Sorafenib tablets were supplied by Bayer HealthCare Pharmaceuticals (West Haven, Conn). Therapy was discontinued if the patient developed unacceptable toxicity, if progression of disease occurred, or if, in the investigator's judgment, the patient was unlikely to benefit from further treatment with sorafenib. Toxicity-related dose modifications of sorafenib were performed in accordance with protocol-specified guidelines.

Because there was no central pathologic review and the original case report forms included the choice “other” rather than “unclassified,” 1 member of the steering committee reviewed pathology reports from a total of 226 patients (162 other, 11 no subtype recorded, 50 nonclear cell, 3 clear cell). Among the nonclear cell reports reviewed, histologic coding was changed in 25, typically to unclassified (22 cases). Patients with “unclassified” histology or those coded as “other” in whom the pathology report could not be reviewed were not included in the nonclear cell category for analysis.

Initial study phase

A baseline radiologic assessment of tumor burden was completed within 28 days before enrollment. Radiologic and/or clinical assessment of the target tumor was repeated at 8-week intervals or in accordance with the local standard of care. Clinical assessments were used only if radiologic assessments were not available. There was no validation of the investigator's assessment of disease progression by an independent radiologist or oncologist. When the expanded access program was discontinued, the protocol specified follow-up for overall survival (OS) every 3 months; however, compliance with this requirement was poor and, at the time of final data analysis, the median follow-up time was only 22.4 weeks. Patients who remained on sorafenib therapy when the program ended were switched to commercial product or enrolled into the extension phase.

Six-month extension phase

When the expanded access program ended in the United States with FDA approval of sorafenib, patients who had not received prior antineoplastic therapy for advanced RCC and/or who were diagnosed with nonclear cell RCC were eligible to enter a 6-month extension study. During the 6-month extension, visit procedures and reporting were the same as during the initial study phase. Tumor assessments and radiologic evaluations were performed every 8 weeks, as described. At the conclusion of the extension phase, patients were observed every 3 months for OS.

Safety and Efficacy Assessments

Assessment of adverse events was based on clinical and laboratory evaluations considered as standard of care, and any reported adverse events were graded with version 3.0 of the National Cancer Institute Common Terminology Criteria for Adverse Events. The best overall response rate was assessed using Response Evaluation Criteria in Solid Tumors (RECIST) guidelines.23 In this study, PFS was defined as the time from administration of the first dose of sorafenib to disease progression or death. OS was defined as the time from administration of the first dose of sorafenib to death and was calculated for all patients 18 months after FDA approval of sorafenib. Kaplan-Meier analyses were performed for PFS and OS.

Safety and efficacy data were evaluated separately for 2 prespecified patient groups: first-line patients with no prior antineoplastic therapy for advanced RCC and those who received prior therapy. In addition, post hoc analyses of safety and efficacy were completed for subgroups of patients who are typically not enrolled in sorafenib clinical trials: patients with nonclear cell RCC, brain metastases, or prior bevacizumab treatment, and patients ≥70 years of age.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES

Between June 2005 and July 2006, 2515 patients with advanced RCC were enrolled in the ARCCS program in North America; 2504 patients (2298 in the United States and 206 in Canada) received at least 1 dose of sorafenib and were included in safety and efficacy analyses. Of these patients, 328 participated in the extension phase of the program. Demographic and disease characteristics in first-line patients were similar to those with prior therapy (Table 1). Among patients with brain metastasis, 36 had no prior treatment for advanced RCC, and 34 had prior treatment. Patient disposition is shown in Figure 1.

thumbnail image

Figure 1. Patient disposition is shown. ARCCS indicates Advanced Renal Cell Carcinoma Sorafenib (ARCCS) program; NCC, nonclear cell.

Download figure to PowerPoint

Table 1. Patient Characteristics
CharacteristicPatient SubgroupsAll Patients, N=2504
First-Line, n=1254Previously Treated, n=1250Nonclear Cell, n=202Brain Metastases, n=70Prior Bevacizumab, n=290Age ≥70 Years, n=736
Sex, No. (%)       
 Men874 (70)860 (69)143 (71)51 (73)188 (65)464 (63)1734 (69)
 Women379 (30)380 (30)57 (28)18 (26)99 (34)269 (37)759 (30)
 Missing1 (<1)10 (1)2 (1)1(<1)3 (1)3 (<1)11 (<1)
Race, No. (%)       
 White1133 (90)1098 (88)174 (86)64 (91)254 (88)685 (93)2231 (89)
 Black47 (4)55 (4)16 (8)3 (4)15 (5)10 (1)102 (4)
 Hispanic31 (3)54 (4)5 (2)2 (3)10 (3)20 (3)85 (3)
 Asian18 (1)20 (2)5 (2)1 (1)2 (1)9 (1)38 (2)
 Native American7 (1)4 (<1)001 (<1)2 (<1)11 (<1)
 Unknown01 (<1)00001 (<1)
 Missing18 (1)18 (1)2 (1)08 (3)10 (1)36 (1)
Median age (range), y65 (13-93)62 (22-92)62 (13-90)59 (30-80)61 (28-89)75 (70-93)63 (13-93)
Prior therapy, No. (%)       
 Nephrectomy970 (77)1111 (89)179 (89)55 (79)267 (92)597 (81)2081 (83)
 Radiotherapy355 (28)530 (42)46 (23)57 (81)139 (48)214 (29)885 (35)
Systemic/targeted therapies       
 Interferon-α0678 (54)48 (24)20 (29)131 (45)188 (26)678 (27)
 Interleukin-20538 (43)34 (17)18 (26)100 (35)88 (12)538 (22)
 Bevacizumab0290 (23)24 (12)6 (9)290 (100)57 (8)290 (12)
 Thalidomide0147 (12)8 (4)8 (11)26 (9)38 (5)147 (6)
 Sunitinib029 (2)2 (1)01 (<1)5 (1)29 (1)
 Other0419 (34)57 (28)17 (24)68 (23)98 (13)419 (17)
 Median time since initial diagnosis (range), y0.6 (<0.1-31.2)2.2 (<0.1-34.2)1.0 (<0.1-31.2)1.1 (<0.1-16.6)2.1 (<0.1-19.3)1.7 (<0.1-34.2)1.4 (<0.1-34.2)
Sites of disease (≥20% of patients), No. (%)       
 Lung832 (66)901 (72)127 (63)59 (84)205 (71)511 (69)1733 (69)
 Kidney416 (33)339 (27)49 (24)31 (44)54 (19)229 (31)755 (30)
 Bone317 (25)380 (30)51 (25)25 (36)106 (37)160 (22)697 (28)
 Liver260 (21)336 (27)63 (31)15 (21)90 (31)146 (20)596 (24)
 Lymph node248 (20)251 (20)58 (29)19 (27)60 (21)138 (19)499 (20)
 Adrenal126 (10)144 (12)15 (7)16 (23)30 (10)65 (9)270 (11)
Disease sites, No. (%)       
 1448 (36)393 (31)63 (31)1 (1)100 (35)273 (37)841 (34)
 2391 (31)361 (29)69 (34)14 (20)78 (27)227 (31)752 (30)
 >2381 (30)469 (38)67 (33)55 (79)108 (37)214 (29)850 (34)
 Missing34 (3)27 (2)3 (1)04 (1)22 (3)61 (2)

The median duration of sorafenib treatment was 12 weeks (range, <1-81 weeks); the median average daily dose was 758 mg. Dose reductions occurred in 869 (35%) patients and interruptions in 1515 (61%) patients, primarily because of adverse events (88% and 79%, respectively). The most common adverse events leading to dose reduction were hand-foot skin reaction (n = 218, 8.7%), rash (n = 104, 4.2%), gastrointestinal events (n = 111, 4.4%), and hypertension (n = 90, 3.6%); similarly, those leading to dose interruption included hand-foot skin reaction (n = 326, 13.0%), rash (n = 203, 8.1%), gastrointestinal events (n = 207, 8.3%), and hypertension (n = 115, 4.6%). The incidence of treatment interruptions because of toxicity was higher in this expanded access protocol than in the prospective definitive phase 3 trial. It is unclear whether this is because of enrollment of patients with greater morbidities, less experience with the drug by investigators, or the availability of other agents for this disease. The proportion of patients who discontinued for adverse events was 10%; the most common causative adverse events were gastrointestinal and dermatologic.

Drug-related grade 2 and ≥3 adverse events reported in >2% of patients are presented in Table 2. The most common grade ≥2 drug-related adverse events were hand-foot skin reaction (18%), rash (14%), hypertension (12%), and fatigue (11%). Rates in elderly patients were similar. Drug-related central nervous system hemorrhages occurred in <1% of the overall population and were not reported in the brain metastases subpopulation. Seizures occurred in 7% of patients with brain metastases and in 1% of the total population; none was attributed to sorafenib therapy. There were 189 (8%) on-study deaths. Incidence was similar across subgroups (range, 6%-8%), except in patients with brain metastases (17%). The most common nonfatal serious adverse events reported in all patients were dyspnea (3%) and dehydration (3%).

Table 2. Drug-Related Grade 2 and ≥3 Adverse Eventsa Reported in ≥2% of Patients
Adverse EventsPatient Subgroups, No. (%)All Patients, N=2504, No. (%)
First-Line, n=1254Previously Treated, n=1250Nonclear Cell, n=202Brain Metastases, n=70Prior Bevacizumab, n=290Age ≥70 Years, n=736
  • a

    Graded according to version 3.0 of the National Cancer Institute Common Terminology Criteria for Adverse Events.

Adverse event grade2≥32≥32≥32≥32≥32≥32≥3
Dermatologic              
 Hand-foot skin reaction104 (8)132 (11)111 (9)106 (8)10 (5)19 (9)4 (6)5 (7)23 (8)32 (11)51 (7)58 (8)215 (9)238 (10)
 Rash/desquamation119 (9)80 (6)97 (8)44 (4)21 (10)15 (7)7 (10)2 (3)18 (6)8 (3)73 (10)38 (5)216 (9)124 (5)
 Alopecia20 (2)1 (<1)18 (1)1 (<1)2 (1)0001 (<1)013 (2)1 (<1)38 (2)2 (<1)
 Dry skin20 (2)017 (1)1 (<1)1 (<1)0005 (2)09 (1)037 (1)1 (<1)
Fatigue86 (7)51 (4)84 (7)62 (5)13 (6)12 (6)7 (10)5 (7)22 (8)18 (6)63 (9)49 (7)170 (7)113 (5)
Gastrointestinal              
 Diarrhea75 (6)32 (3)62 (5)26 (2)9 (4)3 (1)1 (1)2 (3)5 (2)5 (2)41 (6)18 (2)137 (6)58 (2)
 Nausea55 (4)17 (1)42 (3)21 (2)8 (4)5 (2)3 (4)1 (1)15 (5)4 (1)26 (4)9 (1)97 (4)38 (1)
 Anorexia65 (5)13 (1)49 (4)13 (1)7 (3)1 (<1)9 (13)1 (1)15 (5)7 (2)45 (6)10 (1)114 (5)26 (1)
Mucositis, oral cavity (symptomatic)31 (2)9 (1)27 (2)16 (1)6 (3)01 (1)2 (3)4 (1)8 (3)17 (2)7 (1)58 (2)25 (1)
 Vomiting22 (2)12 (1)17 (1)15 (1)4 (2)3 (1)01 (1)4 (1)5 (2)9 (1)7 (1)39 (2)27 (1)
 Constipation15 (1)2 (<1)20 (2)2 (<1)1 (<1)0006 (2)07 (1)1 (<1)35 (1)4 (<1)
 Taste alteration11 (1)07 (1)01 (<1)03 (4)01 (<1)06 (1)018 (1)0
 Dehydration2 (<1)16 (1)7 (1)7 (1)01 (<1)1 (1)06 (2)1 (<1)5 (1)13 (2)9 (<1)23 (1)
Hypertension83 (7)48 (4)114 (9)66 (5)11 (5)9 (4)6 (9)024 (8)14 (5)55 (7)36 (5)197 (8)114 (5)
Hypophosphatemia22 (2)8 (1)8 (1)2 (<1)1 (<1)1 (<1)00006 (1)1 (<1)30 (1)10 (<1)
Pain, abdominal6 (<1)7 (1)9 (1)15 (1)2 (1)7 (3)1 (1)007 (2)5 (1)8 (1)15 (1)22 (1)
Weight loss30 (2)2 (<1)23 (2)2 (<1)5 (2)1 (<1)3 (4)06 (2)1 (<1)19 (3)1 (<1)53 (2)4 (<1)

Overall best response per RECIST guidelines was available in 1891 (76%) patients (Table 3). The median PFS in the overall population was 24 weeks (95% confidence interval [CI], 22-25; censorship rate, 46%); it was similar among first-line and previously treated populations (Fig. 2, Top). For comparison with other clinical trials that typically exclude patients with nonclear cell RCC, PFS was also calculated without data from patients with nonclear cell RCC (n = 202). Results were similar to that in the overall population with a median of 24 weeks (95% CI, 22-25; censorship rate, 46%).

thumbnail image

Figure 2. Kaplan-Meier curves for progression-free survival (PFS) in the entire population (Top) and patients enrolled in the extension protocol (Bottom) are shown, including first-line patients and previously-treated patients. Inset shows PFS for the papillary and chromophobe subgroup. The median PFS was 21 weeks (censoring rate was 43% [81 of 188]). CI indicates confidence interval.

Download figure to PowerPoint

Table 3. Best Response Rates by Patient Subgroupa
Response by Radiologic AssessmentPatient Subgroup, No. (%)All Patients, N=1891, No. (%)
First-Line, n=951Previously Treated, n=940NCC, Papillary, n=107NCC, Chromophobe, n=20Brain Metastases, n=50Prior Bevacizumab, n=199Age ≥70 Years, n=537
  • NCC indicates nonclear cell; CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease.

  • a

    Includes patients valid for radiologic response assessment and subjects valid for clinical and radiologic response assessment.

CR1 (<1)0000001 (<1)
PR38 (4)29 (3)3 (3)1 (5)2 (4)5 (3)19 (4)67 (4)
SD for at least 8 weeks752 (79)759 (81)87 (81)17 (85)34 (68)154 (77)427 (80)1511 (80)
PD160 (17)152 (16)17 (16)2 (10)14 (28)40 (20)91 (17)312 (16)
CR+PR+SD791 (83)788 (84)90 (84)18 (90)36 (72)159 (80)446 (83)1579 (84)

Of the 328 patients enrolled in the extension protocol, 246 patients had data available for analysis of progression; the median PFS in these patients was 36 weeks (95% CI, 33-45; censorship rate, 56%). In the first-line patients from the extension population (n = 220), the median PFS was 36 weeks (95% CI, 33-45; censorship rate, 58%). In the previously treated nonclear cell extension patients (n = 26), the median PFS was 46 weeks (95% CI, 30-59; censorship rate, 38%) (Fig. 2, Bottom). The median OS in the entire population was 50 weeks (95% CI, 46-52; censorship rate, 63%); similar results were obtained for the first-line and previously treated patients (Fig. 3).

thumbnail image

Figure 3. Kaplan-Meier curves for overall survival for the entire population are shown, including first-line patients and previously (Prev) treated patients. Inset shows overall survival for the papillary and chromophobe subgroup. The median progression-free survival was 40 weeks (censoring rate was 57% [108 of 188]). CI indicates confidence interval.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES

The phase 3 TARGET trial established the safety and efficacy of sorafenib in patients with advanced RCC22; the ARCCS expanded access program broadened the experience with sorafenib in this patient population. As the ARCCS program was conducted in hospitals and community practice centers in addition to academic sites and because inclusion criteria were more liberal than for clinical trials, the ARCCS program enrolled a large number of patients who were more representative of the general metastatic RCC population than those in the typical RCC clinical trial.

Sorafenib treatment in the ARCCS program demonstrated a clinically manageable safety profile, with types and severity of adverse events similar to those in the phase 3 TARGET trial. The most common adverse events in both TARGET and ARCCS were hand-foot skin reaction, rash, fatigue, diarrhea, and nausea. The severity of these adverse events was usually low to moderate grade. Incidence and severity of adverse events in different subgroups of patients (first-line patients, previously treated patients, patients with nonclear cell RCC, brain metastases, previous treatment with bevacizumab, and patients ≥70 years of age) were also similar to those in the overall ARCCS population.

Hypertension as a serious adverse event was reported in TARGET as well as other clinical trials with sorafenib.22, 24 In ARCCS, grade ≥3 drug-related hypertension was reported in only 5% of patients. However, other studies indicate that the incidence of sorafenib-induced blood pressure elevation is on the order of 75%,25 and a meta-analysis suggests that the overall incidence of hypertension is about 23%.24

The ARCCS study also suggests that the tolerability profile of sorafenib is not worse in patients with previously treated active or progressive brain disease, although baseline data regarding the extent of brain disease and prior treatment were limited. Overall, the higher rate of seizures in those with brain metastases was not attributed to sorafenib, and there were no cerebral bleeding events reported in this subgroup. In the ARCCS program, 736 (29%) patients were ≥70 years of age, reflective of the age distribution of patients with advanced RCC. As with the overall ARCCS population, the adverse events were predictable and manageable in elderly patients.

The TARGET trial demonstrated that sorafenib therapy improved PFS 2-fold versus placebo in patients with RCC refractory to prior cytokine therapy.22 In a disease state for which the mortality rate is high and median survival is only 13 months,26 this delay in disease progression is notable. In trials with other VEGF inhibitors in patients with advanced RCC regardless of prior therapy status, median PFS ranged from 4 months to 11 months.2 In the ARCCS program, median PFS of 6 months in the entire population and 9 months in patients in the 6-month extension phase is consistent with previous findings. The OS of 12 months observed in the ARCCS program was lower than that in the TARGET study (19 months); however, the patient population was less selective in the former. Furthermore, efficacy results from the ARCCS program are not directly comparable to those from randomized, controlled trials because of inherent limitations in study design and methods of an expanded access program. First, data collection and monitoring were not as rigorous as for a controlled clinical trial. A key limitation was that investigator assessments were not validated by an independent, centralized reviewer. In addition, radiologic assessments were preferred but not always obtained; in such cases, clinical assessments of tumor response were used. Perhaps most importantly, FDA approval of sorafenib for treatment of advanced RCC terminated the expanded access program. This limited the duration of on-study therapy (median, 12 weeks) and precision in the estimate of PFS because of the high rate of censorship (46%-73% of this was because of switching to commercial drug). In a large proportion of cases, confirmatory radiologic assessments of a partial response could not be performed on study.

The extension phase was added as a protocol amendment to allow for longer treatment and follow-up of patients who had not received prior therapy and those with nonclear cell RCC, thus providing additional information about efficacy of sorafenib in these patient types. However, the censorship rate in the extension subset was also quite high (56%-91% of this was because of switching to commercial drug). Nonetheless, within the limits of the program design, these results suggest that sorafenib is effective regardless of prior therapy status. Recently, first-line therapies for advanced RCC with other multikinase inhibitors (sunitinib) and angiogenic inhibitors (combination of bevacizumab and interferon) were shown to be more effective than IFN-α alone. There are no direct comparisons between these treatments and sorafenib.27, 28

The majority (80%) of metastatic RCC is clear cell; therefore, most prior efficacy data, including those obtained with sorafenib, are from patients with this histologic diagnosis. In fact, in many trials, patients with nonclear cell RCC were excluded. Although the extension phase of this protocol was designed in part to assess PFS in patients with nonclear cell disease, the number of patients with nonclear cell enrolled in the extension population was too small (n = 202) to provide meaningful estimates. Nevertheless, the 127 patients with valid response data represent the largest experience with this drug in nonclear cell RCC to date. Within the limitations of local pathology evaluation, albeit with central pathology report review, sorafenib appears to have antitumor activity in papillary and probably chromophobe cancers.

Finally, emerging data suggest that there is not an absolute cross-resistance between different VEGF pathway targeting agents. In fact, a previous study demonstrated activity of sorafenib or sunitinib in patients with advanced RCC with prior therapy with antiangiogenic agents.29 In ARCCS, 290 patients had prior therapy with bevacizumab, and their response to sorafenib was similar to that of other subgroups. Thus, similar to sunitinib, sorafenib has clinical activity after bevacizumab treatment.30

In conclusion, the ARCCS expanded access program provided additional information on sorafenib for the treatment of advanced RCC in a more representative patient population, extending the experience with sorafenib beyond the clinical trial setting in patients who had not been previously examined. This confirms that sorafenib, along with other VEGF and mTOR pathway-directed agents, is an appropriate addition to the therapeutic armamentarium for this disease. This study does not address current controversies regarding relative value or appropriate sequence of the various agents now available. The data generated from the ARCCS program supported results from the TARGET trial and extended the findings to other subsets of patients who had not been previously examined. The tolerability of sorafenib was similar to that observed in previous clinical trials, with the majority of adverse events clinically manageable. Furthermore, these findings were supportive of previous trials demonstrating that sorafenib therapy delays disease progression in patients with advanced RCC. Additional studies are currently underway to investigate how best to combine or sequence sorafenib with other therapeutic agents in a variety of treatment settings.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES

We thank the patients who enrolled in this study and their families. We acknowledge the critical review provided by Sarah Guadagno, PhD and James Partyka, PharmD; Onyx Pharmaceuticals, Inc; and the medical writing assistance of Meenakshi Subramanian, PhD, Envision Pharma, Inc.

CONFLICT OF INTEREST DISCLOSURES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES

Sponsored by Bayer HealthCare Pharmaceuticals and Onyx Pharmaceuticals, Inc. Dr. Stadler has acted in a compensated consultant/advisory role for Bayer, Onyx, Pfizer, Genentech, Novartis, and Wyeth, and has received research support from Bayer and Onyx. Dr. Figlin has acted in a compensated consultant role for Bayer and Onyx. Dr. McDermott has acted in a compensated consultant/advisory role for Bayer, Onyx, Genentech, Wyeth, and Novartis and has received research funding from Novartis. Dr. Dutcher has acted in a consultant role for Bayer/Onyx, has received honoraria and research grant from Bayer/Onyx. Dr. Knox has received honoraria from Bayer, Pfizer, and Novartis, and research funding from Bayer and Pfizer. Dr. Henderson has acted in a consultant role for Amgen, Genentech, Bristol, ImClone, GlaxoSmithKline, Abraxis, and Pfizer. Dr. Drabkin has received honoraria as a member of the advisory board for Nexavar. Dr. Curti has acted in an uncompensated consultant/advisory role for Zymogenetics and has received honoraria from Novartis and Bayer/Onyx. Dr. Ryan has received honoraria and speaking fees from Bayer/Onyx. Dr. Xia is an employee of Bayer HealthCare. Dr. Cupit is an employee of Bayer HealthCare. Dr. Bukowski has acted in a compensated consultant/advisory role for Pfizer, Bayer, Novartis, Wyeth, and Genentech, and has received honoraria from Pfizer, Wyeth, Genentech, Bayer, and Novartis.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. CONFLICT OF INTEREST DISCLOSURES
  8. REFERENCES