Lack of Activity of Docetaxel in Soft Tissue Sarcomas: Results of a Phase II Study of the Italian Group on Rare Tumors

Purpose. The prognosis of advanced soft tissue sarcoma is poor, only a few drugs showing some activity with response rates around 15– 25%. Consequently drug development seems mandatory to improve treatment outcome. Following previous favourable EORTC experience, the Italian Group on Rare Tumors started a phase II study with docetaxel to confirm the activity of this drug in soft tissue sarcoma. Patients and methods. Thirty-seven patients with soft tissue sarcoma resistant to at least one anthracyclinecontaining regimen were enrolled in a phase II multicenter study evaluating docetaxel 100 mg/m2 in a 1-h i.v. infusion q3 weeks. Results.Thirty-seven patients were enrolled onto this phase II study and 36 were evaluable for response. Only one partial remission was observed [2.8% with 95% confidence interval (CI) 0.1– 16.2%]. Median progression-free and overall survival were 42 and 350 days, respectively. Neutropenia and leukopenia as well as cutaneous manifestations were the most common toxicities. Discussion. The results of this phase II study do not confirm a previous EORTC repor t on the activity of docetaxel in soft tissue sarcoma, but are consistent with other more recent phase II studies. The accumulated evidence does not justify the use of this drug in the management of patients suffering from this disease, resistant to anthracyclinecontaining regimens.


Introduction
T he prognosis of advanced soft tissue sarcom a is poor, with only three drugs, nam ely anthracyclines, ifosfam ide and dacarbaz ine, showing som e activity w ith a response rate of aroun d 15± 25% . 1± 4 T he com bination of such drugs does not signi® cantly in crea se th e resp o n se rate o ve r sin g le -d r u g doxoru bicin . 5± 9 C onsequently new dru g developm ent seems m andatory to im prove the treatm ent ou tco m e of patien ts su ffering fro m this d isease. U nfortunately alm ost all pha se II studies of new agents perform ed in recent years failed to show any activity in the treatm ent of refra ctor y soft tissu e sarcom as. 5,9,10 D ocetaxel (Taxotere¾ ) is a sem isynthetic analog of paclitaxel derived from the needles of the E uropean yew, Taxus baccata. D ocetaxel was shown to have superior in vivo antitum or activity as com pared to taxol in the B16 m elanom a m odel 11 and has show n a favorable response rate in breast cancer. 12 A phase II study of docetaxel by the EO RT C Soft T issue an d B one Sarcom a G roup 13 repor ted an overall response rate of 17% in 29 soft tissue sarcom a patients refractory to an anthrac ycline-containing regim en. According to these data, docetaxel has been suggested as the third active agent in the treatm ent of advanced soft tissue sacom a supporting its use as a ® rst-line treatm ent.
In Septem ber 1995, the Italian G roup on Rate Tum ors started a m ulticenter phase II study in refractory soft tissue sarcom a to con® rm previous E ORTC experience.

Patient population
To be eligible for the study, patients were required to have a pathologically con® rm ed diagnosis of soft tissue sa rco m a an d to be resista n t to at least o n e anthracycline-containing regim en.
Resistance to previous chem otherapy was de® ned as either progressio n d uring or relapse within 6 m onths from the end of previous chem otherapy. Eligible patients were to have received no regim ens including docetaxel. O ther inclusion criteria were as fo llow s: a ge b etw een 1 6 a n d 6 5 year s, E C O G perform ance status £ 2, an expected survival duration of 3 m onths, and previous chem otherapy completed at least 4 weeks before study entry. Patients were required to have bidimensionally m easurable indicator lesions, w hile patients with evaluable lesions alone as well as those with pleural effusion, m alignant ascites and previously irradiated lesions were not considered fo r en ro llm en t in th e stu d y. R eq u ir em en ts fo r eligibility also included leukocyte count 3500 cmm ; granulocyte count 1500 cmm ; platelet count 100 000 cm m , an d n or m al liver and re nal fu nction . Patients w ith cardiac disease, brain m etastases, other serious m edical illnesses and other m alignant tumors were excluded. Protocol and inform ation sheet were approved by the Scienti® c and Ethical Com m ittee of all the participating institutions and all patients had to sign a written, inform ed consent.

Treatment plan
Before treatm ent, all patients had a com plete history and physic al exam ination as well as baseline laborato r y tests, ch est X -rays, E C G , an d ab d o m in al ultrasou nd . C T scan , M R I and bo ne scan were perform ed according to the clinical indications. D ocetaxel was supplied by Rhone Poulenc Rorer com pany as a concentrated sterile solutio n w ith 40 mg/m l (80 m g/2 m l, per vial) in polysorbate 80 (Tween 80). Before infusion, docetaxel was diluted in 250 m l of 5% dextrose and was given at a dose of 100 m g/m 2 i.v., over 1 h. C ycles were restarted ever y 21 days in the absence of toxicities requiring treatm ent delay. T herapy was continued until d isease progression or unacceptable toxicity. T he patients did not receive anti-em etic treatment un less th e p atien t h ad exp er ien ced n au se a an d vom iting with a previous dose. Pre-m edication for acute hypersensitivity reactions consisted of prednisone 50 m g orally 24, 16, 8 and 1 h before and 20, 32, 44, 56, 68 and 80 h after docetaxel. Prophylactic colony-stimulating factors were not given.Toxic effects were reported according to the South West O ncology G roup Toxicity C riteria.
14 D ose red uction s were perform ed as follows: in patients with grade 3 neutropenia and/or grade 2 peripheral neuropathy and/or febrile neutropenia the following cycle was lowered of a 25%; in the absence of recover y of myelotoxicity after 3 weeks, the subsequent dose was delayed until toxicity resolved.

Response criteria
Response evaluation was perform ed every two cycles reporting the initial param eter lesions. According to the South West O ncology Group criteria, 14 com plete rem ission was de® ned as the com plete disapp earance of all detectable lesions for at least 4 weeks. Partial rem ission was de® ned as a 50% decrease in the sum of the products of the diam eters of m easurable lesions for at least 4 weeks, in the absence of sim ultaneous increase of any lesion and/or app earance of any new lesion. Progressive disease was de® ned as an increase of 50% in the size of a m easurable lesion or the app earance of a new lesion. D uration of response was determ ined by the interval between the day of the ® rst treatm ent and the date of clear evidence of disease prog ression.

Statistical analysis
T he study was designed as a two-stage trial, according to Sim on 's optim al d esign , 15 w ith the follow ing sp eci® ca tio n s: a 9 0% p ro b ab ility o f a ccep tin g docetaxel for further studies was required if the true response rate was 20% whereas the probability of accepting it for further studies, if the true response rate was 10% , was set at 5% . Accordingly, 12 patients had to be entered and evaluated in the ® rst stage of the trial. If at least one response was observed, 25 m ore patients had to be accrued, and, overall, at least four responses had to be obser ved out of 37 patients in order to consider the drug sufficiently active to warrant further studies. All patients who started the experim ental treatm ent were included in the analyses of response. All patients in whom an objective response was not d em o n strated , in clu d in g early pro gression s an d deaths, those w ho discontinued treatm ent due to toxicity, and those in w hom response was not evaluated according to the protocol were considered as treatment failures and included in the denom inator of the proportion of responses.
Survival and time to disease progression were evaluated according to the Kaplan± M eier m ethod. 16

Patient characteristics
Between Septem ber 1995 an d O ctober 1996, 37 patients were enrolled onto this phase II study by nine participating centres. O ut of these 37 patients, ® ve were not eligible because of performance status 3 (one case), low neutrophiles <2000 cmm (two cases) and abnorm al renal function (two cases), respectively. All patients had m easurable lesions. Pre-treatm ent patient characteristics are listed in Table 1.The median age was 44 years (range 20± 64 years) w ith a m edian W H O perform ance status of 0 (range 0± 3). Thirtyfour patien ts (91.9 % ) prese nted w ith m etastatic disease an d only three (8.1 % ) w ith locoregio nal advanced disease.Twenty-four patients (64.9% ) were pre-treated with only one chemotherapy regim en, nine (24.3% ) with two, and four with four, respectively. Previous radiotherapy had been adm inistered to 19 patients. The histologic subtypes were as follows: leiomyosarcom a in nine cases, synovial sarcom a in eight cases, m alignant ® brous histiocytom a and schwannom a in four cases, ® brosarcom a, liposarcom a and rhabdom yosarcom a in two cases, em ang iopericytoma and epithelioid sarcom a in one case, and unclas-si® ed sarcom a in four cases, respectively.

Response
All 37 patients were considered evaluable for response except one who died before starting chem otherapy. A total of 116 cycles were adm inistered with a m edian of four courses (range 1± 8). The therapeutic results are reported in Table 2. O nly one partial response (2.8% with a 95% con® dence interval 0.1± 16.2% ) was observed. In 10 patients, stable disease was docum ented, while all other patients showed disease prog ression.
T he m edian time to progression was 42 days, and m edian overall survival was 350 days, respectively ( Table 2 and Fig. 1). D eath was due to tum or progression in 31 patients, to infection in one patient not eligible because of both anem ia and abnorm al renal function, and cause of death was not docum ented in two cases.

Toxicity
The highest grade of toxicity according to the SW OG Toxicity C riteria 14 for each patient is listed in Table   3. Grade 3 and 4 neutropenia and leukopenia were the m ost comm on toxicities, while anemia and thrombocytopenia were rarely docum ented. Febrile neutro-pen ia an d in fe ctio n w ere o b ser ved in a sm all proportion of cases although only one patient died of this cause. M ore than half of the patients developed sym ptom atic rash as well as m ore severe cutaneous m anifestations including super® cial dry desquam ation of the hands and feet and dystrophic nail changes.
Hair loss was observed in almost all evaluable patients. Vom iting and diarrhea, m ucositis and neurotoxicity were rarely reported. F luid retention occurred in 21% of patients but only in 6% of cases reached grade 2± 3 level and a hypersensitivity reaction was observed in 9% of cases. repor ts have also suggested a low level of activity of dacarbazin e. 4,9 N oneth eless the resu lts achieve d w ith these dr ugs g iven alon e o r in com bin ation are poor w ith n o clear bene® t in the advan ced d isease as well as in the ad ju van t setting. 1,5,9 T herefo re the search for new active agen ts re m ains an im portan t issue. H owever the m ajority of phase II stu dies fa iled to show any activity by alm ost all investigated dru gs. 1,5,9,10 T he E O RT C S oft T issu e an d B on e S arco m a G rou p repor ted the resu lts of docetaxel in previously treated soft tissu e sarcom a w ith ® ve par tial rem issio ns (17% ) in 29 evalu able patients. 13 Based on these p o sitive resu lts it w as su gg ested th at docetaxel co uld be an effective d rug , warranting ® rst-lin e phase II stud ies. H owever, the results obtained in the present study failed to con® rm the activity of this drug in soft tissue sarco m a w ith o n ly o n e p ar tial rem issio n in 37 anthracycline-resistant patients. O ur results are in agreem ent w ith the low response rate reported by three studies utilizing docetaxel as ® rst-line. 17± 19 Based on our data and that of other authors, the continued use of docetaxel for treatment of soft tissue sarcom a is not justi® ed.