A retrospective comparison of neoadjuvant chemoradiotherapy regimens for locally advanced esophageal cancer

Preoperative chemoradiotherapy (CRT) with carboplatin/paclitaxel has been shown to increase survival in patients with esophageal cancer, including gastroesophageal junction (GE) junction cancer, over surgery alone; however, there have been no studies comparing the different neoadjuvant CRT regimens. We retrospectively evaluated the long-term results of trimodality therapy for patients with locally advanced esophageal cancer treated on several chemotherapy regimens. Between 1999 and 2014, 215 patients with locally advanced esophageal cancer underwent neoadjuvant CRT followed by surgical resection. The median age was 62 years (range 21-84), 80.5% were men and 86% had adenocarcinoma. The following chemotherapy regimens were administered: cisplatin/5FU (14.9%), cisplatin/irinotecan (35.8%), carboplatin/paclitaxel (35.8%), and other (9.7%). The majority of patients (92.1%) received a radiation dose of 50.4 Gy. Predictors of toxicities and surgical complications were assessed using logistic regression. Overall survival (OS) and recurrence-free survival (RFS) were estimated using the Kaplan-Meier method and proportional hazards regression was used to model time-to-event outcomes. The median follow-up among surviving patients was 4.1 years (range 0.4,13). The median OS was 3.0 years from time of diagnosis and OS was 36.8% at 5 years. RFS was 34.9% at 5 years. After neoadjuvant CRT, 34.7% of patients achieved a pathologic complete response including 60.7% of squamous cell carcinoma patients and 18.4% of adenocarcinoma patients (P < 0.001) and 66% were downstaged. Of the variables examined, pathologic stage, preoperative baseline cardiac comorbidity, postoperative cardiac or pulmonary complications, and chemotherapy regimen were associated with OS. Using cisplatin and 5FU as the reference regimen, patients treated with carboplatin/paclitaxel had significantly improved OS (HR = 0.47, P = 0.017 after adjusting for surgery type, radiation modality, baseline cardiac comorbidity, and preoperative stage) with 5-year OS rate of 66%. The most common surgical complications were cardiac in 61 patients (28.5%) and pulmonary in 52 patients (24.3%). Cardiac complications were associated with age (OR 1.05, P = 0.007) and cardiac comorbidity (OR 2.6, P = 0.02) and pulmonary complications with female gender (OR 3.98, P < 0.001). Forty-four patients (20.5%) required readmission within 30 days of discharge, and readmission was associated with cardiac comorbidity (OR 2.7, P = 0.017). Three patients died within 30 days of surgery. We observed an association between neoadjuvant carboplatin/paclitaxel and improved overall survival that requires confirmation in a prospective randomized trial.

Gastric adenocarcinoma treated with radiation with or without epirubicin-based chemotherapy: Evaluation of radiation-induced liver disease

110

Background: Anthracycline chemotherapy has been associated with radiation-induced liver disease (RILD). We sought to compare the incidence of liver toxicity among patients with gastric adenocarcinoma treated with radiotherapy (RT) with or without epirubicin-based chemotherapy.

Methods: We performed a retrospective analysis of 94 patients with gastric adenocarcinoma treated at Massachusetts General Hospital since November 2005 and the Dana-Farber Cancer Institute since August 1998. All patients underwent definitive surgery and RT (median dose 45 Gray) with a minimum follow up of 6 months. Primary endpoints were development of ascites, radiographic liver change and elevations in liver function tests (LFTs), including alkaline phosphatase (ALP), aspartate transaminase (AST) and alanine transaminase (ALT).

Results: In total, 34 patients received epirubicin-based chemotherapy including 9 perioperatively (6 with oxaliplatin and capecitabine [EOX]; 2 with cisplatin and 5-flourouracil [ECF]; 1 with oxaliplatin and 5-flourouracil [EOF]) and 25 postoperatively (2 EOX; 22 ECF; 1 combination). Seven patients were treated with neoadjuvant RT; 87 received adjuvant RT a median of 88 days after surgery (interquartile range 73-108 days). Twenty-one patients developed ascites within 6 months of completing RT, all but one of whom developed peritoneal carcinomatosis or metastatic disease. Among 57 patients that did not develop metastases, maximum elevations in LFTs were similar in patients that received epirubicin-based chemotherapy compared to those who did not (ALP/AST/ALT 150/44/50 vs. 142/41/44, p=0.25/0.36/0.14, respectively), as were rates of radiographic liver change (22% vs. 13%, p=0.44).

Conclusions: Epirubicin-based chemotherapy does not significantly increase the risk of RILD in a recent cohort of patients treated with modern RT techniques and dose-constraints. In this setting, treatment of gastric adenocarcinoma with RT and either pre- or postoperative chemotherapy is well tolerated with low rates of liver toxicities. Development of liver toxicity, particularly ascites, within six months of RT may be a harbinger of metastatic disease.

No significant financial relationships to disclose.

A multicenter phase II trial of single-agent cetuximab in advanced esophageal and gastric adenocarcinoma

BACKGROUND

Epidermal growth factor receptor (EGFR) is overexpressed in a significant proportion of esophageal and gastric carcinomas. Although previous studies have examined tyrosine kinase inhibitors of EGFR, there remains limited data regarding the role of EGFR-directed monoclonal antibody therapy in these malignancies. We carried out a multi-institutional phase II study of cetuximab, a monoclonal antibody against EGFR, in patients with unresectable or metastatic esophageal or gastric adenocarcinoma.

PATIENTS AND METHODS

Thirty-five patients with previously treated metastatic esophageal or gastric adenocarcinoma were treated with weekly cetuximab, at an initial dose of 400 mg/m(2) followed by weekly infusions at 250 mg/m(2). Patients were followed for toxicity, treatment response, and survival.

RESULTS

Treatment with cetuximab was well tolerated; no patients were taken off study due to drug-related adverse events. One (3%) partial treatment response was noted. Two (6%) patients had stable disease after 2 months of treatment. Median progression-free survival and overall survival were 1.6 and 3.1 months, respectively.

CONCLUSION

Although well tolerated, cetuximab administered as a single agent had minimal clinical activity in patients with metastatic esophageal and gastric adenocarcinoma. Ongoing studies of EGFR inhibitors in combination with other agents may define a role for these agents in the treatment of esophageal and gastric cancer.

Phase II and fluorodeoxyglucose positron emission tomography (FDG-PET) study in patients with advanced biliary tract cancers (BTCs) receiving bevacizumab (B) in combination with gemcitabine (GEM) and oxaliplatin (OX)

4578

Background: Vascular endothelial growth factor (VEGF) expression is present in BTCs and is associated with poor survival. We performed a phase II study to examine the efficacy and tolerability of GEM and OX in combination with B (GEMOX-B) in patients (pts) with advanced BTCs. We also assessed the use of FDG-PET as an early indicator of response following treatment. Methods: Eligibility criteria included unresectable or metastatic measurable BTCs, 0–1 prior chemotherapy regimens, performance status ≤ 2, and adequate organ functions. Pts were treated with all 3 drugs intravenously on days 1 and 15 every 28 days (one cycle): B was given first at 10 mg/kg, followed by GEM at 1000 mg/m2 as a dose-rate infusion at 10 mg/m2/minute, and OX at 85 mg/m2. Whole body FDG-PET scan was obtained at baseline and at the end of cycle 2. The primary endpoint of the study was progression-free survival (PFS). Results: The planned 35 pts (25 cholangiocarcinoma and 10 gallbladder carcinoma) were enrolled: median age = 60 (25–82), M/F = 21/14, median ECOG 1. All pts were evaluable for toxicity and 29 were evaluable for response. Treatment related grade 3–4 toxicities included neutropenia (20%), ALT (17%), neuropathy (14%), hypertension (11%), AST (9%), anorexia (9%), and thrombocytopenia (9%). 13 pts (45%) had a partial response (PR) and an additional 10 pts (34%) had stable disease (SD). With a median follow up of 9.9 months, the median overall survival was 13.2 months (95% CI, 7.3 to 20.5 months), and the median PFS was 7.0 months (95% CI, 5.4 to 9.6 months). The mean baseline SUVmax was 5.72±2.01 and post-treatment SUVmax was 3.73±1.88 with a median 36.4% decrease (n=32). The difference in SUV changes between the groups with PR/SD and progressive disease was statistically significant (p=0.006). An increase in adjusted post- to pre-treatment SUV increased the risk for tumor progression (hazard ratio=3.054). Conclusions: GEMOX-B demonstrated significant antitumor activity with tolerable safety profiles in patients with advanced BTCs. FDG-PET showed significant early decreases in SUVmax following treatment, and these changes correlated with tumor response and time to tumor progression.

[Table: see text]

Weekly docetaxel, cisplatin, and irinotecan (TPC): results of a multicenter phase II trial in patients with metastatic esophagogastric cancer

BACKGROUND

Recent studies have examined the addition of docetaxel to fluorouracil and cisplatin in advanced esophagogastric cancer.

PATIENTS AND METHODS

We carried out a phase I dose-escalation study of weekly docetaxel, cisplatin, and irinotecan (TPC), given on days 1 and 8 every 3 weeks, in patients with chemonaive solid tumors. Subsequently, we completed a multiinstitutional phase II study of TPC in patients with previously untreated, metastatic esophagogastric cancer.

RESULTS

Thirty-nine patients were enrolled in the phase I trial; a weekly schedule of TPC was well tolerated. On that basis, docetaxel 30 mg/m(2), cisplatin 25 mg/m(2), and irinotecan 65 mg/m(2) were selected for the phase II trial, where in the first 18 patients irinotecan 65 mg/m(2) caused too much diarrhea and was reduced to 50 mg/m(2). Among 56 eligible patients with previously untreated, metastatic esophagogastric cancer enrolled in the phase II trial, three complete and 27 partial responses were observed (overall response rate=54%), and 15 patients (30%) had stable disease. Median progression-free survival was 7.1 months, and median survival was 11.9 months. At the final irinotecan dose of 50 mg/m(2), grade 3 or higher toxicity included diarrhea (26%), neutropenia (21%), nausea (18%), fatigue (16%), anorexia (13%), and thrombosis/embolism (13%).

CONCLUSIONS

Weekly TPC is an active and well-tolerated regimen for patients with esophagogastric cancer.

Prediction of response to temozolomide (TMZ)-based therapy by loss of MGMT expression in patients with advanced neuroendocrine tumors (NET)

4505

Background: TMZ, an oral alkylating agent, has recently been shown to be active in a subset of patients with NETs. In other tumor types, methyl-guanine methyl transferase (MGMT) expression is associated with TMZ resistance through its ability to remove methyl/alkyl groups from the O6 -position of guanine, thereby preventing TMZ-induced DNA damage. Methods: To better define NET pts who benefit from TMZ, we performed a retrospective analysis of 76 NET pts treated with TMZ-based regimens at 3 institutions, identified using IRB-approved observational and clinical trial databases. Immunohistochemical MGMT expression was measured in a subset of pts with available paraffin-embedded archival tissue specimens in a blinded fashion, utilizing mouse monoclonal antibodies to MGMT. Nuclear MGMT expression was scored as either present or absent in tumor cells and correlated with tumor type and treatment outcome. Results: Of 76 pts, 30 received TMZ/thalidomide and 46 received TMZ/bevacizumab. Both regimens incorporate oral TMZ, administered at a dose of 150 mg/m2 qd for 7d, every other week. 63 pts were enrolled on phase II studies, and 13 pts were treated outside a formal study protocol. Pt characteristics were: M:F=42:34; median age 57.5 (20–75); tumor type carcinoid (38), pancreatic NET (35), paraganglioma/pheochromocytoma (3). PR or CR (as measured by RECIST) was observed in 11/35 (31%) pancreatic NETs and 0/38 carcinoid pts (P<0.001), as well as 1/3 pts with paraganglioma/pheo. In 21 pts with available specimens, complete absence of MGMT expression was observed in 5/8 pancreatic NET and 0/13 carcinoid tumors (P<0.01). In this pt subset, absence of MGMT expression was associated with response to TMZ ( Table 1 ). Conclusions: Treatment with TMZ is associated with radiologic response in approximately one third of pts with pancreatic NET, but not in pts with carcinoid tumors. Absence of MGMT expression appears to define pts with pancreatic NET who achieve significant benefit from TMZ.

[Table: see text]

No significant financial relationships to disclose.

Phase II cisplatin, irinotecan, cetuximab and concurrent radiation therapy followed by surgery for locally advanced esophageal cancer

4064

Background: Weekly irinotecan, cisplatin, and concurrent radiation therapy is a well-tolerated, active regimen in locally advanced esophageal cancer. (Ilson. JCO 2003) Cetuximab, an EGFR inhibitor, is a potent radiation sensitizer in head and neck cancer. (Bonner. Proc ASCO 2004) Methods: In this phase II trial, patients (pts) with T2–4N0–1M0–1A esophageal adenocarcinoma (A) or squamous cell carcinoma (S) receive 5040 cGy/28 fractions of radiation therapy (RT) and concurrent weekly cisplatin 30mg/m2 plus irinotecan 65 mg/m2 on weeks 1, 2, 4, and 5, followed by surgery 4–8 weeks after completion of RT. Additionally, pts receive weekly infusions of cetuximab 250 mg during RT, up to one week before surgery, and for 6 months following surgery. Results: Seventeen pts have been entered: male: female = 14:3, median age 54, ECOG PS 0:1 = 6:11, A:S = 17:0, stage IIA:IIB:III:IVA = 6:1:8:2, tumor location-esophagus-mid:lower:gastroesophageal junction = 1:4:12, >10% weight loss-yes:no = 8:9. Of 17 pts entered, 15 pts have proceeded to surgery, 1 pt died from Aspergillus infection resulting in respiratory failure and sepsis, and 1 pt is pending surgery. Of the 15 pts who underwent surgery, 2 (13%) had a complete pathologic response; pathologic stage for other pts: 0 = 1, I = 3, IIA = 3, IIB = 1, III = 4, IV = 1. Grade III/IV toxicity (17 pts) was: diarrhea 9 pts, neutropenia 9 pts, febrile neutropenia 5 pts, anorexia 5 pts, vomiting 4 pts, fatigue 3 pts, mucositis 1 pt. Chemotherapy dose attenuation was required for diarrhea in 5 pts, for neutropenia in 4 pts, and for folliculitis in 1 pt. One patient was removed from study during week 6 for prolonged diarrhea/ dehydration. Due to the 2-step design of the trial, accrual is on hold pending a 3rd required pathologic CR in the first 17 patients. Conclusions: Compared to other trials of irinotecan, cisplatin, radiation therapy, and surgery in similar groups of esophageal cancer patients, early results for this combination with cetuximab suggest a lower complete response rate and higher overall toxicity. Additional data will be available at ASCO. Supported by Bristol-Myers Squibb.

No significant financial relationships to disclose.

Phase II study of cetuximab in patients with unresectable or metastatic hepatocellular carcinoma

14096

Background: Patients with advanced hepatocellular carcinoma (HCC) have a poor prognosis with no established systemic treatment regimen. Epidermal growth factor receptor/human epidermal growth factor receptor 1 (EGFR/HER1) and ligand expression is frequently seen in HCC. Recent studies suggest that erlotinib, an EGFR/HER1 tyrosine kinase inhibitor, may have benefit in stabilizing HCC. We performed a phase II study with cetuximab, a chimeric monoclonal antibody that binds selectively to EGFR, in advanced HCC. Methods: Eligibility criteria include unresectable or metastatic measurable HCC, up to two prior systemic regimens, performance status ≤ 2, CLIP score ≤ 3, and adequate organ functions. The initial dose of cetuximab is 400 mg/m2 intravenously (IV) administered over 120 minutes, followed by weekly infusions at 250 mg/m2 IV over 60 minutes. Each cycle is defined as 6 consecutive weekly treatments. EGFR expression was assayed by immunohistochemistry and trough serum concentrations of cetuximab were determined during the first cycle. The primary endpoint of the study was progression free survival (PFS). Results: The planned 30 patients have been enrolled: median age = 58 (33–82), M/F = 23/7, ECOG 0/1/2 = 16/12/2, CLIP 0/1/2/3=2/9/14/5. No responses were seen. Five patients had stable disease. The median number of cycles on study per patient was 1 (range, 1–3). 16 patients developed progressive disease following one cycle of treatment. Only one patient remains on study. The median PFS and OS were 41 days (95% CI, 36 to 79) and 157 days (95% CI, 112 to not available), respectively. The treatment was generally well tolerated. No treatment related deaths occurred. Treatment related grade 1–2 toxicities included rash (83%), fatigue (47%), hypomagnesemia (27%), nausea (20%), anemia (13%), diarrhea (13%), anorexia (13%), and elevation of SGOT/SGPT (10%). Grade 3 SGOT, hypomagnesemia, and fever without neutropenia were seen in one patient (3%) each. Conclusions: Cetuximab has no activity in HCC in this phase II study. It can be safely given with tolerable toxicity profiles in HCC patients. Updated information on toxicity, efficacy, EGFR expression and pharmacokinetics will be presented at the meeting. Supported by Bristol-Myers Squibb.

[Table: see text]

A phase II study of temozolomide and bevacizumab in patients with advanced neuroendocrine tumors

4044

Background: Inhibitors of the VEGF pathway have been shown to have activity in neuroendocrine tumors (NETs). Temozolomide (TMZ), an oral analog of dacarbazine is also active in this setting. We performed a prospective, phase II study to assess the safety and efficacy of TMZ, administered in combination with bevacizuamb, in patients (pts) with advanced NETs. Methods: Pts received TMZ, 150 mg/m2/day po for 7 days every other week, and bevacizumab, 5 mg/kg IV every other week. Due to anticipated lymphopenia, pts received prophylaxis with trimethoprim/sulfamethoxazole (1 DS tablet q MWF) and acyclovir (400 mg po TID). Pts were followed for toxicity, response, and survival. Results: Enrolled patients (n=34) had the following characteristics: M:F = 19:15; median age 61 (range 37–75); ECOG PS 0/1/2 = 12/20/2; carcinoid/pancreatic NET = 16/18. Prior treatments included chemoembolization (n=7) chemotherapy (n=12); and octreotide (n=17); pts on octreotide remained on octreotide at stable doses for the duration of the study. Pts had either well-differentiated tumors (n=27) or moderately/poorly-differentiated NETs (n=7); pts with small cell carcinoma were not eligible for the study. Pts have received treatment for a median of 22 weeks. Grade 3–4 toxicities included: lymphopenia (n=21, 62%), leukopenia (n=2, 6%), thrombocytopenia (n=7, 21%), neutropenia (n=2, 6%), hyponatremia (n=1, 3%), vomiting (n=3, 9%), nausea (n=2, 6%), dehydration (n=1, 3%), fatigue (n=2, 6%), constipation (n=1, 3%), and hypertension (n=1, 3%). 20 pts had elevated CGA levels (>36.4 ng/ml) at baseline; 0/9 (0%) carcinoid and 4/11 (36%) pancreatic NET experienced CGA decreases of >50% from baseline on two consecutive assessments. 29 pts are currently evaluable for radiologic response ( Table ). Conclusions: The combination of TMZ and bevacizumab can be safely administered and shows promising activity in pts with advanced pancreatic NETs. Additional studies with this combination are warranted.

[Table: see text]

[Table: see text]

Irinotecan in esophageal cancer

Irinotecan (Camptosar) has shown activity in several solid tumor malignancies, including gastric and pancreatic cancer. In vitro studies suggest antitumor activity in esophageal cancer cell lines. Sequence-dependent synergy has been demonstrated in vitro between irinotecan and cisplatin. A phase I trial conducted at Memorial Sloan-Kettering Cancer Center (MSKCC) has demonstrated the safety and tolerability of cisplatin plus irinotecan. A phase II study of this combination in patients with previously untreated, advanced esophageal cancer, also at MSKCC, has demonstrated promising results. Current trials at MSKCC are attempting to combine this regimen with either paclitaxel (Taxol), fluorouracil (5-FU), or radiation therapy.

Chemotherapy in esophageal cancer

The incidence of esophageal cancer continues to increase due to a rapid increase in adenocarcinoma of the distal esophagus and gastroesophageal junction. At least 50% of patients present with metastatic cancer and most patients with localized disease will develop metastases despite potentially curative local therapy. Thus, the majority of esophageal cancer patients will become candidates for palliative chemotherapy. Traditionally, single agents effective in this disease have included cisplatin, 5-fluorouracil, and mitomycin. The combination of cisplatin and continuous-infusion 5-fluorouracil is the standard for both squamous cell carcinoma and adenocarcinoma, with a 25% to 35% response rate in metastatic disease. More recently, paclitaxel has shown favorable results as a single agent or in combination with cisplatin in both disease histologies. One-hour weekly paclitaxel, a promising schedule with little toxicity, is under active investigation. Weekly irinotecan and cisplatin is a highly effective new regimen in both adenocarcinoma and squamous cell carcinoma with relatively little toxicity. Vinorelbine has demonstrated response in squamous cell carcinoma and has less toxicity than its predecessor, vindesine. Use of newer agents in combination with concurrent radiotherapy in locally advanced disease is the subject of ongoing clinical trials.

Phase II clinical trial of 13-cis-retinoic acid and interferon-alpha-2a in patients with advanced esophageal carcinoma

BACKGROUND

Interferon in combination with 5-fluorouracil has been shown to be active in squamous cell carcinoma (SCC) and adenocarcinoma (AC) of the esophagus. 13-cis-retinoic acid (CRA) has chemopreventive activity in SCC of the head and neck, and, in combination with interferon, has antitumor activity in SCC of the skin and cervix.

METHODS

The activity and toxicity of CRA and interferon-alpha-2a (IFN) in patients with advanced esophageal carcinoma was evaluated in a Phase II single institution trial. Patients had unresectable or metastatic AC or SCC of the esophagus. One prior chemotherapy regimen was allowed. IFN was given by daily subcutaneous injection at a dose of 3 million U and CRA was taken orally at a dose of 1 mg/kg/day in 2 divided doses. Treatment was given in cycles of 4 weeks and continued until documented disease progression.

RESULTS

Of the 19 patients entered, 15 were evaluable for response and toxicity. One patient was evaluable for response only and one patient was evaluable for toxicity only. Evaluable patients were predominantly male (15 patients), and had AC (13 patients). All had AJCC Stage IV disease and 12 were pretreated. Patients completed an average of two cycles of therapy (range, one to six cycles) prior to progression of disease. National Cancer Institute Common Toxicity Criteria Grade 3/4 toxicity was notable for nausea (25%) and fatigue (31%). No major objective responses were recorded. Eleven patients with AC and 3 patients with SCC had rapid progression of disease. One patient with AC was found to have a minor response for 22 weeks and 1 patient with AC had stable disease for 45 weeks.

CONCLUSIONS

This regimen had no significant activity in patients with advanced AC of the esophagus. Further evaluation of IFN plus CRA, using this dose and schedule, is not recommended. In comparison with prior trials of this therapy, a surprising amount of severe nausea and fatigue was observed in this trial.

Phase II clinical trial of 13-cis-retinoic acid and interferon-alpha-2a in patients with advanced esophageal carcinoma

BACKGROUND

Interferon in combination with 5-fluorouracil has been shown to be active in squamous cell carcinoma (SCC) and adenocarcinoma (AC) of the esophagus. 13-cis-retinoic acid (CRA) has chemopreventive activity in SCC of the head and neck, and, in combination with interferon, has antitumor activity in SCC of the skin and cervix.

METHODS

The activity and toxicity of CRA and interferon-alpha-2a (IFN) in patients with advanced esophageal carcinoma was evaluated in a Phase II single institution trial. Patients had unresectable or metastatic AC or SCC of the esophagus. One prior chemotherapy regimen was allowed. IFN was given by daily subcutaneous injection at a dose of 3 million U and CRA was taken orally at a dose of 1 mg/kg/day in 2 divided doses. Treatment was given in cycles of 4 weeks and continued until documented disease progression.

RESULTS

Of the 19 patients entered, 15 were evaluable for response and toxicity. One patient was evaluable for response only and one patient was evaluable for toxicity only. Evaluable patients were predominantly male (15 patients), and had AC (13 patients). All had AJCC Stage IV disease and 12 were pretreated. Patients completed an average of two cycles of therapy (range, one to six cycles) prior to progression of disease. National Cancer Institute Common Toxicity Criteria Grade 3/4 toxicity was notable for nausea (25%) and fatigue (31%). No major objective responses were recorded. Eleven patients with AC and 3 patients with SCC had rapid progression of disease. One patient with AC was found to have a minor response for 22 weeks and 1 patient with AC had stable disease for 45 weeks.

CONCLUSIONS

This regimen had no significant activity in patients with advanced AC of the esophagus. Further evaluation of IFN plus CRA, using this dose and schedule, is not recommended. In comparison with prior trials of this therapy, a surprising amount of severe nausea and fatigue was observed in this trial.

Irinotecan and cisplatin in upper gastrointestinal malignancies

Irinotecan (CPT-11 [Camptosar]), an active agent in the treatment of fluorouacil-refractory colorectal cancer, has antitumor activity in upper gastrointestinal cancers. Clinical trials from Japan indicate antitumor responses in gastric and pancreatic cancers. Cisplatin (Platinol), a central agent in the treatment of upper gastrointestinal malignancies, is a logical drug to study in combination with irinotecan in upper gastrointestinal cancers. In vitro studies have shown important sequence-dependent synergy of cisplatin/irinotecan combination therapy. Irinotecan appears to prevent removal of cisplatin-induced DNA-interstrand cross-links. Initial phase I and III trials of cisplatin plus irinotecan appear to confirm this synergy, with Japanese trials in gastric cancer showing an encouraging rate of response with acceptable toxicity. A phase I trial conducted at Memorial Sloan-Kettering Cancer Center has demonstrated the safety and tolerability of weekly cisplatin and irinotecan. Currently, a phase II trial of this weekly regimen is under way in patients with metastatic or recurrent esophageal cancer. The response proportion compares favorably to standard therapy, with relatively mild toxicity. Other phase II studies, including single-agent irinotecan in esophageal cancer and the combination of cisplatin and irinotecan in gastric cancer, are being initiated at other US institutions.