Adjuvant chemotherapy for adult soft tissue sarcomas of the extremities and girdles: results of the Italian randomized cooperative trial

PURPOSE

Adjuvant chemotherapy for soft tissue sarcoma is controversial because previous trials reported conflicting results. The present study was designed with restricted selection criteria and high dose-intensities of the two most active chemotherapeutic agents.

PATIENTS AND METHODS

Patients between 18 and 65 years of age with grade 3 to 4 spindle-cell sarcomas (primary diameter > or = 5 cm or any size recurrent tumor) in extremities or girdles were eligible. Stratification was by primary versus recurrent tumors and by tumor diameter greater than or equal to 10 cm versus less than 10 cm. One hundred four patients were randomized, 51 to the control group and 53 to the treatment group (five cycles of 4'-epidoxorubicin 60 mg/m(2) days 1 and 2 and ifosfamide 1.8 g/m(2) days 1 through 5, with hydration, mesna, and granulocyte colony-stimulating factor).

RESULTS

After a median follow-up of 59 months, 60 patients had relapsed and 48 died (28 and 20 in the treatment arm and 32 and 28 in the control arm, respectively). The median disease-free survival (DFS) was 48 months in the treatment group and 16 months in the control group (P =.04); and the median overall survival (OS) was 75 months for treated and 46 months for untreated patients (P =.03). For OS, the absolute benefit deriving from chemotherapy was 13% at 2 years and increased to 19% at 4 years (P =.04).

CONCLUSION

Intensified adjuvant chemotherapy had a positive impact on the DFS and OS of patients with high-risk extremity soft tissue sarcomas at a median follow-up of 59 months. Therefore, our data favor an intensified treatment in similar cases. Although cure is still difficult to achieve, a significant delay in death is worthwhile, also considering the short duration of treatment and the absence of toxic deaths.

Ifosfamide in the Adjuvant Therapy of Soft Tissue Sarcomas

Ifosfamide and anthracyclines are the only active agents in advanced soft tissue sarcomas. Doxorubicin was always used in sarcomas, whereas ifosfamide was reintroduced in the clinic after the discovery of mesna which prevents its typical dose-limiting toxicity: hemorrhagic cystitis. In the adjuvant setting, doxorubicin was used alone or in combination in the first-generation trials, whereas its parent compounds epirubicin and ifosfamide were employed in the second-generation adjuvant trials, which started in the early 90s. Other relevant aspects of the second-generation trials are the use of the hematopoietic growth factors and the increase of the dose intensity, the introduction of more restrictive selection criteria and the use of the two most active agents, ifosfamide and anthracyclines. Only the Italian cooperative trial has been concluded, and the results reported and updated. After a median follow-up of 89.6 months (range 56-119), the intention-to-treat analysis still reveals a difference in overall survival which, however, is not statistically significant. However, the 5-year overall survival estimate, which is a reasonable end point for the survival analysis of adjuvant treatment in soft tissue sarcomas, was 66.0 and 46.1% for the treatment and the control groups, respectively (p = 0.04).

Prognostic and predictive role of neutrophil/lymphocytes ratio in metastatic colorectal cancer: a retrospective analysis of the TRIBE study by GONO

Background

Neutrophil/lymphocyte ratio (NLR), defined as absolute neutrophils count divided by absolute lymphocytes count, has been reported as poor prognostic factor in several neoplastic diseases but only a few data are available about unresectable metastatic colorectal cancer (mCRC) patients (pts). The aim of our study was to evaluate the prognostic and predictive role of NLR in the TRIBE trial.

Patients and methods

Pts enrolled in TRIBE trial were included. TRIBE is a multicentre phase III trial randomizing unresectable and previously untreated mCRC pts to receive FOLFOXIRI or FOLFIRI plus bevacizumab. A cut-off value of 3 was adopted to discriminate pts with low (NLR < 3) versus high (NLR ≥ 3) NLR, as primary analysis. As secondary analysis, NLR was treated as an ordinal variable with three levels based on terciles distribution.

Results

NLR at baseline was available for 413 patients. After multiple imputation at univariate analysis, patients with high NLR had significantly shorter progression-free survival (PFS) [hazard ratio (HR) 1.27 (95% CI 1.05-1.55), P = 0.017] and overall survival (OS) [HR 1.56 (95% CI 1.25-1.95), P < 0.001] than patients with low NLR. In the multivariable model, NLR retained a significant association with OS [HR 1.44 (95% CI 1.14-1.82), P = 0.014] but not with PFS [HR 1.18 (95% CI 0.95-1.46), P = 0.375]. No interaction effect between treatment arm and NLR was evident in terms of PFS (P for interaction = 0.536) or OS (P for interaction = 0.831). Patients with low [HR 0.84 (95% CI 0.64-1.08)] and high [HR 0.73 (95% CI 0.54-0.97)] NLR achieved similar PFS benefit from the triplet and consistent results were obtained in terms of OS [HR 0.83 (95% CI 0.62-1.12) for low NLR; HR 0.82 (95% CI 0.59-1.12) for high NLR].

Conclusion

This study confirmed the prognostic role of NLR in mCRC pts treated with bevacizumab plus chemotherapy in the first line, showing the worse prognosis of pts with high NLR. The advantage of the triplet is independent of NLR at baseline.

Prognostic factors in soft tissue sarcomas: a study of 395 patients

AIMS

The aim of this study was to report prognostic factors, end-points of local recurrence, distant recurrence, post-metastasis survival, and overall survival in a cohort of patients with soft tissue sarcomas.

METHODS

We analysed a database of 395 patients affected by primary soft tissue sarcomas of various primary sites, treated and followed up at the Centro di Riferimento Oncologico, Aviano, Italy from January 1985 to January 1997.

RESULTS

Grade, size, stage, surgical margins, distant metastasis, age, sex, performance status, and haemoglobin value were significant for overall survival. Histology, grade, stage, and surgical margins were significant for local recurrence. Grade, size, and stage, were significant for distant recurrence; and surgical margin was significant variable for post-metastasis survival.

CONCLUSIONS

Grade, size, and TNM stage (UICC/AJCC) have stronger prognostic significance for overall survival and distant recurrence than for local relapse. Positive surgical margins are the main predictors for local relapse. Age was the most consistent adverse independent prognostic factor for survival.

A randomized, multicenter, phase II study of vandetanib monotherapy versus vandetanib in combination with gemcitabine versus gemcitabine plus placebo in subjects with advanced biliary tract cancer: the VanGogh study

BACKGROUND

The management of biliary tract cancers (BTCs) is complex due to limited data on the optimal therapeutic approach. This phase II multicenter study evaluated the efficacy and tolerability of vandetanib monotherapy compared with vandetanib plus gemcitabine or gemcitabine plus placebo in patients with advanced BTC.

PATIENTS AND METHODS

Patients were randomized in a 1 : 1 : 1 ratio to three treatment groups: vandetanib 300 mg monotherapy (V), vandetanib 100 mg plus gemcitabine (V/G), gemcitabine plus placebo (G/P). Vandetanib (300 mg or 100 mg) or placebo was given in single oral daily doses. Gemcitabine 1000 mg/m(2) was i.v. infused on day 1 and day 8 of each 21-day cycle. The primary end point was progression-free survival (PFS). Secondary end points were: objective response rate (ORR), disease control rate, overall survival, duration of response, performance status and safety outcomes.

RESULTS

A total of 173 patients (mean age 63.6 years) were recruited at 19 centers across Italy. Median (95% confidence intervals) PFS (days) were 105 (72-155), 114 (91-193) and 148 (71-225), respectively, for the V, V/G and G/P treatment groups, with no statistical difference among them (P = 0.18). No statistical difference between treatments was observed for secondary end points, except ORR, which slightly favored the V/G combination over other treatments. The proportion of patients reporting adverse events (AEs) was similar for the three groups (96.6% in V arm, 91.4% in the V/G arm and 89.3% in the G/P arm).

CONCLUSIONS

Vandetanib treatment did not improve PFS in patients with advanced BTC. The safety profile of vandetanib did not show any additional AEs or worsening of already known AEs.

CLINICAL TRIAL NUMBER

NCT00753675.

Utility of combined use of plasma levels of chromogranin A and pancreatic polypeptide in the diagnosis of gastrointestinal and pancreatic endocrine tumors

BACKGROUND

Chromogranin A (CgA) is considered the most accurate marker in the diagnosis of gastro-entero-pancreatic (GEP) endocrine tumors. Pancreatic polypeptide (PP) has also been proposed to play this role, but then not used due to its low sensitivity. The aim of the present study was to determine whether the assessment of PP would improve the diagnostic reliability of CgA in patients with GEP tumors.

PATIENTS AND METHODS

Both markers were assessed in 68 patients [28 functioning (F), 40 non functioning (NF)]. Twenty-seven patients disease-free (DF) after surgery, and 24 with non-endocrine tumors (non-ETs) were used as control groups.

RESULTS

CgA sensitivity was: 96% in F, 75% in NF, 74% in pancreatic, and 91% in gastrointestinal (GI) tumors. Specificity was 89% vs DF, and 63% vs non-ETs. PP sensitivity was: 54% in F, 57% in NF, 63% in pancreatic, and 53% in GI tumors. Specificity was 81% vs DF, and 67% vs non-ETs. By combining the two markers a significant gain in sensitivity vs CgA alone was obtained: overall in GEP tumors (96% vs 84%, p = 0.04), in NF (95% vs 75%, p = 0.02), and in pancreatic (94% vs 74%, p = 0.04). More specifically, a 25% gain of sensitivity was obtained in the subgroup of NF pancreatic tumors (93% vs 68%, p = 0.04).

CONCLUSION

The combined assessment of PP and CgA leads to a significant increase in sensitivity in the diagnosis of GEP tumors, particularly in pancreatic NF.

Somatostatin receptor scintigraphy versus chromogranin A assay in the management of patients with neuroendocrine tumors of different types: clinical role

BACKGROUND

Current diagnosis and staging of neuroendocrine tumors (NETs) are significantly improved by the introduction of the chromogranin A (CgA) assay in plasma or serum as a tumor marker, and by the use of somatostatin receptor scintigraphy (SRS) for tumor localization. However, the clinical role of CgA assay compared with SRS in the management of NETs has not been well elucidated.

PATIENTS AND METHODS

Sixty-three consecutive patients with a histological diagnosis of NET underwent plasma CgA assay and SRS for tumor staging (23 cases), evaluation of tumor response (18 cases) and evaluation of tumor recurrence on follow-up (22 cases). Twenty-one patients had well-differentiated neuroendocrine tumors (WDNETs: 18 gastroenteropancreatic tumors and three lung NETs); 22 patients had well-differentiated neuroendocrine carcinomas (WDNECs: 17 gastroenteropancreatic carcinomas, two lung neuroendocrine carcinomas and three neuroendocrine carcinomas of unknown origin) and 20 patients had poorly differentiated neuroendocrine carcinomas (PDNECs: 14 extra-pulmonary small-cell carcinomas and six Merkel cell carcinomas). Almost all (58 of 63) NETs were non-functioning. The quantitative determination of CgA was performed in plasma using an enzyme immunoassay with a cut-off value fixed at 34 U/l. Scintigraphies with indium 111-DTPA-octreotide ((111)In-pentetreotide) included whole-body images and single photon emission computed tomography (SPECT) scans of the chest and abdomen.

RESULTS

SRS results were compared with CgA findings and final clinical data. The overall sensitivity of SRS and CgA, based on the final clinical data, was 77% and 55%, respectively, whereas the specificity of both SRS and CgA was 94%. Concerning tumor type, SRS accuracy was 95% for WDNETs, 86% for WDNECs and 60% for PDNECs; CgA accuracy was 76% for WDNETs, 68% for WDNECs and 50% for PDNECs. With regard to disease extent, SRS sensitivity was 100% for limited disease and 72% for advanced disease; CgA sensitivity was 43% for limited disease and 57% for advanced disease.

CONCLUSIONS

In our NET series, SRS proved to be more sensitive than CgA, with equivalent specificity. Tumor differentiation influences the sensitivity of SRS and CgA analysis. In addition, the plasma CgA level is related to tumor secretory activity. Nevertheless both SRS and CgA should be considered useful tools in the diagnostic work-up of NET patients.

FOLFOXIRI or FOLFOXIRI plus bevacizumab as first-line treatment of metastatic colorectal cancer: a propensity score-adjusted analysis from two randomized clinical trials

BACKGROUND

FOLFOXIRI plus bevacizumab is a valid option as upfront treatment for metastatic colorectal cancer (mCRC) patients. While several trials investigated the effect of combining bevacizumab with different chemotherapy regimens, including fluoropyrimidines monotherapy and oxaliplatin- or irinotecan-containing doublets, no randomized comparison assessing the impact of the addition of bevacizumab to FOLFOXIRI is available.

PATIENTS AND METHODS

A total of 122 mCRC patients received first-line FOLFOXIRI in the phase III trial by the GONO (FOLFOXIRI group) and 252 patients received first-line FOLFOXIRI plus bevacizumab in the TRIBE trial (FOLFOXIRI plus bevacizumab group). A propensity score-adjusted method was adopted to provide an estimation of the benefit from the addition of bevacizumab to FOLFOXIRI in terms of survival and activity parameters.

RESULTS

Patients in the FOLFOXIRI group had more frequently Eastern Cooperative Oncology Group performance status of one or two, high Köhne score, metachronous and liver-limited disease, had previously received adjuvant treatments and had their primary tumors resected. The median progression-free survival (PFS) was 12.3 months in the FOLFOXIRI plus bevacizumab group compared with 10.0 months in the FOLFOXIRI group {propensity score-adjusted hazard ratio (HR) 0.74 [95% confidence interval (CI) 0.59-0.94], P = 0.013}. This association was significant also in the multivariable model (P = 0.024). The median OS was 29.8 months in the FOLFOXIRI plus bevacizumab group compared with 23.6 months in the FOLFOXIRI group [propensity score-adjusted HR: 0.72 (95% CI 0.56-0.93), P = 0.014]. At the multivariable model, the addition of bevacizumab was still associated with significantly longer OS (P = 0.030). No significant differences in RECIST response rate (RR) [65.1% versus 55.7%; propensity score-adjusted odds ratio (OR): 1.29 (95% CI 0.81-2.05), P = 0.280], early RR [62.7% versus 57.8%; OR: 1.14 (95% CI 0.68-1.93), P = 0.619] and median depth of response (42.2% versus 53.8%, P = 0.259) were reported.

CONCLUSIONS

Though in the absence of a randomized comparison, the addition of bevacizumab to FOLFOXIRI provides significant benefit in PFS and OS, thus supporting the use of FOLFOXIRI plus bevacizumab as upfront treatment for mCRC patients.

TRIALS' NUMBERS

NCT01219920 and NCT00719797.

Dose-finding study of epidoxorubicin and docetaxel as first-line chemotherapy in patients with advanced breast cancer

BACKGROUND

Anthracyclines and taxanes are the most active drugs against breast cancer and the search after their optimal combination is under intensive investigation in both the advanced and early disease settings. A dose-finding study of epidoxorubicin (E) and docetaxel (D) was conducted in advanced breast cancer (ABC) to define the maximum tolerated dose (MTD) of the combination with and without granulocyte colony-stimulating factor (G-CSF) support and to characterise its toxicity and activity profile.

PATIENTS AND METHODS

Forty-two patients who received neither palliative chemotherapy nor adjuvant anthracyclines (55% with dominant visceral disease and 66% with > or = 2 sites involved) with measurable/evaluable lesions, were treated at four dose levels starting from E 75 mg/m2 and D 75 mg/m2 to E 120 mg/m2 and D 85 mg/m2. A maximum of four cycles of the combination was given every three weeks and four additional cycles of single agent D were allowed in responding patients. Cardiac function was monitored at baseline and at every second course by echocardiography.

RESULTS

Febrile neutropenia (two patients) and prolonged, severe neutropenia (absolute neutrophil count (ANC) < 0.1 x 10(9)/l for more than three days; one patient) defined the MTD of the combination without G-CSF support at E 90 mg/m2 and D 75 mg/m2. G-CSF was then routinely administered from the subsequent dose level of E 120 mg/m2 and D 75 mg/m2. The MTD with G-CSF support was established at E 120 mg/m2 and D 85 mg/m2 (one patient with neutropenic fever together with failure of ANC recovery at day 21, three patients with ANC less than 0.1 x 10(9)/l for more than three days, one patient with both and one patient with grade 4 thrombocytopenia and toxic death from typhlitis while neutropenic). No severe neurotoxicity, mucositis, or fluid retention were observed and there were no clinical signs of cardiotoxicity. Antitumor activity was not a primary endpoint of the study: the overall response rate (ORR) in 40 evaluable patients was 60% (95% confidence interval: 43%-75%, 58% in liver disease, 84% in soft tissue) with no apparent dose-related effect. After a median follow-up of 19 months (range 2-30+), the overall time to progression (TTP) in nine patients without maintenance hormonal therapy was five months.

CONCLUSIONS

The combination of E and D proved to be an effective and safe regimen in poor- prognosis patients with ABC. G-CSF support allowed higher doses to be delivered safely but dose escalation did not translate into improved response rates (RR). The MTD without growth factors support was used, in a phase II trial, which also included patients with previous anthracycline-containing adjuvant regimens.

Increasing 4'-epidoxorubicin and fixed ifosfamide doses plus granulocyte-macrophage colony-stimulating factor in advanced soft tissue sarcomas: a pilot study

PURPOSE

To determine the maximum-tolerated dose (MTD) of 4'-epidoxorubicin (EPI) in combination with full dose of ifosfamide (IFO) when granulocyte-macrophage colony-stimulating factor (GM-CSF) was used, to estimate its clinical efficacy, and to evaluate the mobilization of hematopoietic progenitors.

PATIENTS AND METHODS

Previously untreated advanced patients were treated with fixed doses of IFO at 1.8 g/m2/d for 5 days and escalating doses of EPI. The starting dose level of EPI was 50 mg/m2 bolus on days 1 and 2; subsequent levels were 60 mg/m2 and 70 mg/ m2 given on days 1 and 2. GM-CSF (5 micrograms/kg/d) was administered from days +6 to +19. Clinical evaluation of response was performed after three consecutive cycles. Mobilization of hematopoietic progenitors was evaluated as day 14 CFU-GM after the first cycle only.

RESULTS

Overall, six, 18, and 13 assessable patients were entered onto each EPI dose level, respectively. The first and the second EPI level were considered feasible. Conversely, at the third level, only six of 13 patients [46%] tolerated full EPI doses at the scheduled time. Therefore, the dose-intensity of the three levels was 100%, 99.7%, and 86.1%, respectively. Overall, 20 of 37 patients (54%) obtained an objective response. The response rates for the three EPI dose levels were significantly different [17%, 33%, and 100%, respectively; test for trend, P < .001]. Considering only lung metastases, the overall response rate was 72% (20%, 66%, and 100% for the three EPI levels, respectively). The most relevant mobilization effect was obtained at the third EPI level, when both GM-CSF and IL-3 were used as in vitro-stimulating factors.

CONCLUSION

The third EPI level (70 mg/m2 on days 1 and 2) is the MTD of this program, since it was administered, without dose reduction or treatment delay, for three consecutive cycles in less than half of the patients. Nevertheless, this level proved to be interesting with regard to response rate (13 of 13 objective responses) and in mobilization of the hematopoietic progenitors.

Antithrombin III deficiency as a risk factor for catheter-related central vein thrombosis in cancer patients

The fibrin sleeve of venous catheters (VC) and parietal thrombi represent frequent and dangerous side-effects of central venous catheterization (CVC), due to the risk of embolism. Reduced levels of coagulation clotting factors inhibitors (such as Antithrombin III) are known to be associated with increased thrombogenic risk. The aim of this study was to evaluate the role of Antithrombin III (AT III) deficiency as a risk factor for thrombosis in cancer patients undergoing CVC. The study groups included patients with a reduced AT III activity (< 70%, 20 consecutive patients) and with normal AT III values (> 70%, 20 randomly selected patients), requiring a VC for chemotherapy and/or total parenteral nutrition. The study protocol included evaluation of Hb, PLTs, PT (INR), aPTT, Fibrinogen and AT III at days 0, 1, 3 and 8 after CVC and upon VC removal. Peripheral and pullout phlebographies were performed in all patients on catheter withdrawal. A quantitative scale was developed to evaluate both VC and parietal thrombus degree in each catheter-containing venous segment (subclavian, innominate, superior vena cava); the sum of the mean values was defined as overall thrombus. The average VC dwelling time was similar in both groups. There were no significant differences in Hb, PLTs, PT (INR), aPTT, Fibrinogen and in the remaining parameters of the study between the two groups. The group with AT III deficiency presented a higher degree of both parietal (p < 0.05) and overall thrombus (p < 0.02). Data showed a higher severity of CVC-related thrombosis in patients with AT III deficiency than in the control group. Further studies are needed to evaluate whether the therapeutically-induced normalization of AT III levels can reduce the thrombosis degree.

Cost Evaluation of Irinotecan-Related Toxicities Associated With the UGT1A1*28 Patient Genotype

The adoption of a preemptive UGT1A1*28 genotyping to increase irinotecan safety in clinical practice is still limited. This is the first actual study of costs associated with the management of irinotecan-related toxicities, and their association with UGT1A1*28 genotype. A retrospective analysis of the cost of toxicity management was conducted on 243 metastatic colorectal cancer patients enrolled in a clinical trial and treated with standard of care FOLFIRI (5-fluorouracil combined with irinotecan). The mean predicted cost per patient was higher for *28/*28 (€4,886), vs. *1/*1 (€812), (regression coefficient 1.79, 95% confidence interval (CI) = 1.31-2.28; P < 0.001) and for *1/*28 (€1,119) vs. *1/*1 (regression coefficient 0.32, 95% CI = 0.04-0.60; P = 0.024). This is consistent with a different grade 4 toxicity profile among the three genotypes, and a higher frequency of costly interventions like hospitalization among patients with the *28 allele. A differential toxicity management cost by *28 genotype is herein demonstrated, representing a first step towards the demonstration of the test clinical utility.

Impact of age and gender on the safety and efficacy of chemotherapy plus bevacizumab in metastatic colorectal cancer: a pooled analysis of TRIBE and TRIBE2 studies

BACKGROUND

The phase III TRIBE and TRIBE2 studies randomized metastatic colorectal cancer patients to first-line FOLFOXIRI/bevacizumab or a doublet (FOLFIRI or FOLFOX)/bevacizumab. The studies demonstrated a significant benefit from the triplet at the price of an increased incidence of chemotherapy-related adverse events (AEs). In both trials, males and females aged between 18 and 70 years with ECOG PS ≤2 and between 71 and 75 years with ECOG PS = 0 were eligible. We investigated the effect of FOLFOXIRI/bevacizumab versus doublets/bevacizumab according to age and gender.

PATIENTS AND METHODS

Subgroup analyses according to age (<70 versus 70-75 years) and gender were carried out for overall response rate (ORR), progression-free survival (PFS), and AE rates.

RESULTS

Of 1187 patients, 1005 (85%) were aged <70 years and 182 (15%) 70-75 years; 693 (58%) were males and 494 (42%) females. There was no evidence of interaction between age or gender and the benefit provided by the intensification of the upfront chemotherapy in terms of ORR and PFS, or the increased risk of experiencing G3/4 AEs. Elderly patients and females experienced higher rates of overall G3/4 AEs (73% versus 60%, P < 0.01 and 69% versus 57%, P < 0.01, respectively). Notably, in the FOLFOXIRI/bevacizumab subgroup, G3/4 diarrhea and febrile neutropenia occurred in 27% and 16% of elderly patients, respectively, while females reported high incidences of any grade nausea (67%) and vomiting (50%).

CONCLUSIONS

The improvements in terms of ORR and PFS of FOLFOXIRI/bevacizumab versus doublets/bevacizumab are independent of gender and age, with a similar relative increase in AEs among elderly patients and females. Initial dose reductions and possibly primary G-CSF prophylaxis should be recommended for patients between 70 and 75 years old treated with FOLFOXIRI/bevacizumab, and a careful management of antiemetic prophylaxis should be considered among females.

Epirubicin and ifosfamide in advanced soft tissue sarcomas

BACKGROUND

To evaluate the feasibility, toxicity and efficacy of the combination of (IFO) ifosfamide and epirubicin (EPI) given at conventional doses for monochemotherapy, we started a phase II study in advanced/metastatic soft tissue sarcoma patients.

PATIENTS AND METHODS

Treatment consisted of: epirubicin 75 mg/m2 i.v. day 1; IFO 1.8 g/m2 days 1 to 5; MESNA 20% of the IFO dose at 4-hour intervals three times a day during IFO administration. Cycles were given every 3-4 weeks for at least three cycles.

RESULTS

The overall response rate for non-visceral sarcomas (51 pts) was 31% (95% confidence limits +/- 13%). Among the 13 visceral sarcomas no response was seen for the leiomyosarcomas of the gastrointestinal tract, whereas one complete and one partial remission were observed for the uterine sarcomas. The duration of response was 10 months (range 5-34+) for complete responses and 9 (range 4-42+) for partial responses. The median survival for responders is 18 months (range 2-60+) and for non-responders 10 months (range 1-33) (p < 0.004).

CONCLUSIONS

This combination proved to be feasible and tolerable. The overall response rate does not appear to be superior to those with other standard treatments, but it should be pointed out that our patient population was totally unselected.