Michael Tanaka

Prospective Evaluation of Cancer Clinical Trial Accrual Patterns: Identifying Potential Barriers to Enrollment

PURPOSE Well-conducted cancer clinical trials are essential for improving patient outcomes. Unfortunately, only 3% of new cancer patients participate in clinical trials. Barriers to patient accrual in cancer clinical trials must be identified and overcome to increase patient participation. MATERIALS AND METHODS We prospectively tracked factors that potentially affected patient accrual into cancer clinical trials at the University of California Davis Cancer Center. Oncologists seeing new outpatients were asked to complete questionnaires regarding patient characteristics and the physicians decision-making on patient eligibility, protocol availability, and patient opinions on participation. Statistical analysis was performed to correlate these parameters with subsequent protocol accrual. RESULTS There were 276 assessable patients. At the initial visits, physicians did not consider clinical trials in 38% (105/276) of patients principally because of a perception of protocol unavailability and poor performance status. Physicians considered 62% (171/276) of patients for participation in clinical trials. Of these, only 53% (91/171) had an appropriate protocol available for site and stage of disease. Seventy-six of 90 patients (84%) with available protocols met eligibility criteria for a particular study. Only 39 of 76 patients (51%) agreed to participate in cancer clinical trials, for an overall accrual rate of 14% (39/276). The remainder (37/76, 49%) declined trial participation despite meeting eligibility criteria. The most common reasons were a desire for other treatment (34%), distance from the cancer center (13%), patient refusal to disclose reason (11%), and insurance denial (8%). Patients with private insurance were less likely to enroll in clinical trials compared to those with government-funded insurance (OR, 0.34; P =.03; 95% CI, 0.13 to 0.9). CONCLUSION Barriers to cancer clinical trial accrual can be prospectively identified and addressed in the development and conduct of future studies, which may potentially lead to more robust clinical trials enrollment. Investigation of patient perceptions regarding the clinical trials process and the role of third party-payers is warranted.

Phase III Trial of Fluorouracil-Based Chemotherapy Regimens Plus Radiotherapy in Postoperative Adjuvant Rectal Cancer: GI INT 0144

PURPOSE Adjuvant chemoradiotherapy after or before resection of high-risk rectal cancer improves overall survival (OS) and pelvic control. We studied three postoperative fluorouracil (FU) radiochemotherapy regimens. PATIENTS AND METHODS After resection of T3-4, N0, M0 or T1-4, N1, 2M0 rectal adenocarcinoma, 1,917 patients were randomly assigned to arm 1, with bolus FU in two 5-day cycles every 28 days before and after radiotherapy (XRT) plus FU via protracted venous infusion (PVI) 225 mg/m2/d during XRT; arm 2 (PVI-only arm), with PVI 42 days before and 56 days after XRT + PVI; or arm 3 (bolus-only arm), with bolus FU + leucovorin (LV) in two 5-day cycles before and after XRT, plus bolus FU + LV (levamisole was administered each cycle before and after XRT). Patients were stratified by operation type, T and N stage, and time from surgery. RESULTS Median follow-up was 5.7 years. Lethal toxicity was less than 1%, with grade 3 to 4 hematologic toxicity in 49% to 55% of the bolus arms versus 4% in the PVI arm. No disease-free survival (DFS) or OS difference was detected (3-year DFS, 67% to 69% and 3-year OS, 81% to 83% in all arms). Locoregional failure (LRF) at first relapse was 8% in arm 1, 4.6% in arm 2, and 7% in arm 3. LRF in T1-2, N1-2, and T3, N0-2 primaries who received low anterior resection (those most suitable for primary resection) was 5% in arm 1, 3% in arm 2, and 5% in arm 3. CONCLUSION All arms provide similar relapse-free survival and OS, with different toxicity profiles and central catheter requirements. LRF with postoperative therapy is low, justifying initial resection for T1-2, N0-2 and T3, and N0-2 anterior resection candidates.

Premedication strategy for weekly paclitaxel.

Purpose: Dexamethasone and histamine antagonists have been employed commonly as premedications for prophylaxis of hypersensitivity reactions (HSRs) to paclitaxel. Frequent premedications for weekly administration of paclitaxel are associated with added side-effects and extra cost. We analyzed our experience of HSRs to paclitaxel administered every 3–4 weeks, and designed a treatment algorithm to eliminate premedications in a majority of patients receiving weekly paclitaxel. Patients and methods: The incidence of HSRs was analyzed retrospectively in patients who received 3-hr infusions of paclitaxel (135–225 mg/m2) every 3–4 weeks in our institution over a period of 5 years. On the basis of the results of this analysis, we designed a premedication schema for patients receiving weekly paclitaxel, 50–90 mg/m2/week, as follows: Thirty minutes prior to the first weekly dose of paclitaxel, patients received intravenously (IV) dexamethasone (10 mg), diphenhydramine (25 mg), and cimetidine (300 mg). If no HSRs occurred, all premedications were deleted for subsequent weekly paclitaxel doses. For patients who experienced HSRs, 20 mg of dexamethasone was given orally 12 and 6 hr prior to re-challenge with paclitaxel in addition to diphenhydramine and cimetidine. Results: Over a period of 5 years, 358 patients received 1608 3-hr infusions of paclitaxel (135–225 mg/m2). Hypersensitivity reactions, which occurred exclusively during the first cycle of paclitaxel administration, were observed in 14 patients. Of these 14 patients, 11 were successfully retreated with paclitaxel without HSRs, two had recurrent HSRs upon paclitaxel re-challenge, and one refused further treatment. These observations indicate that HSRs to paclitaxel occurred in 4% of patients upon first exposure, that most of these patients can be retreated successfully with paclitaxel without recurrent HSRs, and that if no HSRs occur during the first cycle, HSRs are unlikely to occur with subsequent paclitaxel administration. The premedication schema was applied to 30 patients receiving 205 one-hr infusions of weekly paclitaxel. Using this premedication strategy, no HSRs have been observed during the initial or subsequent administration of paclitaxel. Conclusion: We conclude that this premedication strategy is feasible and worthy of further study for patients receiving weekly paclitaxel.

A Phase II Study of Paclitaxel, Carboplatin, and Radiation with or without Surgery for Esophageal Cancer

Background: Cisplatin-based chemoradiotherapy (CRT) has been a standard treatment for patients with locally advanced esophageal cancer. However, cisplatin is associated with significant toxicity. We conducted a phase II clinical trial of concurrent paclitaxel, carboplatin, and radiation with or without surgery as an alternative to the standard cisplatin-based CRT for localized and metastatic esophageal cancer. Methods: Fifty patients with esophageal cancer were enrolled: 16 patients with stage II, eight patients with stage III, and 26 patients with stage IV disease. Two thirds (67%) of patients had adenocarcinoma and one third (33%) with squamous histology. Patients with resectable disease were treated with paclitaxel 30 mg/m2, twice weekly for 10 doses, carboplatin AUC (area under the curve) 1.5 weekly for five doses; and concurrent radiation, 1.8 Gy/day, for a total of 45 Gy, followed by esophagectomy. Without surgery, patients received an additional dose each of paclitaxel and carboplatin with concurrent radiation for a total of 50.4 Gy, followed by two consolidation cycles of paclitaxel (200 mg/m2) and carboplatin (AUC 6). Results: During CRT, common stage III/IV toxicities included nausea/emesis (19%), esophagitis (9%), and neutropenia (4%). For consolidation chemotherapy, neutropenia (23%), neuropathy (8%) and nausea/emesis (4%) were the most common stage III/IV side effects. After CRT, 26% had a complete response, 17% had a partial response, and 41% had stable disease. Ninety-one percent of patients had clinical improvement of dysphagia. With a median follow-up of 32 months, the median survival was 12 months for patients with metastatic disease, 44 months for localized disease treated with esophagectomy, and >44 months for localized disease treated with definitive CRT. Conclusions: The regimen of paclitaxel, carboplatin, and radiation with or without surgery is well tolerated with promising efficacy for patients with esophageal cancer.

Weekly 24-hour continuous infusion interleukin-2 for metastatic melanoma and renal cell carcinoma : a phase I study

Twenty-nine patients with biopsy-confirmed metastatic melanoma (17) or metastatic renal cell carcinoma (12) were treated with escalating doses or recombinant human interleukin-2 (IL-2) administered as weekly 24-h intravenous infusions. Patients received from 3 to 12 x 10(6) C.U./m2 (18-72 x 10(6) I.U./m2) weekly over a treatment period of 1 to 16 weeks, with a median of eight weekly cycles administered. Patients in all treatment groups experienced non-life-threatening systemic side effects consisting of fever, nausea, vomiting, fluid retention, and diarrhea. Grade III hypotension was seen in four of six patients (67%) at 12 x 10(6) C.U./m2, and represented the dose-limiting toxicity. Grade IV hypotension occurred in 1 of 14 patients at 6 x 10(6) C.U./m2; no other grade IV toxicities were observed. Grade III fever occurred in 3 of 11 patients (27%) treated at 3 x 10(6) C.U./m2, 3 of 14 patients (21%) at 6 x 10(6) C.U./m2, and 3 of 6 patients (50%) at 9 x 10(6) C.U./m2. An objective response was observed in 3 of 28 evaluable patients (10%): 1 complete response and 1 partial response in renal cell cancer, and 1 partial response in a melanoma patient. We conclude that for future studies, the recommended dose of IL-2 given as a weekly 24-h infusion is 9 x 10(6) C.U./m2 and that a low rate of objective tumor response can be obtained in patients with melanoma and renal cell carcinoma using this regimen.

Phase I Study of Two Different Schedules of Bortezomib and Pemetrexed in Advanced Solid Tumors with Emphasis on Non-small Cell Lung Cancer

Introduction: Bortezomib and pemetrexed are approved anticancer agents with non-overlapping mechanisms of action and toxicity. We examined the safety and tolerability of pemetrexed in combination with two different schedules of bortezomib in patients with advanced solid tumors. Methods: Two separate dose-escalating arms (arm A and arm B) were conducted simultaneously. Patients received pemetrexed on day 1 (D1) (500–600 mg/m2 IV) every 21 days. In arm A, bortezomib was given twice weekly (0.7–1.3 mg/m2 on D1, 4, 8, and 11). In arm B, bortezomib was given weekly (1.0–1.6 mg/m2 on D1 and 8). Results: We treated 27 patients on four dose levels in arm A and three dose levels in arm B. Tumor types included lung (n = 16), adenoid cystic carcinoma (n = 2), prostate (n = 2), sarcoma (n = 2), breast (n = 1), thymus (n = 1), head and neck (n = 1), and gastrointestinal(n = 2). Dose-limiting toxicities were seen in arm A only; grade 3 asthenia (n = 2), grade 3 transaminitis and dehydration (n = 1). The most common grade 3/4 toxicity was neutropenia. Of 26 evaluable patients, 2 patients had partial response (1 in arm A and 1 in arm B), 13 had stable disease (7 in arm A and 6 in arm B), and 11 had progression (6 in arm A and 5 in arm B). Of the 16 patients with non-small cell lung cancer, 2 (12.5%) had partial response and 9 had stable disease, for a disease control rate of 68.8%. Recommended phase II dose for arm A is pemetrexed 500 mg/m2 and bortezomib 1.3 mg/m2 twice weekly. For arm B, the recommended dose is pemetrexed 500 mg/m2, bortezomib 1.6 mg/m2 weekly. Conclusions: Pemetrexed with bortezomib is feasible and tolerable at recommended single-agent doses. Based on the observed efficacy, a phase II study in non-small cell lung cancer is warranted.

An evaluation of barriers to accrual in the era of legislation requiring insurance coverage of cancer clinical trial costs in California

PURPOSEClinical trials are essential to improve cancer therapy, but only 3% of newly diagnosed adult cancer patients enroll annually. We previously conducted a prospective analysis of factors affecting trial accrual at the UC Davis Cancer Center between 1997 and 2000. It was found that the accrual rate was 14% and that patients with private insurance were significantly less likely than patients with government insurance to enroll, suggesting that fear of insurance denial was a barrier. In 2002, a new California law (SB37) required insurers to reimburse routine costs of care for cancer trials. METHODSTo assess the impact of SB37 on accrual, we repeated our study using the same survey instrument. Oncologists seeing new patients at the UC Davis Cancer Center from August to November 2002 completed questionnaires that inquired about patient characteristics and eligibility, protocol availability, and patient willingness to participate. RESULTSPhysicians considered clinical trials for 55% (118/216) of patients, but trials were available for only 53% (62/118). Eligibility criteria were met by 82% (51/62). Of these, 69% (35/51) agreed to participate (vs 51% previously). No patient declined to participate because of insurance limitations (vs 8% previously). Furthermore, insurance type was no longer a significant factor in determining whether patients would enroll. This suggests that although the overall rate of accrual is only slightly increased after passage of SB37, patients may be more willing to enroll. Efforts to increase participation must include enhancing physician and patient awareness of SB37.

Phase II trial of R115777 (NSC #70818) in patients with advanced colorectal cancer: A Southwest Oncology Group study

SummaryPurpose: The purpose of this Phase II multi-institutional trial was to determine the efficacy and toxicity of R115777 in previously untreated patients with metastatic colorectal carcinoma. Patients and methods: Patients were required to have histologically confirmed colorectal cancer with distant metastatic disease that was not surgically curable. They could not have received prior chemotherapy for metastatic disease. R115777 was given at a dose of 300 mg p.o. twice a day for 21days every 28 days until tumor progression or toxicity or other reason for discontinuation occurred. The primary endpoint was to determine the confirmed response probability with this treatment. Results: There were 55 eligible patients accrued to the study. There were no complete responses, but one confirmed partial response for a confirmed response probability of 2% (95%CI 0–10%). Three additional patients had an unconfirmed partial response for an overall response probability of 7%. The time to treatment failure was 1.7 months and the estimated median survival was 8.1 months. One patient died of treatment related infection and there were 7 other patients with grade 4 toxicities consisting of neutropenia, leukopenia, febrile neutropenia and thrombocytopenia, depression, increased bilirubin, anemia, and pneumonitis/infiltrates. Conclusion: R115777 given as a single agent by this dose and schedule is ineffective in patients with metastatic colorectal cancer.

Phase I trial of gemcitabine and paclitaxel in advanced solid tumors.

Purpose. Gemcitabine and paclitaxel are chemotherapeutic agents with clinical antitumor activity in a broad range of malignant solid tumors. Because of preclinical synergy, unique mechanisms of action and resistance, and nonoverlapping toxicities, gemcitabine and paclitaxel combinations are attractive for testing in clinical trials. Prior weekly gemcitabine and paclitaxel regimens administered on a 28-day cycle have been limited by cumulative hematological toxicity on day 15, thus reducing the planned gemcitabine dose intensity. We therefore conducted a phase I trial of a 21-day schedule of weekly gemcitabine and paclitaxel to determine the tolerability, maximum tolerated dose (MTD), and preliminary estimates of efficacy of this regimen. Patients and Methods. Forty-one patients with advanced malignant solid tumors were accrued. Gemcitabine was given at a fixed dose of 1000 mg/m2 while paclitaxel was administered at an initial dose of 60 mg/m2, then escalated by 15 mg/m2 increments over seven dose levels to a prospectively planned maximum dose of 150 mg/m2. Both agents were infused intravenously on days one and eight every 21 days. At least three patients were enrolled per dose level. No intrapatient dose escalation was allowed. Results. All patients were assessable for toxicity and 31 were assessable for response. The regimen was generally well-tolerated. Dose-limiting thrombocytopenia was observed in one patient at a paclitaxel dose of 135 mg/m2/week (dose level 6). After expansion of this dose level by 14 additional patients, no further dose-limiting toxicities were observed although one patient at dose level seven died of neutropenic sepsis after completing three cycles. There were eight partial responders for an overall response proportion of 26% (95% CI: 11, 41). Twelve patients (39%) had stable disease. Conclusion. This 21-day schedule of gemcitabine and paclitaxel is safe, well-tolerated, and active. The recommended phase II dose is gemcitabine 1000 mg/m2 and paclitaxel 150 mg/m2 on days one and eight every 21 days. The antitumor activity observed with this regimen warrants further investigation.

Phase II study of dovitinib in patients progressing on anti-vascular endothelial growth factor therapy

BACKGROUND Prior work identified the fibroblast growth factor (FGF) pathway as a mediator of resistance to anti-vascular endothelial growth factor (VEGF) therapy. We tested dovitinib, an inhibitor of both FGF and VEGF receptors, in patients progressing on anti-VEGF treatment. METHODS Patients with measurable advanced colorectal or non-small cell lung cancer with progression despite anti-VEGF treatment within 56 days, good performance status and adequate organ function were eligible. A research tumor biopsy was followed by treatment with dovitinib 500 mg on a 5-day on/2-day off schedule for 28-day cycles. The primary endpoint of tumor response was evaluated every 2 cycles. Secondary endpoints included toxicity and 8-week disease control rate. Intratumor mRNA expression of angiogenic mediators was analyzed using a next generation sequencing based expression array. RESULTS Ten patients treated previously with bevacizumab or ziv-aflibercept enrolled. The study closed with termination of dovitinib development. No responses were observed in 7 evaluable patients. The best response was stable disease in 1 patient. Common toxicities included gastrointestinal, metabolic, and biochemical derangements. All patients experienced at least one grade ≥ 3 treatment-related adverse event, most commonly fatigue, elevated GGT, and lymphopenia. Expression of multiple angiogenic mediators was common in tumors progressing on anti-VEGF therapy including high levels of FGFR1 and VEGFA. CONCLUSIONS We found no evidence for the activity of dovitinib in patients who had recently progressed on anti-VEGF therapy and toxicities were significant. In tumors progressing despite anti-VEGF therapy, a multitude of pro-angiogenic mediators are expressed, including members of the FGF pathway.