Long-term efficacy and safety of addition of carboplatin with or without veliparib to standard neoadjuvant chemotherapy in triple-negative breast cancer: 4-year follow-up data from BrighTNess, a randomized phase III trial.

BACKGROUND: Primary analyses of the phase III BrighTNess trial showed addition of carboplatin with/without veliparib to neoadjuvant chemotherapy significantly improved pathological complete response (pCR) rates with manageable acute toxicity in patients with triple-negative breast cancer (TNBC). Here, we report 4.5-year follow-up data from the trial. PATIENTS AND METHODS: Women with untreated stage II-III TNBC were randomized (2 : 1 : 1) to paclitaxel (weekly for 12 doses) plus: (i) carboplatin (every 3 weeks for four cycles) plus veliparib (twice daily); (ii) carboplatin plus veliparib placebo; or (iii) carboplatin placebo plus veliparib placebo. All patients then received doxorubicin and cyclophosphamide every 2-3 weeks for four cycles. The primary endpoint was pCR. Secondary endpoints included event-free survival (EFS), overall survival (OS), and safety. Since the co-primary endpoint of increased pCR with carboplatin plus veliparib with paclitaxel versus carboplatin with paclitaxel was not met, secondary analyses are descriptive. RESULTS: Of 634 patients, 316 were randomized to carboplatin plus veliparib with paclitaxel, 160 to carboplatin with paclitaxel, and 158 to paclitaxel. With median follow-up of 4.5 years, the hazard ratio for EFS for carboplatin plus veliparib with paclitaxel versus paclitaxel was 0.63 [95% confidence interval (CI) 0.43-0.92, P = 0.02], but 1.12 (95% CI 0.72-1.72, P = 0.62) for carboplatin plus veliparib with paclitaxel versus carboplatin with paclitaxel. In post hoc analysis, the hazard ratio for EFS was 0.57 (95% CI 0.36-0.91, P = 0.02) for carboplatin with paclitaxel versus paclitaxel. OS did not differ significantly between treatment arms, nor did rates of myelodysplastic syndromes, acute myeloid leukemia, or other secondary malignancies. CONCLUSIONS: Improvement in pCR with the addition of carboplatin was associated with long-term EFS benefit with a manageable safety profile, and without increasing the risk of second malignancies, whereas adding veliparib did not impact EFS. These findings support the addition of carboplatin to weekly paclitaxel followed by doxorubicin and cyclophosphamide neoadjuvant chemotherapy for early-stage TNBC.

Neoadjuvant weekly paclitaxel and carboplatin with trastuzumab and pertuzumab in HER2-positive breast cancer: a Brown University Oncology Research Group (BrUOG) study.

PURPOSE: In HER2-positive breast cancer (HER2+ BC), neoadjuvant chemotherapy (NACT) with dual HER2-targeted therapy achieves high pathologic complete response (pCR) rates. Anthracycline-free NACT regimens avoid toxicities associated with anthracyclines, but every 3-week TCHP also has substantial side effects. We hypothesized that a weekly regimen might have equivalent efficacy with less toxicity; we also investigated whether poorly responding patients would benefit from switching to AC. METHODS: Patients with clinical stage II-III HER2+ BC received weekly paclitaxel 80 mg/m2 and carboplatin AUC2 with every 3-week trastuzumab and pertuzumab (wPCbTP), with the option of splitting the pertuzumab loading dose. After 12 weeks, responding patients continued wPCbTP for another 6 weeks, while non-responders switched to AC. Dose modifications and post-op therapy were at investigator discretion. RESULTS: In 30 evaluable patients, the pCR rate was 77% (95% CI 58-90%); 12/14 (86%) in ER-negative and 11/16 (69%) in ER-positive. Only two patients transitioned to AC for non-response, of which one achieved pCR. There were no episodes of febrile neutropenia or grade ≥ 3 peripheral neuropathy, though several patients who continued wPCbTP stopped before week 18. Split-dose pertuzumab was associated with less grade ≥ 2 diarrhea (40%) than the standard loading dose (60%). CONCLUSION: pCR rates with our regimen were as high as reported with TCHP with fewer grade ≥ 3 toxicities, though diarrhea remains a concern. Too few patients had a suboptimal response to adequately test switching to AC. The wPCbTP regimen should be considered an alternative to TCHP as neoadjuvant therapy for HER2+ BC. TRAIL REGISTRATION: ClinicalTrials.gov identifier: NCT02789657.

SWOG S0800 (NCI CDR0000636131): addition of bevacizumab to neoadjuvant nab-paclitaxel with dose-dense doxorubicin and cyclophosphamide improves pathologic complete response (pCR) rates in inflammatory or locally advanced breast cancer.

SWOG S0800, a randomized open-label Phase II clinical trial, compared the combination of weekly nab-paclitaxel and bevacizumab followed by dose-dense doxorubicin and cyclophosphamide (AC) with nab-paclitaxel followed or preceded by AC as neoadjuvant treatment for HER2-negative locally advanced breast cancer (LABC) or inflammatory breast cancer (IBC). Patients were randomly allocated (2:1:1) to three neoadjuvant chemotherapy arms: (1) nab-paclitaxel with concurrent bevacizumab followed by AC; (2) nab-paclitaxel followed by AC; or (3) AC followed by nab-paclitaxel. The primary endpoint was pathologic complete response (pCR) with stratification by disease type (non-IBC LABC vs. IBC) and hormone receptor status (positive vs. negative). Overall survival (OS), event-free survival (EFS), and toxicity were secondary endpoints. Analyses were intent-to-treat comparing bevacizumab to the combined control arms. A total of 215 patients were accrued including 11 % with IBC and 32 % with triple-negative breast cancer (TNBC). The addition of bevacizumab significantly increased the pCR rate overall (36 vs. 21 %; p = 0.019) and in TNBC (59 vs. 29 %; p = 0.014), but not in hormone receptor-positive disease (24 vs. 18 %; p = 0.41). Sequence of administration of nab-paclitaxel and AC did not affect the pCR rate. While no significant differences in OS or EFS were seen, a trend favored the addition of bevacizumab for EFS (p = 0.06) in TNBC. Overall, Grade 3-4 adverse events did not differ substantially by treatment arm. The addition of bevacizumab to nab-paclitaxel prior to dose-dense AC neoadjuvant chemotherapy significantly improved the pCR rate compared to chemotherapy alone in patients with triple-negative LABC/IBC and was accompanied by a trend for improved EFS. This suggests reconsideration of the role of bevacizumab in high-risk triple-negative locally advanced breast cancer.

Zoledronic acid preserves bone mineral density in premenopausal women who develop ovarian failure due to adjuvant chemotherapy: final results from CALGB trial 79809.

BACKGROUND: Chemotherapy-induced ovarian failure (CIOF) is a frequent side-effect of adjuvant chemotherapy that results in rapid bone loss. We hypothesised that zoledronic acid (ZA), a third-generation amino bisphosphonate, would prevent bone loss in premenopausal women who developed CIOF. METHODS: Women (439) were randomised to intravenous (i.v.) ZA 4 mg every 3 months for 2 years starting within 1-3 months after randomization (arm A) or 1 year after randomization (arm B, controls). CIOF was prospectively defined as ≥ 3 months of amenorrhoea, follicle-stimulating hormone (FSH) ≥ 30 MIU/ml and non-pregnant at 1 year. The primary end-point was the percentage change in bone mineral density (BMD) in the lumbar spine (LS) from baseline to 12 months in the ZA and in control groups in women who developed CIOF; the secondary end-point was BMD in LS at 3 years in all randomised women. FINDINGS: One hundred and fifty (56%) met the definition of CIOF at 1 year. Overall, grade 3 toxicities of ZA were fatigue (1%) arthralgias (21%) and pain (84%). The median percent change (interquartile range, IQR) at 1 year was +1.2% (-0.5% to +2.8%) and -6.7% (-9.7% to -2.9%) p<0.001 and at 3 years was +1.0% (-1.6% to +5.2%) and -0.5% (-3.7% to +3.2%) p=0.019 in arms A and B, respectively. INTERPRETATION: ZA every 3 months is well tolerated and prevents rapid bone loss in premenopausal women that develop CIOF. Giving ZA with rather than 1 year after the start of adjuvant chemotherapy is the preferred sequence to prevent bone loss.

Splenosis presenting as occult gastrointestinal bleeding.

A 48-year-old man presented with recurrent gastrointestinal bleeding and anemia. Routine endoscopic evaluation was nondiagnostic. Angiography demonstrated multiple apparent arteriovenous malformations. Exploratory laparotomy revealed numerous splenic implants along the small and large bowels, some of which had apparently eroded through the bowel mucosa and bled. Excision of these penetrating lesions prevented further bleeding. An incidentally noted renal cell cancer was also resected. The patient's splenosis was the result of childhood trauma that caused splenic rupture and precipitated splenectomy. Splenosis develops frequently following traumatic splenic rupture. Experimental evidence suggests that the presence of an intact spleen suppresses the growth and development of splenic implants. Following splenectomy, splenules may replace some of the "housekeeping" and immunologic functions of the spleen, but even patients with documented splenosis should be considered functionally hyposplenic. While in most cases splenules cause no symptoms, splenosis must be considered in the differential diagnosis of previously splenectomized patients who present with unexplained masses or occult bleeding.

Locally advanced breast cancer.

Over the past 20 years, the prognosis for women diagnosed with locally advanced breast cancer (LABC; clinical stages IIB through IIIB) has improved significantly with recognition of the efficacy of multimodal therapy for reducing both local and distant recurrences, even in patients with inflammatory breast cancer (IBC). Most patients will respond to induction, or neoadjuvant, chemotherapy (NAC) with an anthracycline-based regimen, enabling many patients with large but operable tumors to undergo breast-conserving surgery (BCS) and enabling resection in most patients with inoperable disease. However, only a small percentage of patients achieve a pathologic complete response (CR) with this approach. Long-term disease-free survival (DFS) and overall survival (OS) correlate with the extent of residual disease in the breast and axillary nodes following NAC. The addition of paclitaxel or docetaxel, either in combination with an anthracycline or as a separate regimen administered before or after anthracycline-based therapy, increases clinical and pathologic response rates and may improve DFS. With the possible exception of patients with IBC, BCS does not compromise outcome. Partial mastectomy should be accompanied by standard nodal dissection in patients with clinically or radiographically positive axillae; in patients with negative axillae, sentinel lymph node (SLN) sampling, with subsequent axillary dissection reserved for patients with involved nodes, may reduce postoperative morbidity. Patients who received only anthracycline-based NAC who are found to have significant residual disease in the breast or involved axillary nodes at surgery should receive adjuvant chemotherapy with paclitaxel. Postoperative radiation to the residual breast or chest wall and regional nodal areas reduces locoregional recurrences, but its impact on OS remains controversial. Adjuvant hormonal therapy with tamoxifen improves DFS and OS in patients with hormone receptor (HR)-positive tumors, and ovarian ablation should be considered in premenopausal patients with HR-positive tumors and multiple involved nodes or stage IIIB disease. Neoadjuvant hormonal therapy with either tamoxifen or an aromatase inhibitor may benefit frail or elderly patients with HR-positive tumors for whom chemotherapy is not an option. No advantage has been demonstrated for high-dose chemotherapy requiring hematopoietic stem-cell support as either NAC or adjuvant therapy in LABC. Newer treatment approaches, including trastuzumab (Herceptin, Genentech, Inc., South San Francisco, CA), in patients with Her-2-overexpressing tumors or other biologic agents, do not have a proven role in the management of LABC at this time.

Multicentric warfarin-induced skin necrosis complicating heparin-induced thrombocytopenia.

Two patients developed catastrophic multicentric skin necrosis while receiving warfarin to treat venous thromboembolism complicated by immune-mediated heparin-induced thrombocytopenia (HIT). Patient 1 developed skin necrosis involving the breasts, thighs, and face, as well as venous limb gangrene and bilateral hemorrhagic necrosis of the adrenal glands, resulting in death. The second patient developed bilateral mammary necrosis necessitating mastectomies, as well as skin necrosis involving the thigh. Neither patient had an identifiable hypercoagulable syndrome, other than HIT. HIT may represent a risk factor for the development of multicentric warfarin-induced skin necrosis (WISN).

Spontaneous venous thrombosis in a young patient with combined factor V Leiden and lupus anticoagulant.

We describe a case of a 28-year-old man who developed an extensive spontaneous deep venous thrombosis. Testing revealed heterozygotic factor V Leiden mutation, and the presence of both lupus anticoagulant (LA) and elevated IgM anticardiolipin antibody (ACA). Several family members were found to be heterozygous for factor V Leiden. A paternal aunt had the factor V Leiden mutation, an elevated plasma homocysteine and a borderline increased IgG ACA level. No other family member had a history of a venous thrombotic event. This case illustrates that evaluation of young patients who present with venous thrombosis should be performed for both hereditary and acquired thrombophilic defects. The family studies suggest that the presence of a lupus anticoagulant may be more clinically significant than elevated ACA in risk assessment. Although screening family members when the proband carries factor V Leiden is controversial, psychological reassurance of those who test negative and simple advice on occupations or social habits (e.g., smoking) for those who test positive may be important benefits.

A multimodal approach in the diagnosis of patients with hematopoietic disorders.

Myelodysplastic syndromes (MDS) are a group of relatively ill-defined hematopoietic disorders in which both qualitative and quantitative defects of the hematopoietic cells cause bone marrow dysfunction. With an incidence estimated to be approximately 1 per 100,000 persons per year, MDS mainly affects the elderly. Myelodysplastic syndromes share many features with acute nonlymphocytic leukemia; in fact, a proportion of patients with MDS eventually develop acute myeloid leukemia. To illustrate a multimodal approach in the diagnosis of patients with hematopoietic disorders, we describe a 66-year-old patient with a question of myelodysplastic syndrome, leukemia, and two translocations involving chromosome 10:t(5;10) and t(7;10). These structural rearrangements effectively gave rise to monosomy for part of the long arm of chromosome 5 and for the long arm of chromosome 7. Findings of del(5q) and del(7) in MDS have been reported in the literature. The results of chromosome morphometry, which was conducted to compare the lengths of all relevant chromosome segments, are consistent with the hypothesized chromosomal abnormalities. The investigational technique of fluorescence in situ hybridization (FISH), using both painting and alpha-satellite probes, was used as an adjunct to conventional cytogenetics to further delineate the nature of the chromosome abnormalities observed in the GTG-banded studies. Confirmatory studies utilizing the new technique of spectral karyotyping (SKY) were also carried out. Thus, the multimodal approach of hematopathology, GTG-banding, chromosome morphometry, FISH, and SKY can be very useful for delineating complex cytogenetic cases.

Fluorescent in situ hybridization for identifying cytogenetic abnormalities in inadequate and suboptimal specimens.

Conventional cytogenetic analysis is often hampered by its dependency on the evaluation of dividing cells. When the mitotic index is low, or the cytogenetic preparation suboptimal, an accurate diagnosis often cannot be achieved using standard GTG banding. Molecular cytogenetics using fluorescent in situ hybridization (FISH) can be extremely useful in such a situation because it can be performed on nondividing interphase nuclei, and can usually be carried out on existing slides without requiring an additional specimen. The present paper presents two case histories to illustrate the utility of FISH in the clinical setting.

Weekly high-dose paclitaxel in metastatic and locally advanced breast cancer: a preliminary report.

The optimal dose and schedule for paclitaxel (Taxol; Bristol-Myers Squibb Company, Princeton, NJ) in the treatment of patients with advanced breast cancer are not known. Based on our phase I study in non-small cell lung cancer, in which the dose intensity of paclitaxel was successfully escalated by using a weekly schedule, we initiated a phase II study of weekly paclitaxel in previously untreated patients with metastatic breast cancer (MBC) and locally advanced breast cancer (LABC). Treatment consists of weekly paclitaxel 175 mg/m2 intravenously over 3 hours for 6 weeks, followed by a 2-week break. Doses are modified for neutropenia (absolute neutrophil count < 1,500/microL), bilirubin levels greater than 1.5 times normal, or greater than grade 1 neuropathy. Patients with MBC continue treatment until disease progression. Patients with LABC receive one to two cycles before proceeding to surgery if resectable. Thus far, 15 patients, eight with MBC and seven with LABC, are assessable for response and/or toxicity. Most patients have required dose modification, with median delivery of 75% (cycle 1) and 50% (cycle 2) of the planned dose of paclitaxel. Neutropenia has been the most common cause of dose reductions, although only one patient required treatment for neutropenic fever. Six patients have developed grade 2/3 peripheral sensory neuropathy, but with dose reductions many have continued treatment with stable or improving neurologic symptoms. Objective responses have been seen in 12 of 14 assessable patients, including six with MBC (one complete response, five partial responses) and six with LABC (two complete responses, four partial responses), for an overall response rate of 86% (95% confidence interval, 66% to 96%). All responding LABC patients have been rendered free from disease at surgery. These preliminary results are very encouraging. Accrual to the study continues.

Weekly paclitaxel in patients with advanced lung cancer: preliminary data from a phase II trial.

We conducted a phase II trial in chemotherapy-naive patients with advanced non-small cell lung cancer to determine the efficacy of paclitaxel (Taxol; Bristol-Myers Squibb Company, Princeton, NJ) delivered at a maximum tolerated dose of 175 mg/m2 on an extended weekly schedule. Patients with stage IIIB/IV non-small cell lung cancer were eligible if they had an Eastern Cooperative Oncology Group performance status of 0 to 2, had received no previous chemotherapy, demonstrated normal hematologic and hepatic function, and could provide informed consent. Paclitaxel 175 mg/m2 was administered as an intravenous infusion weekly over 3 hours with standard premedication, for 6 weeks of an 8-week cycle. Doses were modified for absolute neutrophil count less than 1.5 x 10(9)/L or neuropathy greater than grade I on the day of therapy. Patients without progressive disease received subsequent cycles at the same dose. To date, 30 patients have been enrolled; data are available for 25. The median age was 65 years (range, 38 to 78 years), 23 patients were performance status 0 or 1, and 14 had received prior radiation. Sites of disease included the lung (23 patients), central nervous system (11), bone (seven), liver (one), kidney (one), and soft tissue (eight). Eighty-three percent, 75%, 58%, and 50% of intended doses were delivered during cycles 1 though 4, respectively. Grade 2/3 neuropathy occurred in nine patients, but improved in all nine following dose reduction. Grade 3/4 neutropenia occurred in 10 patients. Partial responses occurred in 14 of 25 patients (56%; 95% confidence interval, 46% to 66%). Median duration of response was 6.5 months, and the 1-year survival rate was 53%. The extended weekly paclitaxel schedule results in enhanced dose intensity, marked activity, and acceptable toxicity. Paclitaxel given weekly at maximum dose intensity may be more effective than conventional paclitaxel administration schedules.

Weekly paclitaxel as a radiation sensitizer for locally advanced gastric and pancreatic cancers: the Brown University Oncology Group experience.

Many patients with cancer of the stomach or pancreas have locally advanced, unresectable disease at diagnosis or will develop an early local or regional recurrence despite potentially curative surgery. Effective local treatment could increase the proportion of patients able to undergo surgery and decrease locoregional recurrences, which should improve overall survival. External beam radiation (RT) by itself has little effect. Standard treatment, such as RT with concurrent administration of 5-fluorouracil-based chemotherapy as a radiation sensitizer, has, at best, a modest impact on locoregional recurrences and survival. The use of a more effective radiosensitizer might improve the efficacy of local treatment. Paclitaxel synchronizes cells at G2M, the phase of the cell cycle during which cells are most sensitive to the effects of ionizing radiation, and has been demonstrated to sensitize a variety of human cell lines to the effects of RT. In patients with locally advanced non-small cell lung cancer (NSCLC), the Brown University Oncology Group (BrUOG) has demonstrated a high response rate to low-dose weekly paclitaxel with concurrent RT. In addition, we demonstrated that the response to paclitaxel/RT was not affected by mutations in the p53 tumor suppressor gene. This suggested that paclitaxel/RT would be a rational treatment approach for other malignancies with a high frequency of p53 mutations, such as gastric and pancreatic cancers. We have completed a phase I study of weekly paclitaxel and concurrent radiation for locally advanced gastric and pancreatic cancers. The maximum tolerated dose of paclitaxel was 50mg/m2/week for six weeks with 50 Gray (Gy) abdominal radiation. The dose limiting toxicities were abdominal pain, nausea and anorexia. Preliminary response data from ongoing phase II studies suggest that preoperative paclitaxel/RT has substantial activity in patients with locally advanced gastric and pancreatic cancers, though whether this will translate into improved disease-free and overall survival in these patients is not known.

Paclitaxel and concurrent radiation for locally advanced pancreatic and gastric cancer: a phase I study.

PURPOSE: To determine the maximum-tolerated dose (MTD), dose-limiting toxicities, and potential antitumor activity of weekly paclitaxel with concurrent radiation (RT) for locally advanced pancreatic and gastric cancer. PATIENTS AND METHODS: Thirty-four patients with locally advanced adenocarcinoma of the pancreas or stomach were studied. The initial dose of paclitaxel was 30 mg/m2 by 3-hour intravenous (I.V.) infusion repeated every week for 6 weeks with 50 Gy RT. Doses were escalated at 10-mg/m2 increments in successive cohorts of three new patients until dose-limiting toxicity was observed. RESULTS: The dose-limiting toxicities at 60 mg/m2/wk were abdominal pain within the RT field, nausea, and anorexia. Of 23 patients with assessable disease, 11 (seven with gastric, four with pancreatic cancer) had objective responses for an overall response rate of 48%. CONCLUSION: Concurrent paclitaxel with upper abdominal RT is well tolerated at dosages that have substantial activity. A phase II trial of neoadjuvant paclitaxel and RT at the MTD of 50 mg/m2/wk is underway.

Phase II study of 5-fluoruracil leucovorin and azidothymidine in patients with metastatic colorectal cancer.

The primary objective of this study was to determine the response rate of patients with metastatic colorectal cancer to combined therapy with 5-fluorouracil (5-FU), leucovorin, and intravenous azidothymidine (AZT), a thymidine nucleoside analog. By itself, AZT has limited antineoplastic efficacy. However, experimental studies indicate that 5-FU enhances the antitumor activity of AZT by inhibiting synthesis of normal thymidine nucleotides with which AZT competes for incorporation into nucleic acids. A phase I study defined the maximum tolerated dose of AZT as 7 g/m2 with hypotension during the infusion being the dose-limiting toxicity. A phase II study was performed with oral leucovorin (100 mg p.o. hourly for 4 h prior to 5-FU and 4 h and 8 h after 5-FU), bolus 5-FU (400 mg/m2) followed 1 h later by a 2-h infusion of AZT (7 g/m2). Treatment was given weekly for 4 weeks followed by a 1-week break, which constituted a cycle of therapy. Responses were evaluated after every two cycles. Patients continued on therapy as long as they tolerated treatment and did not have progressive disease. Of 15 evaluable patients who had received no chemotherapy there was 1 complete response and 4 partial responses (a 33% response rate), whereas only 1 of 6 patients who had received prior adjuvant chemotherapy had a partial response (17%). An additional 10 patients had stable disease lasting 2-14 months. Therapy was well tolerated with the only one instance each of grade 3 nausea and vomiting, diarrhea, anemia, and hypotension. Approximately 50% of treatments were accompanied by mild hypotension, which was easily corrected by increasing the rate of normal saline infusion. There was no difficulty administering this regimen in the outpatient setting. While the overall response rate (29%) is comparable to that seen with combinations of 5-FU and leucovorin alone, in most reported series a considerably higher dose of 5-FU was utilized than in this study. Since patients in the present study experienced relatively little 5-FU toxicity, increasing the dose of 5-FU in this regimen would appear to be feasible and might result in a higher response rate.

Fluorescent in situ hybridization for assessing the proportion of cells with trisomy 4 in a patient with acute non-lymphoblastic leukemia.

A case of acute non-lymphoblastic leukemia is described in which fluorescent in situ hybridization (FISH) helped to resolve initially conflicting conventional cytogenetic results. Identification and assessment of the proportion of cells exhibiting trisomy of chromosome 4 in the patient's bone marrow were made using a probe which hybridizes to the centromeric region of chromosome 4. These FISH results were consistent with our retrospective GTG-banded analysis but differed from another study conducted elsewhere. Factors such as culture conditions and duration of culture which may have influenced the proportion of leukemic to nonleukemic cells are discussed. Fluorescent in situ hybridization is a powerful adjunct to conventional cytogenetic analysis and may prove to be a useful tool for monitoring minimal residual disease in this patient.

A phase II trial of continuous infusion cisplatin and 5-fluorouracil with oral calcium leucovorin in colorectal carcinoma.

Twenty previously untreated patients with advanced colorectal adenocarcinoma were entered on a Phase II trial of 3-day continuous infusion cisplatin (25 mg/m2/day) and 5-fluorouracil (800 mg/M2/day) with oral calcium leucovorin (30 mg/dose) every 6 hours. There were four partial responses (20%) and two complete responses (10%) for a total response rate of 30% (95% confidence limits +/- 20%). Patients received a median of 4.5 cycles of therapy (range 2-9 cycles). Three patients experienced neutropenia; one had a life-threatening infection. One developed neuropathy at 375 mg/M2 cumulative dose. Four patients developed mucositis. Treatment was stopped for one patient with stable disease after 5 cycles because of anorexia and nausea and vomiting; treatment was stopped for four patients because of excessive fatigue. The median duration of responses was 4 months (range 3-6 months). Although this regimen is active, the response rate, cumulative nature of the toxicity, and the requirement for hospitalization led us to conclude that this regimen does not warrant Phase III testing but might be a basis for further Phase II therapeutic trials.