Prevalent cases in observational studies of cancer survival: do they bias hazard ratio estimates?

Observational epidemiological studies often include prevalent cases recruited at various times past diagnosis. This left truncation can be dealt with in non-parametric (Kaplan-Meier) and semi-parametric (Cox) time-to-event analyses, theoretically generating an unbiased hazard ratio (HR) when the proportional hazards (PH) assumption holds. However, concern remains that inclusion of prevalent cases in survival analysis results inevitably in HR bias. We used data on three well-established breast cancer prognosticators - clinical stage, histopathological grade and oestrogen receptor (ER) status - from the SEARCH study, a population-based study including 4470 invasive breast cancer cases (incident and prevalent), to evaluate empirically the effectiveness of allowing for left truncation in limiting HR bias. We found that HRs of prognostic factors changed over time and used extended Cox models incorporating time-dependent covariates. When comparing Cox models restricted to subjects ascertained within six months of diagnosis (incident cases) to models based on the full data set allowing for left truncation, we found no difference in parameter estimates (P=0.90, 0.32 and 0.95, for stage, grade and ER status respectively). Our results show that use of prevalent cases in an observational epidemiological study of breast cancer does not bias the HR in a left truncation Cox survival analysis, provided the PH assumption holds true.

Demographics, response, and overall survival of patients with advanced renal cell cancer to sunitinib in a cohort of minority patient population

e16164

Background: Sunitinib is a tyrosine kinase inhibitor active in renal cell cancer (RCC). There is scanty literature of its efficacy in minority population. Methods: 21 patients (pts) with RCC who received sunitinib between February 2006-September 2007 were identified and studied as a retrospective cohort. Clinical and survival data were analyzed using fisher's test, chi square test, Kaplan Meier analyses. Results: Of the 21 patients, 11 (52%) were female and 10 (47%) male, 7 (33.3%) African American, 7 (33.3%) Hispanic, and 5 (23.8%) Caucasian. Median age at diagnosis was 59 years (32–74). 7 (33.3%) had clear cell and 3 (14.3%) sarcomatoid pathology. Mixed, poorly differentiated, papillary and unknown histology were 2 (9.5%) each. 12 (57%) pts had stage 4 disease at diagnosis, stage 3 in 3 (14.3%), stage 2 in 1 (4.8%) and 5 (23%) had missing data. 14(66.7%) pts underwent nephrectomy while 7 (33.3%) did not. 6 (28.6%) pts has good MSKCC risk score, 11 (52.4%) intermediate risk and 3 (14.3%) poor risk. Sunitinib was given at a dose of 50 mg daily for 4 wks followed by 2 wks off. Median duration of treatment was 2.5 months (0–9 mts) and median follow up was 13 mts (1–21 mts). Common grade 3–4 toxicities observed were hand foot syndrome (n = 2), hypertension (n = 2) and thrombocytopenia (n = 1). 4 pts discontinued therapy due to adverse events. 5 (23.8%) has stable disease and 13 (61.9%) had progressive disease. Response to sunitinib was not influenced by sex, race, performance status, MSKCC Score, serum calcium level, LDH and hemoglobin level. Median survival of the group was 4 mts with no difference based on gender (p = 0.8), ethnicity (p = 0.8) or histologic type (p = 0.7). Survival of pts with ECOG performance status (PS) 1 was 8 mts, PS 2 was 4 mts, PS 3 was 2 mts (p = 0.001), MSKCC good risk was 9.4 mts, intermediate score was 9.4 mts and poor risk was 2 mts (p = 0.18). Hemoglobin (p = 0.6), LDH (p = 0.6), calcium (p = 0.2) did not affect the survival. Conclusions: In this minority cohort of pts with RCC treated with sunitinib, response and median survival is much lower than the historical controls. Tolerability and side effect profile are similar to reported literature. Prospective studies are warranted in the treatment of RCC with sunitinib in ethnic minority population.

No significant financial relationships to disclose.

A phase I study of flavopiridol using an alternative schedule in patients (pts) with advanced solid tumors

2580

Background: A phase I study of flavopiridol, a cyclin-dependent kinase inhibitor, using an alternative schedule was conducted in pts with solid tumors given its promising activity in pts with chronic lymphocytic leukemia (CLL). Methods: Using standard 3x3 ph I dose escalation design, NCI-sponsored trial was performed to determine the safety and dose-limiting toxicity (DLT) of flavopiridol given as a 30-min IV loading dose followed by a 4-hr infusion weekly for 4 wks repeated every 6 wks. DLT was defined as Gr 4 hematologic toxicity (HT) for > 7 days, > Gr 3 non-HT except Gr 3 fatigue or diarrhea resolving <4 days and cytokine release syndrome (CRS) > Gr 3 despite steroids. Blood samples were obtained at pre-dose and 0.5, 1, 3, 4.5, 6, 8, 24, and 48-hr after start of first bolus dose for pharmacokinetics (PK). Results: 26 pts with advanced solid tumors with a median age of 63 (44–75) yrs were enrolled. Median no. of doses was 7.5 (1–24). Table 1 outlines the PK parameters, DLTs and CRS. Due to a grade 5 CRS/death in cohort 3, the protocol was amended to include 20 mg IV dexamethasone prior to flavopiridol to prevent CRS (cohorts 2B, 1B). Of the 20 evaluable pts, 35% had stable- and 65% had progressive-disease. Results of serum cytokines (IL-2, IL-4, IL-6, TNF-a, IFN-g, IL-10) levels will be presented. Conclusions: There was a higher frequency of CRS, despite prophylactic steroids seen our pts with solid tumors compared to previous studies with CLL and this correlated with AUC. PK and toxicity profile in our pt population differs from pts with hematologic malignancies administered flavopiridol on the same schedule. Protein binding and serum albumin levels are under evaluation as potential contributors. This work is supported by NCI U01-CA76576.

[Table: see text]

[Table: see text]

Phase II trial of preoperative bevacizumab (Bev), irinotecan (I), cisplatin (C), and radiation (RT) in esophageal adenocarcinoma: Preliminary safety analysis

4573

Background: Preo chemoRT with weekly I/C and 5040 cGy followed by surgery is well tolerated [JCO 24: Abstract 4032; 2006]. ECOG trial E1201recently reported a median survival of 34 months with this preop regimen [JCO 26: Abstract 4532; 2008]. Bev + chemo improves response rate (RR) and time to progression (TTP) when added to weekly I/C in advanced esophagogastric cancer but does not increase chemo toxicity [JCO 24: 5201; 2006]. We are now combining in a Phase II trial Bev/I/C with concurrent radiotherapy (RT) in esophageal adenocarcinoma (EA) with the primary endpoint of safety. Methods: Patients (pts) with resectable Siewert's I or II EA were staged by EUS, PET, and CT. Induction chemo consisted of I-50–65 mg/m2 and C-30 mg/m2 weeks 1,2,4,5, Bev-7.5 mg/kg weeks 1 and 4; and, during RT (180 cGy daily to 5040 cGy), I/C was given weeks 7,8,10,11 and Bev weeks 7,10. Esophagectomy was 6–8 weeks after RT. A planned toxicity analysis was made in 10–15 pts completing chemoRT, and in 10 pts undergoing surgery: toxicity was acceptable if grade 3 / 4 hematologic toxicity remained < 72% and non hematologic toxicity < 40% during combined chemoRT (based on our prior phase II trial of I/C/RT [JCO 24: Abstract 4032; 2006]); and if pts undergoing surgery had no surgical complication related to Bev. Results: 18 pts have been enrolled, 12 male: 6 female; 7 Siewert I: 11 Siewert II; T3N1 12: T3N0 5: T2N0 1. 14 are evaluable for toxicity, 2 are too early, one progressed prior to RT, and one was taken off due to a CVA from a patent foramen ovale. Grade 3/4 neutropenia occurred in 4 pts (29%). Grade 3/4 non heme toxicity occurred in 5 pts (36%), including esophagitis 2 pts (14%), neutropenic fever 1 pt (7%), and pulmonary embolism 1 pt (7%). No grade 3 / 4 hypertension was seen, and 3 pts (21%) developed grade 1 proteinuria. Ten pts underwent surgery, and there were no unexpected surgical or wound complications; there were 2 anastomotic leaks. Pathologic responses: 1 pathologic CR and 1 T0N1. Conclusions: In a preliminary analysis of pts treated with Bev + preop chemoRT in EA, there was no increase in hematologic/non hematologic toxicity or Bev related surgical complications. Accrual will continue to 33 patients. Supported by Genentech.

[Table: see text]

Results of autologous transplants for lymphomas from a tertiary cancer center in India

7106

Background: Autologous stem cell transplanation is the standard of care for patients of relapsed and refractory non-Hodgkin's lymphoma (NHL) and Hodgkin's lymphoma (HL). We report the results of transplants in lymphomas from our center and role of possible prognostic factors. Methods: All 50 consecutive patients who underwent transplant for HL (70%) and NHL (30%) from August 1994- August 2008 were included in this retrospective study. Fifty eight percent of patients received BEAM (carmustine, etoposide, ara-c and melphalan), 30% LACE (lomustine, ara-c, cyclophosphamide and etoposide), 8% ICE (ifosfamide,carboplatin and etoposide) and 4% high dose melphalan (200mg/m2) conditioning regimens. Seventy eight percent of patients received peripheral blood stem cells (PBSC), 8% bone marrow (BM) and 14% both PBSC and BM. Prognostic factors evaluated for progression free survival (PFS) were serum albumin level and body mass index (BMI) at the time of transplant, stage at diagnosis and source of stem cells, while for over all survival (OS), status of disease at transplant was also included. Results: The median time to transplant was 2.25 years from the time of diagnosis. The median age at transplant was 25 years. Seventy four percent of patients were male. At the time of transplant, thirty two percent were in complete remission (CR), 50% in partial remission (PR) and 18% had refractory disease (RD). The median serum albumin and BMI at the time of transplant were 4 g/dl and 22.5 kg/m2 respectively. The best disease response rate was 86% (CR+PR) in patients evaluable for response. Thirteen patients relapsed at a median interval of 11 months post transplant. The cumulative probability of OS and progression free survival PFS at 5 years were 40% and 34% respectively for the whole group. Multivariate analysis using cox regression identified serum albumin greater than 4 g/dl and those receiving PBSC grafts as independently associated with improved OS and PFS. Conclusions: These data provide the first published report of outcomes of autologous transplants in lymphomas from India. Our data suggests that serum albumin level at the time of transplant and stem cell source are important prognostic factors for PFS and OS.

No significant financial relationships to disclose.

Response, time to progression, and overall survival of patients with stage III lung cancer in a minority-based cohort

7558

Background: Lung cancer is the leading cause of cancer death. Response and survival of patients (pts) with stage III lung cancer in minority population is not well studied. Methods: 79 pts treated between 2001 and 2006 were studied as a retrospective cohort. Clinical and survival data were analyzed using fisher's test, chi square test, Kaplan Meier analyses. Results: 33 Pts had Stage IIIA; Median age at diagnosis was 58 yrs (37–75). 14 were males (42.4%) 19 females (57.6%) 26 African American (78.8%) and 7 Caucasians (21.2%). Median number of co morbidities 2 (0–4). Mean follow up was 25 months (mo) (2 - 93), 19 had surgery (12 Lobectomy (36.4%), 7 Pneumonectomy (21.2%)) and 14 were unresectable (42.4%). All pts received adjuvant chemotherapy. Unresectable pts received chemotherapy and radiation, 8 carboplatin and gemcitabine and 6 cisplatin and etoposide. Median time to progression in resectable IIIA was 23 mo (2- 67), unresectable IIIA was 12 mo (3–93). Median survival in pts with resection was 26 mo (2 - 67), unresectable was 12 mo (5 - 93). Overall survival and time to progression was not influenced by sex, race, tumor type. 46 pts had stage IIIB, median age at diagnosis 57.5 yrs (40–68). 31 were males (67.4%) 15 females (32.6%) 29 African Americans (63%) 11 Caucasians (23.9) 3 Asians (6.5%) and 3 Hispanics (6.5%). Median follow up was 10 mo (3–97). All pts received chemotherapy and radiation. 23 received cisplatin and etoposide (50%), 14 carboplatin and gemcitabine (30%), 5 carboplatin and etoposide (9%), 5 carboplatin and paclitaxel (9%). 13 had complete response (28.3%), 15 partial response (32.6%), 2 stable disease (4.3%) and 10 Progression (21.7%). Median time to progression was 9 mo (3–97). Median overall survival was 10 mo (3–97). Overall survival and time to progression was not influenced by sex, race, tumor type, chemotherapy regimen. Conclusions: In this minority based cohort response, time to progression, overall survival in both IIIA and IIIB pts is much lower then historical controls. The overall survival and time to progression in both IIIA and IIIB is not influenced by race, sex, tumor type and type of chemotherapy regimen. Further investigations of disease and healthcare disparities in the underserved minority population, are warranted.

No significant financial relationships to disclose.

A phase I clinical trial of FOLFIRI in combination with the pancyclin–dependentkinase (CDK) inhibitor flavopiridol

e14511

Background: Preclinical data indicate that sequential treatment with flavopiridol (F) increases irinotecan- and 5-FU-induced apoptosis. Clinically, F showed promising activity when combined with irinotecan (Shah et al CCR 2005). Methods: We conducted a phase I trial of FOLFIRI + F every 2 weeks in patients (pts) with advanced solid tumors. Based on sequence- dependent inhibition, F was given 3 h after irinotecan but before 5-FU. Two maximum tolerated doses (MTD) were determined: MTD1 (F over 1 h) and MTD2 (F 30-min bolus + 4 h infusion). F pharmacokinetics (PK) were determined. Results: Of 74 pts treated, 63 were evaluable for toxicity and 56 for response. Pt characteristics: median age 60 (range 19–83), KPS 90 (70–90), prior regimens 3 (1–10). 39 pts received prior irinotecan. Tumor types: colorectal, gastric, HCC, GE junction, small bowel, pancreas, bile duct, breast, bladder, ovarian, sarcoma, melanoma, anal, urethral, thymic, head & neck, unknown primary. MTD1: F 80mg/ m2 with irinotecan 180mg/m2, LV 400mg/m2, 5FU: 400mg/m2 bolus + 2400mg/m2 over 48 h. Dose-limiting toxicities (DLT) were diarrhea, fatigue, neutropenia, neuropathy. MTD2: F 35mg/m2 bolus + 35mg/m2 over 4 h with the same FOLFIRI dose. DLTs were diarrhea, neutropenia, and fatigue. Clinical activity included 2 partial responses (small bowel cancer, 10.3 m; bladder cancer, 10 m) and 1 complete response (mucosal melanoma 10.3 m). 22 pts had stable disease (median 5.9 m; range 1.5–25.7 m). Clinical benefit rate (CR + PR + SD for > 3 m) was 39% (22/56). Of 25 pts with colorectal cancer, 11 had as best response SD for > 3m (median 6 m, range 4.2–15.4 m), despite failing ≥ 1 irinotecan-containing regimen. 6 of those had significant decreases (36–78%) in CEA. F PK showed interpatient variability with no significant interaction between FOLFIRI dose and F Cmax. F Cmax increased with increasing F dose. At MTDs, there was higher Cmax in pts who experienced DLT (3.48μM) vs those who did not (2.21μM). Conclusions: F can be safely given as a bolus (80mg/m2) or split dose (35mg/m2 bolus + 35mg/ m2 over 4 h) in combination with irinotecan 180mg/m2, LV 400mg/m2, 5FU: 400mg/m2 bolus + 2400mg/m2 over 48 h. Promising clinical activity is seen in mucosal melanoma and irinotecan-refractory colon cancer. (Supported by NCI R01CA67819)

[Table: see text]

Preparation and physicochemical characterization of surfactant based solid dispersions of ezetimibe

Solid dispersions of the poorly water-soluble drug ezetimibe were prepared with a surfactant, Pluronic 188 in different ratios and characterized by FTIR, XRD, DSC and dissolution studies. The melting method was employed to prepare the solid dispersions whereas a physical mixture (1:3) was prepared by co-grinding the individual components in a mortar. Physical studies demonstrated a significant reduction in crystallinity with a possibility of presence of amorphous character of drug in the solid dispersions of ezetimibe. Among all binary systems studied, the 1:3 proportion of ezetimibe: Pluronic 188 showed fastest dissolution rate (DE90: 73.38% +/- 3.95) suggesting optimum ratio of the surfactant used.

A case of diffuse pilar leiomyoma or acquired smooth muscle hamartoma?

A 62-year-old woman presented with a skin-coloured indurated asymptomatic plaque, 150 mm in diameter on the knee. Histological analysis of a skin biopsy taken from the lesion showed haphazardly arranged bundles of smooth muscle in the deep dermis, characteristic of a pilar leiomyoma. Cutaneous pilar leiomyomas are rare, benign smooth muscle tumours arising from the arrector pili muscle, which usually appear as red or brown papules < 15 mm in diameter. This case is unusual in its clinical appearance and size, being at least 10 times larger than the largest previously documented cutaneous pilar leiomyoma. Similar lesions have also been previously reported under the term 'acquired smooth muscle hamartoma' (ASMH). These cases are also rare and have no known consistent clinical features. Histologically it is difficult to differentiate between ASMH and leiomyomas, and some authors would consider this case as a new report of ASMH. We feel, however, that as the lesion is composed of only one element and has grown out of proportion with the growth of the limb, it is more in keeping with a true tumour rather than a hamartoma. We suggest the term 'diffuse pilar leiomyoma' may be more appropriate in this case, representing a new clinical variant of cutaneous pilar leiomyoma.

The Physiological Effects of Short-term Smoking Cessation in Claudicants

Introduction

Smoking contributes to atherosclerosis and causes significant postoperative morbidity. New antismoking law forces short-term pre-operative abstinence. Demonstrable clinical benefit might motivate complete cessation. Our aim was to determine the effects of 24-hr smoking cessation on cardiorespiratory function and claudication distance.

Methods

Smoking claudicants were randomized to 24hr smoking or abstinence. Following these separate periods, cardiopulmonary exercise testing was performed. Pre- and post-exercise, serum lactate and ankle brachial pressure index (ABPI) were measured. During exercise, cardiorespiratory function, initial and absolute claudication (IC, AC) distances and visual analogue scores (VAS) of pain were recorded.

Results

16 patients completed both tests. IC, AC and VAS were unchanged with abstinence ( P = .43, .66, .96, .83). ABPI drop post-exercise was unchanged with abstinence ( P = .08, .09). Cardiorespiratory function was not affected by smoking cessation.

Conclusion

Cardiorespiratory function and claudication symptoms are unchanged following 24-hr smoking cessation., No deterioration in respiratory function is important when considering anaesthetic administration. However, lack of symptomatic improvement may discourage patients from abstaining. Further investigation should determine correlation between short-term abstinence and postoperative morbidity.

High levels of fusidic acid-resistant Staphylococcus aureus despite restrictions on antibiotic use

BACKGROUND

High rates of fusidic acid (FA)-resistant Staphylococcus aureus (FRSA) in patients with skin disease have been previously attributed to high usage of topical FA.

AIMS

To assess whether local community guidelines to restrict topical FA has affected its prescription and use and the level of FRSA in patients with skin disease.

METHODS

Stapylococcus aureus isolates from microbiology samples received over a 4-month period in 2004 were tested for antibiotic sensitivities. Comparison was then made with the results of a previous study carried out in 2001.

RESULTS

A significant fall was seen in the use of topical FA in dermatology patients. In 2001, 62% of patients had used FA-containing preparations within the previous 6 months, compared with just 15% of patients in 2004 (P < 0.001). The number of topical FA prescriptions in primary and secondary care dropped between 2001 and 2004. The proportion of S. aureus isolates resistant to FA in dermatology patients had not significantly fallen between 2001 (50%) and 2004 (41%) (P = 0.4). However, there was a significant increase in FA resistance within hospital inpatients, nondermatology outpatients and primary-care patients (P < 0.05). The FRSA level had doubled in hospital inpatients (20%) and almost tripled in nondermatology outpatients (28%) and primary care patients (25%).

CONCLUSION

Persistent high levels of FA resistance may represent the development of an FRSA reservoir in the community. Continued restriction of FA is still recommended.

Development and characterization of bicalutamide-poloxamer F68 solid dispersion systems

The objective of the present work was to improve the dissolution rate of a poorly water-soluble drug, bicalutamide, by a solid dispersion technique. The solid dispersion systems of bicalutamide were prepared with poloxamer F68 in 1:1, 1:3, and 1:5 ratios using the melting method. The interaction of drug with polymer was evaluated by TLC, FTIR, and powder XRD. The results of powder XRD showed a significant decrease in the crystallinity of drug in the binary systems of bicalutamide. All binary systems of bicalutamide showed faster dissolution than pure drug alone (p < 0.001). However, among all binary systems studied, 1:1 proportion of bicalutamide : poloxamer was found to be excellent for dissolution enhancement (DP30: 99.98% +/- 3.9) of bicalutamide. The higher ratios of poloxamer F68 (1:3 and 1:5) had retarded the release of drug from their corresponding binary systems which might be due to its gelling property in higher concentration.

Voriconazole plasma monitoring

AIMS

Very little information is available regarding the use of voriconazole drug monitoring in children with invasive fungal infections. The purpose of this study was to report the cases of five paediatric patients treated with voriconazole, in which plasma levels were used to monitor therapy.

METHODS

Five children treated with voriconazole were included in this case series. Voriconazole plasma levels were determined using either a bioassay or liquid chromatography-tandem mass spectrometry.

RESULTS

The patients' ages ranged from 2 to 10 years old (mean 6.2 years). Three patients had acute leukaemia and two had suffered severe burn injuries. Doses administered varied from 3.4 mg/kg every 12 h to 8.1 mg/kg every 8 h. Plasma voriconazole concentrations were unpredictable for these paediatric patients. Subtherapeutic levels were frequently observed, despite progressive increments in dosage. For others, voriconazole levels markedly increased after a small increment in dosage. Phenobarbitone caused important drug interactions with voriconazole for one [corrected] of the patients.

CONCLUSIONS

The dose administered did not correlate with exposure as measured by plasma levels of voriconazole. While the optimal dosage for voriconazole in children is still unknown, drug monitoring seems warranted to ensure adequate exposure, and after dose increments to prevent excessive exposure. Drug interactions significantly altered exposure.