Duration Of PFS Research Articles

Durable Responses in Patients Treated with Galiximab (Anti-CD80) in Combination with Rituximab for Relapsed or Refractory Follicular Lymphoma: Long-Term Follow-up of a Phase II Clinical Trial.

Galiximab is an anti-CD80 monoclonal antibody with human IgG1 constant regions and macaque variable regions. CD80 is an immune costimulatory molecule that is constitutively expressed on the surface of follicular lymphomas. Modest single-agent clinical activity (ORR 11%) was demonstrated in a single agent Phase I study of galiximab in relapsed/refractory, follicular NHL with the observation of late and prolonged responses. We previously conducted a phase II multicenter study evaluating the combination of galiximab and rituximab for relapsed/refractory follicular NHL, and demonstrated an ORR of 64% (17% CR, 14% CRu, and 33% PR). Here, we analyze long-term safety and efficacy data from this trial. Patients received galiximab (500 mg/m2 qwk x 4 weeks) concurrently with a standard course of rituximab (375 mg/m2 qwk x 4 weeks). No maintenance therapy was allowed. Sixty-four patients received treatment. Mean age at study entry was 59 yrs. The majority of patients (88%) were Stage III/IV, and FLIPI risk groups were distributed as low (27%), intermediate (39%), and high (34%). All patients had received at least 1 prior lymphoma therapy; 42% were rituximab naïve; rituximab-refractory patients (no response or a response with TTP<6 months) were excluded. The median follow-up of responding patients is 45 months (range 9–59 months). The median PFS was 12.2 months. 20% of patients had PFS durations exceeding 2 years. 37% of patients did not require any additional lymphoma therapy for at least 2 years following treatment (TTNT); 28% had a TTNT of more than 3 years. Grade, tumor bulk, age, stage, prior rituximab exposure, and FLIPI score did not predict for PFS duration of > 2 years, or TTNT duration of > 2 years (p=NS, chi-square analysis). Obtaining a CR predicted for PFS and TTNT duration >2 years (p=0.001). Response duration to prior therapy was not correlated to outcome following rituximab/galiximab therapy (p=0.18). Quantitative immunoglobulin levels (IgA, IgM and IgG) did not differ between prolonged responders and other patients, and did not significantly change over the first year of follow-up. Median Cmax, half-life, AUC and Cl were similar between prolonged responders and other patients. No late opportunistic infections, secondary malignancies, or infusion-associated deaths have been reported. We conclude that the combination of rituximab and galiximab is well-tolerated in long-term follow-up, with a substantial number of durable responses. Almost one-third of patients treated with only 4 weeks of this combination do not require additional lymphoma therapy for more than three years. There is no clear prognostic marker that predicts for prolonged benefit from the combination. These durable responses provide strong rationale for ongoing phase III clinical trials of the galiximab/rituximab combination.

A phase II trial of sunitinib (S) in previously-treated pancreas adenocarcinoma (PAC), CALGB 80603

4515 Background: S is an oral broad-spectrum tyrosine kinase (TK) inhibitor with potential utility in PAC given the perturbation of multiple signaling pathways and stromal alterations noted in PAC. CALGB conducted an NCI-sponsored phase II study of S in patients (pts) with previously-treated PAC. Methods: Eligibility included: PAC previously treated with front-line gemcitabine (gem) therapy, or progression of disease (PD) during or within 3-months of completion of adjuvant therapy, or newly metastatic disease following therapy for locally advanced PAC; performance status 0–2; measurable disease by RECIST; no prior anti-VEGF therapy; no evidence of duodenal invasion on CT. S was dosed at 50 mg daily for 28 days followed by 14 days of rest (1 cycle). Re-staging was performed after each of the first 4 cycles of therapy. Study design: Single-arm, multi-center cooperative group, 3-stage design, with response criteria of at least stable disease (SD) in 1/19, 2/39, for expansion. Planned accrual was 64. Primary endpoint: Response rate (RECIST). Secondary endpoints: Duration of response, toxicity, PFS, and survival. Results: From 11/06 to 11/07, 77 pts were accrued with the criteria met for full accrual (1 withdrew consent). Pt characteristics were: male 54%, female 46%, age: median 65 yrs, range 42- 87, and PS 0/1/2 45%/48%/7%. Prior therapy included: gem 61%, gem-based combination 37%, TK inhibitor 14%, radiation therapy 14%. Sites of disease were: liver 84%, lung 30%, other 19%. As of 12/07, 52 pts are evaluable for response. Response criteria for continuation were met at the two interim analyses. SD occurred in 7 pts [13%, 95% CI (3.6%, 22%)] and PD in 29 (56%). Number of cycles received was: 1- 86%; 2- 12%, 3- 2%. Dose modification was needed in 21%. Median PFS was 41 days (CI 37–57d), and median OS 97 days (CI 79–127d). Grade 3–5 toxicities included: hematologic 13% (G3), fatigue 12%, bleeding 6% (G3), nausea 4% (G3), Thrombosis/embolism 2% (G3), TTP/renal failure 2%, GI perforation 2% (G5). Conclusions: Preliminary analysis of a large cooperative group phase II study of S in previously-treated PAC suggests modest single agent activity and no new safety signals in this patient population. Final results will be presented. Author Disclosure Employment or Leadership Consultant or Advisory Role Stock Ownership Honoraria Research Expert Testimony Other Remuneration Pfizer

A multicenter phase II study of erlotinib and bevacizumab in patients with metastatic melanoma

8539 Background: Erlotinib and bevacizumab have demonstrated activity in a number of malignancies by virtue of interrupting interdependent signaling pathways thought important in tumorigenesis. Melanoma may be an appropriate target based on its expression of EGFR and VEGF. We conducted a phase II multi-institutional trial evaluating erlotinib and bevacizumab in advanced melanoma patients. Methods: Eligibility included measurable disease, ECOG PS = 0–1, adequate organ function, no more than one prior therapy for metastatic disease, and CNS metastases were allowed if limited and controlled. Patients received oral erlotinib 150 mg/day and bevacizumab 10 mg/kg IV Q 2 weeks with tumor evaluation every 8 weeks. The primary outcomes were response rate (RR), response duration, and frequency of PFS &gt;6 months. Secondary outcomes included overall survival, safety, and tolerability. A two-stage accrual design was employed ensuring that = 3/21 patients had PFS &gt;6 months before additional patients were accrued. Results: As of Nov 2006, 29 patients with metastatic melanoma were enrolled. A total of 23 patients were evaluable for response. The majority was male 19/29 (65%) and had a median age = 62 yrs (range 35–78 years). Fifteen of the 29 had stage M1c disease (51.7%) and 18/29 (62.1%) had a PS = 1. Ten patients (34%) had prior adjuvant therapy and 6 patients (21%) prior therapy for metastatic disease. There were 2/23 (9%) partial responses lasting &lt; 6 months and 5/23 (22%) had stable disease lasting &gt; 6 months. The median progression free survival of evaluable patients was 96 days (95% CI: 50 - 142 days). A total of 25 grade III toxicities were observed with the most common being rash/pruritis (n=4), pain (n=4), fatigue (n=3), hypertension (n=2) and diarrhea (n=2). Two grade IV toxicities were observed (myocardial infarction and bowel perforation) both thought to be due to bevacizumab. Conclusion: The combination of erlotinib (150 mg/day) and bevacizumab (10 mg/kg) appears potentially active in patients with metastatic melanoma with largely tolerable toxicities. Accrual of a total of 41 patients will be completed shortly. [Table: see text]

Predictors of toxicity to concurrent chemoradiation (CRT) in patients (pts) with locally advanced head and neck squamous cell cancer (HNSCC)

16540 Background: A number of studies have evaluated the impact of prognostic factors on survival but very few have studied their role as determinants of toxicity, especially outside of the context of clinical trials. We set out to identify patient characteristics that are likely to predict toxicity to chemotherapy (CT) with daily cisplatin 6mg/m2 and radiation therapy (RT) in pts with locally advanced HNSCC. Methods: Retrospective chart review of above pts at the Juravinski Cancer Center from 2000 through 2004. A number of pt characteristics were analysed. Toxicity outcomes evaluated were completion of CT and of RT, rise in serum Cr , hematologic suppression, need for extra hydration, and need for extra antiemetics. Results: The charts of 108 pts were reviewed. Median follow-up was 36 months. Five year projected OS and PFS were 78% and 57% in our study compared to 53% and 47% in a published EORTC trial (NEJM, 2004). Median duration of PFS was 64 months in our study compared to 55 months in the EORTC trial. Median OS was not reached in our study. Nineteen percent completed CT and 81% completed RT, 17% had full course of CRT. Thirty-nine percent required extra hydration and 50% required extra antiemetics. Twenty four percent had elevated creatinine levels and 20.4% had hematologic suppression. Logistic regression was used to search for significant correlations between pt characteristics and outcomes. Pts with occasional alcohol intake were more likely to complete RT than heavy alcohol abusers (OR 8.9; P 0.04). Pts with occasional alcohol intake were more likely to get extra hydration than heavy alcohol abusers (OR 5.1; P 0.008).Hematologic suppression was more likely with increasing age (OR 1.08;P 0.015).No significant difference in toxicity outcomes was observed between males and females. Conclusions: This review demonstrated that CRT with daily cisplatin is associated with significant toxicity and most pts are unable to receive full treatment . A number of pt variables might predict increased susceptibility to toxicity. This should be validated in a prospective study. Median PFS in our series appear to be at least as good as previously published reports of CRT using daily cisplatin as well as dose schedules but toxicity does not appear greatly reduced. No significant financial relationships to disclose.

Sorafenib (S) and dacarbazine (D) combination in patients (pts) with advanced malignant solid tumors: Phase I study with tumor biopsy genomic analysis and dynamic contrast enhanced ultrasonography (DCE-US)

3556 Background: S is a multikinase inhibitor targeting VEGFR-1, -2, -3, PDGFR-a and -β and Raf kinase. The combination of S (400 mg bid) and D (1,000 mg/m2, 21-day [d] cycle [C]) was previously defined as maximum tolerated dose. Objectives were to define the pharmacokinetic (PK) profile of D ± S, assess safety of the combination and collect antitumor activity data by genomic analysis and by DCE-US. Methods: All pts received D (1,000 mg/m2 by 1-hr infusion) with continuous oral S 400 mg bid (except Day1C1). PK was performed during 1st and 2nd infusions. RNA expression profiling was performed on sequential tumor biopsies at baseline (BL) and Day 21. DCE-US was assessed at BL and Day 8, 21 of C1. Results: 23 pts treated (median age 59 y) had melanoma (4), leiomyosarcoma (4), adenocarcinoma (5), mesothelioma (3), neuro-endocrine tumor (3) and other (4). 22 had = 1 prior chemotherapy (range, 1–7). Median (range) number of D cycles was 4 (1–9), median (range) duration of S treatment was 15 wks (3–48). Toxicities were manageable and included diarrhea (48% any grade [G], no G3/4), hand-foot skin reaction (48% any G, 4% G3), fatigue (26% G3), thrombopenia (22% G4), and neutropenia (13% G4). Of 22 evaluable pts, 1 (melanoma) had confirmed PR, 13 pts had SD as best response, with a median (range) duration of 161 d (75–258). 7 pts continue with SD. PK (n=15) showed increased mean AUC by 41% and mean Cmax of AIC (an inactive D metabolite that is formed in an equal proportion to the active moiety, diazomethane, that cannot be analytically measured) by 45%. Genomic analysis in 10 pts with PFS =3 mos and in 7 with PFS &lt;3 mos identified a subset of 4 genes whose modulation differentiates these groups with 100% accuracy. 99 DCE-US performed in 17 pts showed 3/7 with short PFS had increased volume on Day 21 and in 3 an increase in maximal intensity &gt;50% on Day 8. Of 10 pts with long PFS, maximal intensity decreased by &gt;50% on Day 21 in 8 pts. Conclusion: Combination of D and S is manageable with clear evidence of antitumor activity. Preliminary data strongly suggest a new paradigm based on sequential tumor genomic analysis allowing early identification of responders. Early stage DCE-US may predict PFS duration. No significant financial relationships to disclose.

Phase II study of RAD001 (everolimus) and depot octreotide (sandostatin LAR) in advanced low grade neuroendocrine carcinoma (LGNET)

4503 Background: Effective systemic therapy for advanced LGNET is lacking. Mutations involving the mTOR pathway including TSC2, NF-1, and vHL are associated with development of LGNET. mTOR also mediate signaling downstream of IGF1 and VEGF. RAD001 is a novel oral mTOR inhibitor. Octreotide has been described to inhibit VEGF and IGF1 production in solid tumors. Methods: Patients (pts) received depot octreotide 30 mg q28 days, and RAD001 5 (pts 1–30) or 10 mg (pts 31–60) po daily. Response was evaluated every 12 weeks (wks). Results: 30 carcinoid (C) and 30 islet cell (I) patients (pts) were enrolled between 2/05–7/06. 38 pts (23 C, 15 I) had prior octreotide. By RECIST criteria, there were 10 PR (17%), 45 SD (75%) (15 pts had &gt; 15% reduction), 5 PD (8%). The response rates within C, and I groups were 13% and 20%. The response rates within the 5 mg and 10 mg groups were 13% and 20% respectively. Of 39 pts with PD prior to study entry, there were 7 (18%) PR, 27 (69%) SD (9 pts had &gt; 15% reduction), 5 (13%) PD. PFS rate at 24 wks was 86%. The median PFS duration is 59 wks. The median PFS duration among the C and I groups were 69 and 39 wks. Among the 39 pts with progression at study entry, the median PFS duration was 38 wks. Of 24 pts with elevated chromogranin A at entry, 19 (56%) had &gt; 50% reduction. Treatment was well tolerated. The most common toxicity was mild aphthous ulceration. CTC G3/4 toxicities included: thrombocytopenia (3), neutropenia (3), hypophosphatemia (5), hyperglycemia (4), hypoglycemia (2), hypokalemia (2), fatigue (6), diarrhea (6), aphthous ulcer (5), rash (3), and 1 each of anemia, hypertriglyceridemia, bilirubin, AST/ALT, pain, nausea, nail, dysgeusia, pneumonitis. Conclusions: RAD001 at 5 or 10 mg daily was well tolerated in combination with octreotide. Promising anti-tumor activity has been observed. The RADIANT (RAD001 In Advanced Neuroendocrine Tumors) trials are underway to evaluate RAD001 in larger patient groups. No significant financial relationships to disclose.

Serum beta2-Microglobulin before Transplantation Is an Independent Predictor of Outcome in Patients with Multiple Myeloma Treated Using Autologous HCT; a Single Centre Multivariate Analysis in a Series of 81 Patients.

High-dose therapy combined with autologous stem cell transplantation-AHCT has been found effective both for up-front therapy and for management of chemotherapy sensitive relapses in multiple myeloma (MM). The results of five randomized trials are controversial; most have confirmed the advantage of AHCT in terms of CR and DFS but a significantly better OS was reported in only one study. It is unclear whether the entirety of MM patients benefit from the higher efficacy of AHCT versus standard chemotherapy or the better results reflect rather a high efficacy in certain subset of patients. It is of crucial importance therefore to define predictive factors characterizing patients who may benefit from therapy including AHCT.The main objective was to select pre-transplant factors predictive for the efficacy of autoHCT in patient with MM treated using defined standard operating procedures in single center from January 1993 to April 2005.Material and methods: 81 MM patients, undergoing first autoHSCT in single transplant centre: median age at autoHSCT 51 y (31–70), F/M 38/43, median time from diagnosis to autoHSCT 12 months (6–68), MM subtype: IgG n=51, IgA −13, non-secretory-7, not-defined −10; source of stem cells: PB −79, BM -2, conditioning: melphalan 200 mg /m2 (n=76), other (n=5), disease status: CR −30, PR-51.Factors submitted for multivariate analysis: serum beta2-microglobulin (b2-mgl), lactate dehydrogenase (LDH), monoclonal protein level (MoP), bone marrow plasma cell percentage (BM-pc), hemoglobin level (Hb), age ≤60 vs. > 60y and the number of transplanted CD34 positive cells ≤ median value equaling 7,18 x 106/kg vs. > this number. Factors were measured within the median time of 7 d before initiation of conditioning regimen.Results: The transplant related mortality at day 100 was 3,7%; the relapse incidence - RI at 9,2 y equaled 74% for patients with elevated b2m and 32% for those with b2m within normal range, p=0.02. Other factors included in multivariate analysis were: BM - pc ≥ 5% vs. < 5% (100% vs. 55%, p=0.07) and MoP elevated vs. within normal limits (77% vs. 59%, p=0.1).The progression free survival - PFS was decreased in patients with elevated b2-mgl (29% vs. 68%, p=0.02), in these with BM - pc ≥5% (0% vs. 45%, p=0.03) and with elevated MoP (22% vs. 40%, p=0.1). Coincident observations in our MM patients suggest that elevated b2-mgl is more frequent in patients displaying 13q−.In multivariate analysis b2-mgl remained the only factor associated with increased risk of relapse (hazard ratio=3.29, p=0.03) and reduced probability of PFS (hazard ratio=3.29, p=0.03)Conclusion: Serum beta2-microglobulin before first autoHCT is a reliable independent predictor for RI and PFS duration in MM patients, ipso facto it is helpful in appraisal of patients who may benefit from myeloablative therapy followed by autoHCT.

Bevacizumab plus irinotecan/5-FU/leucovorin for treatment of metastatic colorectal cancer results in survival benefit in all pre-specified patient subgroups

3617 Background. Bevacizumab [Avastin (BV)], a recombinant, humanized monoclonal antibody against VEGF, has demonstrated efficacy and safety when added to irinotecan/5-fluorouracil/leucovorin (IFL) chemotherapy as first-line therapy for metastatic colorectal cancer (CRC). The results of the primary analysis of a randomized phase III trial demonstrated that the addition of BV to IFL resulted in significant improvement in survival compared with IFL alone (p=0.00004; median survival increased from 15.61 to 20.34 months). Additional sub-group analyses were performed to evaluate the effects of baseline risk factors on survival. Methods. The effects of demographic and baseline prognostic characteristics and selected baseline laboratory values on duration of survival, PFS, and objective response rate were examined. Patient characteristics included baseline ECOG performance status, number of organ sites with metastases, site of primary tumor, age, sex, race, prior adjuvant chemotherapy, prior radiotherapy, duration of metastatic disease, years since diagnosis of CRC, baseline sum of longest diameters of target lesions, baseline levels of albumin, alkaline phosphatase, and LDH. Descriptive summaries of duration of survival and PFS were produced for each level of the categorical variables listed above for each treatment arm. These descriptive summaries consisted of the hazard ratio from unstratified Cox regression and Kaplan-Meier estimates of median time to the event. Results. 815 patients were randomized to receive either IFL + placebo or IFL + BV. A survival benefit in the IFL/BV arm was seen across all patient subgroups compared with the IFL/placebo arm (see table). Conclusion. The addition of Bevacizumab to IFL chemotherapy results in prolonged survival in multiple sub-groups. Author Disclosure Employment or Leadership Consultant or Advisory Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Genentech Genentech Genentech Genentech Genentech

A phase II trial of imatinib in patients with advanced carcinoid tumor

4124 Background: Effective systemic therapy options for carcinoid tumors are lacking. Carcinoid tumors co-express PDGF and PDGFR, a receptor/ligand system important in cell growth and survival. A phase II trial was conducted to assess the response rate and safety profile of PDGFR inhibitor, imatinib (Gleevec ™), at a dose of 400 mg PO BID in patients with advanced carcinoid tumors. Methods: Previously treated or untreated patients with advanced carcinoid tumors with adequate organ and bone marrow function were included. Response was evaluated every 12 weeks by CT or MRI. Results: 15 men and 12 women with a median age of 60 (22–74) enrolled. 21 patients received concurrent octreotide. 10, 6, 3 and 2 patients received prior chemotherapy, hepatic artery embolization, interferon, and radiation, respectively. Median number of weeks on study was 16 (range, 12–78+). By RECIST criteria, 1 patient, receiving concurrent octreotide, had a radiologic PR (remains on study after 1.5 years), 17 had SD, and 9 with PD. Among 14 patients with PD at entry, 8 patients remained progression free for at least 12 weeks (range, 18–52+). 4 patients had > 50% reduction in markers (chromogranin, 5-HIAA, pancreatic polypeptide). Median PFS duration is 24 weeks (range, 12–68+). PFS duration was significantly better in patients receiving concurrent octreotide (14 weeks vs 38 weeks; P=0.05). 1-year OS rate is 88%. There was no correlation between plasma VEGF and bFGF, CT flow studies and patient outcome. Grade 3–4 toxicities included fatigue (7), hypophosphatemia (5), diarrhea (3), fluid retention (3), nausea (2), granulocytopenia (2), hypokalemia (2), rash (1), anorexia (1), thrombocytopenia (1), hyperglycemia (1), and hyperbilirubinemia (1). Conclusions: Our data suggest a modest level of biologic effect of imatinib in carcinoid, therefore, further development of imatinib in conjunction with other agents may be warranted. Supported by a grant from Novartis. No significant financial relationships to disclose.

Bevacizumab plus irinotecan/5-FU/leucovorin for treatment of metastatic colorectal cancer results in survival benefit in all pre-specified patient subgroups

3617 Background. Bevacizumab [Avastin (BV)], a recombinant, humanized monoclonal antibody against VEGF, has demonstrated efficacy and safety when added to irinotecan/5-fluorouracil/leucovorin (IFL) chemotherapy as first-line therapy for metastatic colorectal cancer (CRC). The results of the primary analysis of a randomized phase III trial demonstrated that the addition of BV to IFL resulted in significant improvement in survival compared with IFL alone (p=0.00004; median survival increased from 15.61 to 20.34 months). Additional sub-group analyses were performed to evaluate the effects of baseline risk factors on survival. Methods. The effects of demographic and baseline prognostic characteristics and selected baseline laboratory values on duration of survival, PFS, and objective response rate were examined. Patient characteristics included baseline ECOG performance status, number of organ sites with metastases, site of primary tumor, age, sex, race, prior adjuvant chemotherapy, prior radiotherapy, durati...

Analysis of factors associated with outcome in patients with malignant peritoneal mesothelioma undergoing surgical debulking and intraperitoneal chemotherapy.

Malignant mesothelioma (MM) arising in the peritoneal cavity is a rare neoplasm characterized by peritoneal progression and for which there are limited therapeutic options. We evaluated the peritoneal progression-free and overall survival (PFS and OS, respectively) for patients with peritoneal MM after surgical resection and regional chemotherapy. Forty-nine patients (28 males, 21 females; median age, 47 years; range, 16 to 76 years) with MM underwent laparotomy, tumor resection, continuous hyperthermic peritoneal perfusion with cisplatin (median dose 250 mg/m2), and a single postoperative intraperitoneal dwell of fluorouracil and paclitaxel (n = 35) on protocols approved by the Institutional Review Board. Standard techniques for actuarial analyses of potential prognostic variables (Kaplan-Meier method with two-tailed log-rank test and Cox proportional hazards model) were performed. At a median potential follow-up of 28.3 months, median actuarial PFS is 17 months and actuarial OS is 92 months. Factors associated with improved PFS and OS by the Cox proportional hazards model were a history of previous debulking surgery, absence of deep tissue invasion, minimal residual disease after surgical resection (OS only), and age younger than 60 years (OS only). Surgical resection and regional chemotherapy for MM results in durable PFS and OS. Favorable outcome is associated with age, tumor biology (selection of patients with a history of previous debulking), lack of invasive tumor growth, and minimal residual disease after tumor resection.