Phase II Study of Paclitaxel in Patients With Soft Tissue Sarcomas

Purpose. Patients with soft tissue sarcoma (STS) who have previously received standard chemotherapy including adriamycin (doxorubicin), ifosfamide, cyclophosphamide and DTIC (dacarbazine) have very limited therapeutic options. It is important to identify new drugs with some activity in this disease and we therefore undertook this trial to determine the antitumor activity of paclitaxel (Taxol). Methods. We conducted a phase II study of paclitaxel in patients with STS who had received prior standard chemotherapy. Paclitaxel was administered at a starting dose of 200 mg m-2 as a 24-h infusion with STS premedication, every 21 days or upon hematologic recovery (absolute granulocyte count (AGC) ≥ 1500/μl, platelets ≥ 100 000/μl). Neupogen was not used routinely. The study was conducted based on a two-stage design proposed by Simon. Responses were assessed radiographically using standard criteria. Results. Nineteen eligible patients were treated in the first stage of the study. The median age was 50 years (range 20–68 years), and there were nine females and 10 males with Zubrod performance status of 1 or 2. One patient achieved a minor response. Median AGC nadir was 0.1/μl on day 12 with absolute neutropenia lasting 5 days. Median platelet nadir was 171 000/μl on day 9. There were no grade 3/4 non-hematologic toxicities and no deaths related to treatment. Discussion. Paclitaxel, at this dose and schedule, is well tolerated but inactive in this patient population.


Introduction
Paclitaxel (Taxol; Bristol-M eyers Squibb C o., Princeton, N J, U SA) is a novel antimicrotubule agent with a unique m echanism of action. It has been studied in a w ide variety of m alignancies with varyin g levels of ef® cacy. 1 G iven its`broad spectrum' activity, we perform ed a phase II study of paclitaxel in patients with soft tissue sarcom a (STS) w ho had received prior standard chem otherapy. Since leiomyosarcom as of gastrointestinal (G I) origin do not respond well to standard chemotherapy including adriam ycin and ifosfam ide, patients with this histology w ere eligible without prior exposure to chem otherapy. The m ajor objectives of the study w ere to evaluate the ef® cacy and the toxicity pro® le of paclitaxel in patients with ST S.

E ligibility
Patients older than 16 years of age w ith histologic proof of a ST S were eligible. All patients had received prior standard chemotherapy including adriamycin, ifosfam ide, cyclophospham ide and D T IC (dacarbazine) with the exception of patients with G I leiomyosarcom a who could be chem otherapy naõ È ve. Patients were required to have a Zubrod perform ance status of 0± 2, a life expectancy of at least 12 weeks, m easurable or evaluable disease, and relatively norm al organ function de® ned as abso lute granulocyte count (AG C) of $ 1500/m l, platelet count of $ 100 000/m l, total bilirubin , 1.5 m g dl 2 1 serum creatinine , 1.6 m g dl 2 1 and cardiac ejection fraction of . 50% w ithout evidence of congestive heart failure. N o other concurrent therapy was allowed and all patients signed an inform ed consent. Exclusion criteria included pregnancy, serious non-m alignant intercurrent illnesses, serious conduction abnorm alities or cardiac arrhythm ias requiring anti-arrhythm ic m edications.

Treatment plan
T o minim ize the risk of anaphylactoid reactions, all patients were prem edicated with 20 m g of dexamethasone given orally, 12 and 6 h prior to paclitaxel, and 50 m g of diphenhydram ine and 300 m g of C ycles were repeated every 21 days or upon complete recovery. G ranulocyte colony-stim ulating factor was not used prohylactically, but w as allow ed for therapeutic indications. D ose m odi® cation was perform ed based on hem atologic and non-hem atologic toxicities as outlined in T able 1.

Pretreatment evaluation and follow -up studies
Prior to enrolm ent on the protocol, all patients underwent a com plete history and physical examination. Laboratory studies included a com plete blood count (CBC) with differential and platelets, chemistry pro® le (SM A 12), electrolytes and m agnesium, repeated prior to each cycle. Appropriate radiographic studies were perform ed to de® ne the extent of tumor. A cardiac scan or 2D echocardiogram was perform ed to docum ent the cardiac ejection fraction prior to initiation of treatment. F ollowing paclitaxel, patients were followed with at least once weekly C BC w ith differential and platelet counts. C rosssectional radiographic im aging to assess response w as perform ed every two cycles.

Statistical considerations
T he trial was conducted in two stages, using the optim al two-stage design proposed by Sim on. 2 Based on the hypo thesis that a response rate of < 5% would be of no interest and a response rate of $ 20% would be signi® cant, 15 patients were required and 19 patients were entered in the ® rst stage of the study. If no responses were seen in these patients, the study would have to be term inated, otherwise a total of 35 patients needed to be accrued. T his design afford ed a power of 92% to detect a response rate of at least 20% with a rejection error of 10% .

Patient characteristics
N ineteen patients were entered on the study ( Table  2). The m edian age w as 50 years (range 20± 68 years). T here were nine fem ales and 10 m ales with a Zubrod perform ance status of 1 or 2. Eight patients had leiomyosarcom a, three had unclassi® ed sarcom a, three had strom al sarcom a of the breast, two had pleom orphic rhab dom yosarcom a, and one each had alveolar soft-part sarcom a, m alignant ® brous histiocytom a and synovial sarcom a. Seventeen patients had metastatic disease and tw o had persistent or locally recurrent disease. M edian num ber of cycles of paclitaxel chem otherapy adm inistered was tw o (range 1± 7).

D iscussion
According to the Am erican C ancer Society' s estim ates, approxim ately 6400 new ST S were diagnosed in 1996. 3 C hem otherapeutic agents with signi® cant activity in these tum ors include adriam ycin, ifosfam ide, cyclophosph am ide and D TIC. Patients whose disease recurs or progresses after a trial of these standard agents have a guarded prognosis, and trials of new and interesting drugs are warranted. W e have previously published our results with paclitaxel in bone sarcom as. 4 In this similar trial, we chose to evaluate the activity of pactitaxel at a dose of 200 m g m 2 2 administered as a 24-h infusion, every 3 weeks in patients failing prior treatment with standard chem otherapy. N o objective responses were noted in our trial suggesting a low level of activity of this drug in this patient population. A recently reported trial of paclitaxel in patients with previously untreated advanced ST S revealed a response rate of 12.5% . 5 T he patient population selected for our trial w as refractory to standard chem otherapy, and therefore less likely to respond to paclitaxel. Six patients had leiom yosarcom a of GI origin which did not respond to paclitaxel.
In conclusion, paclitaxel at a dose of 200 m g m 2 2 over 24-h every 3 w eeks in patients w ith ST S refractory to standard chem otherapy is well tolerated but inactive. Phase II trials of new agents are w arranted.