Abstract OT2-2-01: SOFT and TEXT: Trials of tamoxifen and exemestane with and without ovarian function suppression for premenopausal women with hormone receptor-positive early breast cancer

Background: The SOFT and TEXT randomized phase 3 trials address two primary questions for endocrine treatment of premenopausal women with hormone receptor-positive breast cancer. 1) In combination with ovarian function suppression (OFS), does an aromatase inhibitor (exemestane, E) improve outcome compared with tamoxifen (T)? 2) Does addition of OFS to T improve outcome compared with T alone?

Trial Designs: SOFT compares 5y of T to OFS+T or OFS+E. OFS can be GnRH analog (triptorelin) × 5y, oophorectomy or ovarian irradiation. Median age was 43y (11% <35y); 35% had N+ disease. Prior neo/adjuvant chemotherapy (CT) given in 53%.

TEXT compares 5y of OFS+T to OFS+E. Patients were enrolled prior to CT (if planned). Median age was 43y (9% <35y); 48% had N+ disease. Adjuvant CT was planned in 60%.

Major Eligibility Criteria – Premenopausal, confirmed by estradiol levels– ER≥10% and/or PgR≥10%– Invasive early breast cancer

Specific Aims: Evaluate the role of aromatase inhibitors and the addition of OFS to T in this population.

Statistical Methods (amended 2011): The primary analysis will be intention-to-treat of all randomized patients. The primary endpoint, invasive disease-free survival (DFS), is defined as time from randomization to invasive local, regional, or distant relapse, contralateral breast cancer, appearance of a second non-breast malignancy, or death. DFS will be compared using a 2-sided stratified logrank test with an overall experiment-wise alpha level equal to at most 0.05.

In the original protocol, anticipated 5y DFS was 67% with T alone, 74% with OFS+T, and 79.8% with OFS+E, and required a total of 396 DFS events in TEXT and 783 DFS events in SOFT to reach 80% power. Because the enrolled population had more favorable characteristics and the event rates were lower than anticipated (2%/yr vs 8%/yr), the revised analysis plan is ‘time-driven’ rather than ‘event-driven.’

The comparison of OFS+E to OFS+T across both trials (n = 4717) is planned at median follow-up (MFU) >5y. The estimated power to detect a 20%, 25%, or 30% reduction in the hazard with OFS+E vs OFS+T is 63%, 84%, and 95%, respectively.

The comparison of OFS+T to T alone is planned at 5y MFU (SOFT, n=2045). The estimated power to detect a 20%, 25%, 30%, or 33.5% reduction in the hazard is 34%, 52%, 69%, and 80%, respectively.

Accruals

SOFT Target: 3000; Final: 3066

TEXT Target: 2639; Final: 2672

Enrollment 2003–2011; primary analyses expected late 2013/early 2014.

Related Research

Quality of Life (QL) component evaluates QL, menopausal symptoms and sexual impairment.

TEXT Translational Research investigates patient and tumor features that may contribute to treatment effectiveness and side effects.

TEXT-Bone investigates changes in bone mineral density and the role of serial serum markers of bone remodeling as predictors of bone side effects.

Co-SOFT evaluates changes in cognitive function during the first year.

SOFT-EST evaluates estrogen levels during the first 4y of GnRH analogue and whether there is a suboptimally estrogen-suppressed subgroup.

North American Pharmacogenetics study investigates whether genetic variations that affect T and E metabolism influence efficacy.

Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr OT2-2-01.

Measuring clinical competence in preregistration trainees by OSCEs

Focal points

Bloodspot acylcarnitine and amino acid analysis in cord blood samples: efficacy and reference data from a large cohort study

BACKGROUND

In order to test the feasibility of cord blood screening for inherited metabolic disease, a two-year cohort study of births in six obstetric units from five towns in the north of England was undertaken. These towns have a high prevalence of consanguineous marriages, largely among the immigrant Asian community. The purpose of the study was to determine whether early detection of metabolic disease was possible and whether early intervention would improve prognosis.

METHODS

Following parental consent, cord blood samples were collected at birth and analysed for acylcarnitine and amino acid profiles by tandem mass spectrometry in one of two laboratories. One laboratory used butylated derivatives, the other used underivatized samples. The same laboratories performed routine blood spot neonatal screening at 5-7 days of age on these babies. Patients with positive results were investigated and treated by a metabolic paediatrician as soon as possible.

RESULTS

24,983 births were examined. 12,952 samples were analysed as butyl derivatives, 12,031 samples were analysed underivatized. The following disorders were detected: medium-chain acyl-CoA dehydrogenase (MCAD) deficiency (1 case), 3-methylcrotonyl-CoA carboxylase (MCC) deficiency (2 cases), maternal carnitine transporter defect (2 cases), maternal MCC (1 case). The following disorders were diagnosed subsequently but were not detected by the cord blood screening: phenylketonuria (PKU) (1 case), maple syrup urine disease (MSUD) (2 cases), argininosuccinic aciduria (1 case), methylmalonic acidaemia (MMA) (1 case), glutaric aciduria type 2 (1 case), MCAD deficiency (2 cases), 3-hydroxy-3-methylglutaryl-CoA lyase deficiency (1 case). Comprehensive reference data for all analytes by both methods were obtained.

CONCLUSIONS

Cord blood testing is of limited value in detecting inherited metabolic disease. The metabolites associated with most disorders examined were not elevated in cord blood. Some maternal disorders, carnitine transporter defect and 3-methlycrotonyl-CoA carboxylase deficiency, are detected. These remain of uncertain clinical significance. Comprehensive reference data have been obtained that will facilitate future interpretation of studies in cord blood.

Phase I study of preoperative chemoradiation therapy in unresectable locally advanced breast cancer (LABC)

11097

Introduction: Therapy of unresectable LABC is challenging due to the need to resect local disease and prevent distant metastases. We conducted a phase I trial to determine the maximum tolerated dose of preoperative vinorelbine (VIN) combined with paclitaxel (P) and radiation (XRT). Methods: Eligible patients (pts) had unresectable LABC, ECOG performance 0–2 and adequate organ function. Prior chemotherapy and metastatic disease were permitted. All pts received weekly bolus P 80 mg/m2 and concurrent XRT at 2.0 Gy daily for 5 days in a week-on/week-off (WO/WO) schedule for 12 weeks. A planned 59 -70 Gy were delivered to the breast, chest wall and supraclavicular regions. In dose level 2, VIN 15 mg/m2 /week was added. Results: Twenty-six pts were enrolled from July 1999-July 2005 with a median age of 51 years (range 24–69); 9 pts had stage IV disease, 17 pts had stage IIIB disease, and 13 pts had received prior chemotherapy. VIN was not tolerated due to grade (G) 4 neutropenia despite GCSF; however single-agent P and concurrent XRT was well-tolerated. Thus, 7 pts in the final cohort received P/XRT without GCSF. This regimen was well-tolerated; toxicities seen included: G3 in-field skin desquamation 4/7; G4 lymphopenia 7/7, G3 infection 1/7; G3 wound dehiscence 1/7. After therapy, 20/26 pts underwent mastectomy: 6 pts had a pathologic complete response (no invasive disease), 12 had a pathologic partial response (scattered microscopic disease), and 2 pts had no response. Overall pathologic response rate was 90% for pts undergoing mastectomy; only 3 of the 20 resectable pts are alive without relapse a mean of 19.5 months after mastectomy. Conclusions: Neoadjuvant P 80 mg/m2 with full dose XRT in a WO/WO schedule is tolerable and effective allowing 20/26 unresectable pts to undergo successful mastectomy. Pts do not require routine GCSF support. Although most pts eventually relapse distantly, this regimen does render some pts not responding to initial doxorubicin-based therapy resectable.

[Table: see text]

No significant financial relationships to disclose.

A phase II trial of trastuzumab and low dose interleukin-2 in patients with metastatic breast cancer who have previously failed trastuzumab

3028

Background: Trastuzumab (TZB) mediates lysis of Her2-pos. breast cancer cells by interleukin-2 (IL2) primed natural killer (NK) cells. We hypothesized that IL2 would augment the effects of TZB. The objective of this study was to determine the response rate to and toxicity of low dose IL2 plus TZB in patients with Her2-pos. metastatic breast cancer who had progressed within 12 months of receiving TZB. Also, we measured the ability of patient (pt) peripheral blood mononuclear cells (PBMC) to produce cytokines and conduct antibody-directed cellular cytotoxicity (ADCC) against Her2-pos. target cells. Methods: Pts received low (1 million IU/m2) and intermediate dose (12 million IU/m2) IL2 plus TZB (4 mg/kg) in each cycle. Low dose IL2 was given on days 2–7 and days 12–21 of cycle 1, and days 4–14 of later cycles. Intermediate dose IL2 was given on days 9–11 of cycle 1, and on days 1–3 of later cycles. TZB was given on day 1 and 8 of cycle 1, and on day 1 of later cycles. Pt plasma and PBMCs were analyzed for levels of serum cytokines and ADCC against a TZB-coated cell line, respectively. Results: Thirteen pts with median age of 52 (range 30–71), and a median of 1 (range 1–2) prior TZB-containing regimens were enrolled. The median number of cycles completed was 4. Five pts had grade 3 or greater toxicities, including fever, nausea, vomiting, diarrhea, dyspnea, and hypercalcemia. Twelve pts had progression of disease, and 1 pt withdrew consent. ADCC of pt PBMCs against a TZB-coated cell line was enhanced in only 1 pt. Two pts had elevated plasma levels of interferon-gamma (IFNg), and one of these pts had a 20-fold increase in IFNg transcript levels. The antiangiogenic chemokines MIG and IP-10 rose significantly over baseline in 11 pts. Conclusions: TZB given with low and intermediate pulse-dose IL2 did not produce a robust immune or clinical response in this pt population. A clinical trial in a TZB-naive population may help to determine the immune effects of this combination of IL2 plus TZB.

No significant financial relationships to disclose.

Validation and reduction of FACT/GOG-Ntx subscale for platinum/paclitaxel-induced neurologic symptoms: a gynecologic oncology group study

The FACT/GOG (Gynecologic Oncology Group) Neurotoxicity (Ntx) subscale for assessing platinum/paclitaxel-induced neurologic symptoms was evaluated. The 11-item questionnaire was administered to patients with advanced endometrial cancer treated with doxorubicin/cisplatin (AP) or doxorubicin/cisplatin/paclitaxel (TAP) prior to 1-7 cycles of treatment in GOG 177. The subscale was evaluated in 134 patients in the TAP group for internal reliability, construct validity, criteria validity, sensitivity to treatment differences, and change over time. Cronbach coefficients for internal consistency prior to cycles 1-7 were 0.85, 0.80, 0.84, 0.82, 0.82, 0.85, and 0.84, respectively. The area under the receiver operating characteristic curve was 0.81 for the Ntx score prior to cycle 3. The TAP arm Ntx scores increased significantly from 3.67 at baseline to 8.13 prior to cycle 7; these were higher than the AP arm Ntx scores, which increased from 3.54 at baseline to 4.72 prior to cycle 7. The four sensory items accounted for 80% of treatment differences and 63% of longitudinal changes in the observed subscale score. The 11-item Ntx subscale reliably and validly assesses platinum/paclitaxel-induced peripheral neuropathy. A reduced four-item version is an efficient alternative in measuring this toxicity in clinical trials without compromising its performance.

Study participants’ perceptions of the process and impact of receiving results of N9831

518

Background: There has been growing interest in providing clinical trial participants with study results. We sought to evaluate the process of sharing results from a large cooperative group trial in an effort to guide clinicians and clinical investigators. Methods: We mailed surveys to a subset of women who participated in NCCTG 9831, Phase III Trial of Adjuvant Chemotherapy with or without Trastuzumab for Women with HER2+ Breast Cancer, after the preliminary study results were mailed to participants. Surveys were sent to all trial participants enrolled through 9 CALGB/ECOG institutions. Results: Of 228 surveys sent, 160 (70%) have been returned. Average age of respondents was 51 years (range 26–76); 84% were white; 61% were college graduates; 4% reported recurrent disease. Women reported receiving results by mail (84%), from a health care provider in person or by phone (43%), and/or from the media (47%); 2% reported that they were not informed of the results. 29% heard the results first from the media; 27% first heard by mail. 35% of women might have preferred to be offered results, with the option of not receiving them, but only 4% of women indicated that they would have declined results had they been offered first. 89% of women found the results information easy to understand; 69% correctly interpreted the results of the study; 31% either had an incorrect interpretation or were unsure of the findings. 81% of women were satisfied with how results were shared; 63% of women felt that learning results had an impact on their lives, 24% were more anxious after learning the results; 36% were less anxious. Multivariable analyses evaluating factors associated with greater satisfaction and increased anxiety will be presented. Conclusions: Sharing results is met with overwhelmingly favorable responses from patients, although a substantial proportion of patients may not initially understand the findings. Some patients desire to be offered results first, but few would decline them. The potential for increased anxiety should be considered, and psychosocial support may be required by some. A plan to share results should be routinely and prospectively included in the design of clinical trials. (Supported in part by an ASCO Career Development Award (AHP)

No significant financial relationships to disclose.

Angiogenesis pathway gene polymorphisms associated with clinical outcome in recurrent ovarian cancer treated with low dose cyclophosphamide and bevacizumab: A California Consortium Trial

5017

Background: Despite advances in chemotherapy, ovarian cancer remains a major cause of cancer mortality worldwide. It has therefore become essential to identify novel therapeutic targets, such as angiogenesis which is a complex process regulated by the delicate balance between various local proangiogenic and antiangiogenic proteins. Bevacizumab, a monoclonal antibody binding to VEGF, has shown significant activity in colon, breast, lung and ovarian cancer. There are no established molecular markers to predict response or time to tumor progression for Bevacizumab based chemotherapy. The key enzymes of the VEGF pathway are: Vascular Endothelial Growth Factor (VEGF), VEGF Receptor (VEGFR), Hypoxia Inducible Factor1 (HIF α and β-subunit), Neuropilin1 (NRP), Interleukin-8 (IL-8), Adrenomedullin (AM) and Leptin. Methods: Seventy patients with refractory ovarian cancer were enrolled in a Phase II clinical trial and treated with Cyclophoshamide 50 mg po/Bevacizumab 10 mg/kg IV every 14 days. From 52 patients blood samples were available for gDNA extraction and PCR-RFLP assays. Results: 13 patients had a PR (25%) and 39 were non responders. 31 pts had progressed. Median follow-up of 8.3 months with a median progression-free survival of 6.6 months. Patients who were homozygous A/A or heterozygous A/T genotype at the −251 locus in the IL-8 gene had a lower response rate than those who were T/T (P = 0.047 Fisher’s exact test). Patients with Vegf936 C/C had a median TTP of 6.5 months, pts with any T (T/T, C/T) had a median TTP of 17.2 months. Pts carrying both AM 3’end alleles <14 CA repeats had 3.4 months median TTP, patients with at least one allele >14 showed a median TTP of 6.6 months; for both alleles >14 patients showed 8.7 months of median TPP (P = 0.0006 Log-rank test) Conclusions: Our data suggest for the first time, that IL-8 may be a potential molecular predictor of response to Bevacizumab based chemotherapy. We also demonstrate that both VEGF 936 and the AM 3’ dinucleotide repeat polymorphisms are potential molecular markers for time to tumor progression. A larger prospective study is needed to validate and confirm our preliminary findings. This study was supported by NCI grant NO1 CM 17101.

[Table: see text]

Does adjuvant chemotherapy (CT) exacerbate cognitive impairment in elderly breast cancer (BC) patients? Results of a prospective, longitudinal study

10530

Background: Over the last few years, several retrospective studies documented cognitive impairment in post-chemotherapy BC patients. This decline was hypothesized to be related to CT (“chemo-brain”). More recently, a prospective study (Wefel et al. 2005) documented significant cognitive impairment at baseline, which may impact response to CT, along with treatment related effects. While all previous studies focused on the cognitive response of young BC women, older patients may be much more susceptible to any CT related neurotoxicity. Thus, the current study investigated the incidence and course of cognitive impairment in older, post-menopausal BC patients undergoing CT. Methods: 19 postmenopausal women (≥50 yrs) with no prior chemo or hormonal cancer treatment, and baseline MMSE >23 completed a comprehensive neurocognitive battery of tests along with anxiety and depression measures, prior to treatment. At 6 months, 12 patients returned for post-chemotherapy evaluation, with 8 evaluated at 2 years. Ten DCIS patients were also evaluated at baseline and 6 months. Results: Mean age at baseline was 67 yrs with 14 years of education. Prior to treatment, 37% displayed significant cognitive impairment not accounted for by other variables. At six months, within and between group analyses revealed only improved fine motor speed (Finger tapping, p.01) related to practice effects, but no decline in cognitive functions. There were no significant differences across measures between BC and DCIS groups. However, RCI revealed 3 of 12 BC patients improved on memory tasks. Two year longitudinal analyses again revealed improved motor speed and possible cognitive flexibility (TMT-A, p.02), along with decreased anxiety (STAI-S, p = .03). Conclusion: This small study does not support a finding of treatment-associated cognitive impairment among elderly women receiving standard CT for breast cancer. However, like Wefel’s sample, we documented significant cognitive impairment prior to initiation of treatment. Studies addressing the etiologic mechanism related to this cognitive dysfunction are warranted. Limitations of the current study will be discussed.

No significant financial relationships to disclose.

A phase I and pharmacokinetic study of irinotecan in patients with hepatic or renal dysfunction or with prior pelvic radiation: CALGB 9863

BACKGROUND

To ascertain if hepatic or renal dysfunction or prior pelvic radiation (XRT) leads to increased toxicity at a given dose of irinotecan and to characterize the pharmacokinetics of irinotecan and its major metabolites in patients with hepatic or renal dysfunction.

PATIENTS AND METHODS

Adults with tumors appropriate for irinotecan therapy and who had abnormal liver or renal function tests or had prior radiation to the pelvis were eligible. Patients were assigned to one of four treatment cohorts: I, aspartate aminotransferase (AST) > or = 3x upper limit of normal and direct bilirubin <1.0 mg/dl; II, direct bilirubin 1.0-7.0 mg/dl; III, creatinine 1.6-5.0 mg/dl with normal liver function; IV, prior pelvic XRT with normal liver and renal function. Starting with reduced doses of either 145 or 225 mg/m(2), irinotecan was administered every 3 weeks to at least three patients within each cohort. Irinotecan and its metabolites in the blood were measured in all patients.

RESULTS

Thirty-five patients were evaluable for toxicity. No dose-limiting toxicity was seen in cohort I, although only three patients were treated and at a dose of 225 mg/m(2). Patients with elevations of direct bilirubin had dose-limiting toxicities, even though the starting dose was 145 mg/m(2). These same patients appeared to have comparable exposure to the active metabolite SN-38 as normal patients treated with full-dose irinotecan. Patients with elevations of creatinine or with prior pelvic radiotherapy did not appear to have increased risk of toxicity at the doses explored in this study.

CONCLUSIONS

Patients with elevated bilirubin treated with irinotecan have an increased risk of toxicity and a dose reduction is recommended. Patients with elevated AST, creatinine or prior pelvic radiation do not appear to have increased sensitivity to irinotecan, but the data are not adequate to support a specific dosing recommendation.

Fertile transgenic pearl millet [ Pennisetum glaucum (L.) R. Br.] plants recovered through microprojectile bombardment and phosphinothricin selection of apical meristem-, inflorescence-, and immature embryo-derived embryogenic tissues

Pearl millet [ Pennisetum glaucum (L.) R. Br.] is a drought-tolerant cereal crop used for grain and forage. Novel traits from outside of the gene pool could be introduced provided a reliable gene-transfer method were available. We have obtained herbicide-resistant transgenic pearl millet plants by microprojectile bombardment of embryogenic tissues with the bar gene. Embryogenic tissues derived from immature embryos, inflorescences and apical meristems from diploid and tetraploid pearl millet genotypes were used as target tissues. Transformed cells were selected in the dark on Murashige and Skoog medium supplemented with 2 mg/l 2,4-D and 15 mg/l phosphinothricin (PPT). After 3-10 weeks in the dark, herbicide-resistant somatic embryos were induced to germinate on MS medium containing 0.1 mg/l thidiazuron and 0.1 mg/l 6-benzylaminopurine. Plants were transferred to the greenhouse after they were rooted in the presence of PPT and had passed a chlorophenol red assay (the medium turned from red to yellow). Transgenic plants were recovered from bombardments using intact pAHC25 plasmid DNA, a gel-purified bar fragment, or a mixture of pAHC25 plasmid or bar fragment and a plasmid containing the enhanced green fluorescent protein ( gfp) gene (p524EGFP.1). Analyses by the polymerase chain reaction, Southern blot hybridization, GFP expression, resistance to herbicide application, and segregation of the bar and gfp genes confirmed the presence and stable integration of the foreign DNA. Transformed plants were recovered from all three explants, although transformation conditions were optimized using only the tetraploid inflorescence. Time from culture initiation to rooted transgenic plant using the tetraploid inflorescence ranged from 3-4 months. Seven independent DNA/gold precipitations were used to bombard 52 plates, 29 of which produced an average of 5.5 herbicide-resistant plants per plate. The number of herbicide-resistant plants recovered per successful bombardment ranged from one to 28 and the frequency of co-transformation with gfp ranged from 5% to 85%.

Initial clinical experience with intensity-modulated whole-pelvis radiation therapy in women with gynecologic malignancies

OBJECTIVE

Our goal in this article to describe our initial experience with intensity-modulated whole-pelvis radiation therapy (IM-WPRT) in gynecologic malignancies.

METHODS

Between February and August 2000, 15 women with cervical (9) or endometrial (6) cancer received IM-WPRT. All patients received a treatment planning computed tomography (CT) scan. On each scan, the target volume (upper vagina, parametrial tissues, presacral region, uterus, and regional lymph nodes) and normal tissues (small bowel, bladder, and rectum) were identified. Using commercially available software, an IM-WPRT plan was generated for each patient. The goal was to provide coverage of the target with the prescription dose (45 Gy) while minimizing the volume of small bowel, bladder, and rectum irradiated. Acute gastrointestinal (GI) and genitourinary (GU) toxic effects in these women were compared with those seen in 25 patients treated with conventional WPRT.

RESULTS

IM-WPRT plans provided excellent coverage of the target structures in all patients and were highly conformal, providing considerable sparing of the bladder, rectum, and small bowel. Treatment was well tolerated, with grade 0-1 GI and GU toxicity in 46 and 93% of patients, respectively. IM-WPRT patients had a lower rate of grade 2 GI toxicity (53.4% vs 96%, P = 0.001) than those treated with conventional WPRT. Moreover, the percentage of women requiring no or only infrequent antidiarrheal medications was lower in the IM-WPRT group (73.3% vs 20%, P = 0.001). While grade 2 GU toxicity was also lower in the IM-WPRT patients (6.7% vs 16%), this difference did not reach statistical significance (P = 0.38).

CONCLUSION

IM-WPRT provides excellent coverage of the target structures while sparing critical neighboring structures in gynecology patients. Treatment is well tolerated with less acute GI toxicity than conventional WPRT. More patients and longer follow-up are needed to evaluate the full merits of this approach.

Phase I trial of concomitant chemoradiotherapy for cervical cancer and other advanced pelvic malignancies

Purpose. The aim of this study was to determine the feasibility and toxicity of concomitant hydroxyurea (HU) and escalating doses of 5-fluorouracil (5-FU) in locally advanced cervical cancer and other pelvic malignancies undergoing radiation therapy (RT). Methods. Treatment consisted of 5-FU, HU, and pelvic RT delivered in an alternate-week fashion. 5-FU was administered as a continuous intravenous infusion at a starting dose of 600 mg/m2/day and was escalated to 1000 mg/m2/day in cohorts of three patients. The HU dose was 500 mg twice daily. Chemoradiotherapy was administered on a 5-day cycle. Following a 9-day rest, the cycle was repeated until the completion of the pelvic RT. Results. Twenty-one patients (18 cervix, 1 bladder, 1 vagina, 1 ovary) were enrolled. 5-FU escalation to 1000 mg/m2/day was well tolerated. No patients developed grade 3-4 hematologic toxicity. Grade 2 leukopenia was noted in 3 patients (14.3%). Grade 3 mucositis, diarrhea, and dermatitis occurred in 10, 10, and 5% of patients, respectively. None of the 99 treatment cycles were delayed secondary to acute toxicity. The overall response rate in the 18 cervical cancer patients was 89% (78% complete, 11% partial). Conclusions. Concomitant continuous infusion 5-FU, twice daily HU, and pelvic RT delivered in an alternate-week fashion is well tolerated. Further study is necessary to evaluate the therapeutic efficacy of this regimen in patients with advanced cervical and other pelvic malignancies.