SARC 005: Adjuvant treatment of high risk uterine LMS with gemcitabine/docetaxel followed by doxorubicin: a phase II multi- center trial PI: Martee L. Hensley, MD

Objectives: Determine 2-year PFS among women with uterine LMS treated with gem-doce x 4, followed by doxorubicin x 4 Determine tolerability/toxicity Explore predictors of PFS: age, tumor size, grade, serosal involvement, STS stage v. FIGO stage, mitotic rate, ER, PR, menopausal status at dx

Schema Gemcitabine 900 mg/m 2 over 90 minutes days 1, 8 Docetaxel 75 mg/m2 day 8 q 3 wk x 4 cycles Repeat CT scan Doxorubicin 60 mg/m 2 q 3 w x 4 Repeat CT scan within 6 weeks after CT c/a/p every 3 mo for 2 y, then every 6 mo

Eligibility > 18 years FIGO stage I or II, high grade LMS s/p hysterectomy (serosal involvement IS eligible even though this is FIGO IIIA) <12 weeks from surgery NED by CT within 3 weeks of enrollment Good marrow, kidneys, liver No other cancer within 5 years No prior gem, doce, or dox No prior pelvic RT No current HRT or anti- hormone therapy EF > 50%

Correlative studies Provide tumor details: size, serosal disease, mitotic rate Provide patient details: age, menopausal status at dx and at start of adjuvant therapy Send unstained slides to MSKCC -ER and PR -institutions are paid $75 when slides are received

Statistical issues Target accrual 45 patients Bayesian model for continuous assessment of PFS and safety Accrue at least 15 patients per year Stop early if data suggest 2 year PFS will be no better than 30%

Calculating futility Event: death or evidence of progression Stop if number of events is too many for the total patient-disease-free-days-on study: Number of eventsMinimum total time on test in days

Data capture and management On line SARC registration On line data entry On line CRFs—easy to use - all grade 3 and 4 - selected grade 2 (neuro, hypersensitivity, pulmonary) Data monitored by SARC Some detail for management plan after recurrence Vital status after recurrence every 6 months

Drug details Gemcitabine and docetaxel both supplied Drug distribution from SARC via Biologics to institutions

Open Sites MSKCC Washington Cancer Institute U Michigan Dana Farber U Chicago Penn Onc/Hem Moffitt St. Vincent’s Mass General MedStar MDACC

Results First accrual: 2/13/06 Accrual to date: 22 patients Recurrence/Death events = 0 9 patients have completed all planned therapy Progression-free days = 5211 (as of 28 Sep 07)

Results—are we on target? Target accrual goal 15 patients per year: in 19. months we have 22—seems okay Treatment is highly unlikely to be futile: up to 7 events could have happened in 2477 days and we have had 0 events in over 5700 days

Results: toxicity Gr 3 G/T heme 4 pts (6 events) Gr 3 Dox heme 6 pts (11 events) Gr 4 Dox heme 3 pts (3 events) Gr 2 G/T hypersensitivity3 pts (3 events) Gr 3 G/T hypersensitivity 1 patient (1 event) No pulmonary toxicity 1 patient off study treatment for abnl AST/ALT after C3 G/T— proceeded on to doxorubicin. Remains on study for f/u purposes

For discussion: Clarify in next amendment that patients that must discontinue Gem-Doce for toxicity reasons may remain on study treatment to complete the doxorubicin

Discussion of next steps: Is a phase III trial with a no-chemotherapy control arm accruable? Scientifically reasonable control arm options are: –Pelvic RT (would likely appeal to Gyn Onc/GOG) –Observation (could be hard to accrue)

Phase III size considerations: Home run assumption: 50% PFS at 2 y for “no chemo” group (arm B) and 80% PFS at 2 y for “chemo” group (arm A). -two sided 0.05 level calculation with continuity corrections Type I error Power N for arm A or arm B Total N for two arms

Phase III size considerations: Chemo improves PFS by 30% assumption: 50% PFS at 2 y for “no chemo” group (arm B) and 65% PFS at 2 y for “chemo” group (arm A). -two sided 0.05 level calculation with continuity corrections Type I error Power N for arm A or arm B Total N for two arms

Are there other options?: Increase the sample size of SARC 005 in order to narrow the confidence interval? Open the eligibility to non-uterine LMS (which will of course increase heterogeneity)? Even if we put our hearts and time behind a phase III, funding for such large trial may be very challenging