Karmanos Cancer Institute, Wayne State University, Detroit, MI, USA
Received 10 June 2003; revised 6 February 2004; accepted 17 February 2004
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
This study was conducted to evaluate the efficacy and toxicity of combination carboplatin and paclitaxel in patients with esophageal cancer.
Materials and methods:
Thirty-five patients were enrolled. Patients were treated with paclitaxel 200 mg/m2 intravenously (i.v.) over 3 h and carboplatin i.v. at an AUC of 5 mg/h/ml. Thirty-three patients were assessable for toxicity and objective response.
Results:
A total of 166 treatment courses were administered with a median of five courses per patient. The objective response rate was 43% [90% confidence interval (CI) 0.30.58] by the intention-to-treat analysis. The median response duration was 2.8 months (90% CI 2.15.4). The median survival time was 9 months (90% CI 713.8) and the 1-year survival rate was 43% (90% CI 0.290.57). The major grade 34 toxicity observed was neutropenia, occurring in 17 patients (52%). There were no treatment-related deaths.
Conclusions:
The combination of carboplatin and paclitaxel is an moderately active and tolerable regimen in advanced esophageal cancer.
Key words: carboplatin, combination chemotherapy, esophageal cancer, paclitaxel
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
As a single agent, paclitaxel has demonstrated encouraging activity in a phase II trial of 50 patients with unresectable esophageal cancer with an objective response rate of 32% [4]. In vitro experiments demonstrate a synergic interaction in human cancer cell lines treated with paclitaxel followed by cisplatin [5]. In a phase I trial of cisplatin and paclitaxel, less myelotoxicity was seen when paclitaxel was administered first [6]. The combination of cisplatin and paclitaxel, however, resulted in significant neurotoxicity, which was the dose-limiting toxicity.
In comparison with cisplatin, carboplatin has lower incidence of neurotoxicity, and is easier to administer in an outpatient setting. Pharmacokinetic studies of paclitaxel administered with carboplatin demonstrate that area under the curve (AUC)-guided dosing can be accurately predicted by the Calvert formula [7], allowing carboplatin to be combined at full doses with paclitaxel [8, 9].
Based on these considerations, we conducted this phase II study to evaluate the efficacy and tolerability of the paclitaxel and carboplatin combination in patients with non-metastatic and metastatic esophageal cancer. The primary end point of this study was objective tumor response.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Patients were excluded from study participation if they had another active malignancy within the preceding year except for adequately treated basal cell, squamous cell skin cancer, or in situ cervical cancer. Pregnant or lactating women were excluded from the study. Additional exclusion criteria included previous chemotherapy for esophageal cancer within the preceding 6 months.
Study design and treatment plan
All treatment was delivered in the outpatient setting. Patients were premedicated with either oral or intravenous dexamethasone prior to paclitaxel. Diphenhydramine hydrochloride 50 mg intravenously (i.v.), ondansetron 1624 mg orally and ranitidine 50 mg i.v. were given 3060 min prior to paclitaxel. Paclitaxel infusion preceded the administration of carboplatin. The doses of paclitaxel (Taxol; Bristol-Myers Squibb, Princeton, NJ, USA) and carboplatin (Paraplatin; Bristol-Myers Squibb) for the first course were 200 mg/m2, and targeted AUC of 5 mg/h/ml (as per the Calvert formula), respectively. Cycles were to be repeated every 21 days. Dosages of paclitaxel and carboplatin were escalated to 225 mg/m2 and an AUC 6 mg/h/ml, respectively, after the first course if the nadir granulocyte count was 1000/mm3, nadir platelet count was
100 000/mm3 and there was acceptable non-hematological toxicity. Patients with non-metastatic disease were treated until best response and then received local therapy according to the investigators discretion. Local treatment could consist of chemoradiotherapy, surgery or a combination of both. Patients with non-metastatic disease were treated with curative intent.
Paclitaxel dose was reduced for grade 2 or 3 neurotoxicity. Patients with grade 4 neurotoxicity were removed from the study. Doses of carboplatin and paclitaxel were attenuated for grade 3 or 4 non-hematological toxicity, a nadir granulocyte count <500/mm3, or platelet count <50 000/mm3. The initial dose reduction was 25 mg/m2 for paclitaxel and 1 mg/h/ml of AUC for carboplatin. If toxicity persisted, a second dose reduction of paclitaxel to 135 mg/m2 was allowed. Once a dose had been reduced during a treatment cycle, re-escalation was not permitted during any other subsequent cycles.
A new cycle of therapy could begin if the neutrophil count was 1500/mm3, the platelet count was
100 000/mm3, and all relevant non-hematological toxicities were grade
2. Patients requiring a delay in therapy of >2 weeks or more than two dose reductions were removed from the study. Additionally, patients were removed from the study if any of the following occurred: disease progression, unacceptable toxicity, withdrawal of consent, or initiation of chemoradiotherapy or surgery for patients with locally advanced disease.
On-study evaluation
Standard efficacy end points of objective response, duration of response, and survival were assessed. Assessment of response was based on computed tomography. These imaging studies were performed at baseline and repeated after every two cycles of therapy. Tumor responses were categorized as complete response, partial response, disease progression or stable disease. Complete response was defined as disappearance of all measurable and evaluable disease without the appearance of any new lesions. Partial response was defined as a 50% reduction in the sum of the products of all measurable lesions compared with baseline measurements. Progressive disease was defined as a
25% increase in the sum of the products of any measurable lesion relative to either the baseline or the maximal response, or the appearance of any new lesion. Stable disease was defined as changes in tumor size that did not meet criteria for either progressive disease, or complete or partial response. Objective responses required one confirmatory follow-up evaluation at least 3 weeks after the initial response was determined. Duration of response was defined as the time from partial response until first evidence of relapse or death from any cause. Duration of response was censored for patients who were removed from the study or received antitumor therapy other than carboplatin and paclitaxel including chemoradiotherapy or surgery. Progression-free survival (or time to progression) was defined as the time from treatment start until disease progression or death from any cause. For patients still progression-free at last tumor assessment, progression-free survival was censored as of that date. Overall survival was defined as the time from treatment start until death from any cause. For patients still alive at last follow-up for vital status, overall survival was censored as of that date. Toxicities were evaluated at a minimum on day 1 of each cycle and graded according to WHO toxicity criteria.
Statistical methods
This phase II trial was planned with a Fleming two-stage design [10]. We wished to distinguish these regions of the true, unknown response rate: 0.40 versus
0.60. The two-stage design called for a maximum of 40 response-evaluable patients, 20 each in stage 1 and stage 2, and had type I error of 0.04 and power of 0.78. Nine to 12 (complete or partial) responders among the first 20 response-evaluable patients were needed to justify beginning stage 2, and nine responses were observed. Accrual slowed considerably during the later part of the trial such that it became necessary to close the trial because of it, with a total of 35 eligible patients enrolled, of whom 33 were response-evaluable. With 20 + 13 = 33 response-evaluable patients, the revised inferential error rates were: type I error of 0.03 and power of 0.67. After completing the revised stage 2, at least 19 responders were needed to conclude that the true, unknown response rate was at least 0.60. Exact, minimum-width 90% confidence intervals (CIs) for response and toxicity rates were calculated using the Casella method [11], as implemented in StatXact software [12]. Standard KaplanMeier estimates of the censored response duration and survivorship functions were computed. Owing to the small sample sizes, survival statistics (e.g. median) were estimated more conservatively using linear interpolation [13] between successive event times on the KaplanMeier curves.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Response and survival
Fifteen of the 35 (43%; 90% CI 0.300.58) registered patients had a partial response. Among the 33 response-evaluable patients, the partial response rate was 45% (90% CI 0.310.61). With only 15 responders among 33 response-evaluable patients, we concluded that the sample response proportion (15/33 = 45%) better supported the null hypothesis that the true unknown response rate was at most 0.40. Response rates were not statistically significantly different in the patients with adenocarcinoma versus squamous cell histology (50% versus 31%, respectively), non-metastatic versus metastatic disease (33% versus 50%, respectively), or in patients with gastroesophageal junction cancer versus mid-esophageal cancer (70% versus 32%, respectively). For the 15 partial responders, the median response duration was 2.8 months (90% CI 2.15.4). The KaplanMeier estimate of the response duration curve for these 15 patients is given in Figure 1. The 6-month relapse-free survival was 22% (90% CI 0.030.41). Two patients were still in partial remission at the time of surgery for a censoring rate of two of 15 (13%). In the 15 responders, 12 of 13 relapses occurred within 7 months of completion of chemotherapy. The remaining relapse occurred at 15.2 months.
|
|
|
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Paclitaxel has demonstrated significant single-agent activity in esophageal cancer, with response rates of 32% [4]. The rationale for combining paclitaxel with platinum compounds is the synergy between paclitaxel and cisplatin in the preclinical models [5]. Encouraged by these findings, a phase II trial of paclitaxel added to cisplatin and 5-FU in metastatic esophageal carcinoma was conducted [18]. Although the response rate of 48% was a potential improvement over historical controls, the frequency of grade 34 toxicities was significant. Hospitalizations were required in 48% of the patients due to gastrointestinal and hematological toxicity. A subsequent phase II trial evaluated a 24-h infusion paclitaxel with cisplatin in patients with metastatic esophageal cancer [19]. The response rate was 44%, but toxicity remained significant. Hospitalization was required in half of the patients. Of even more concern was the 11% treatment-related mortality secondary to cardiac toxicity, stroke or cardiac arrhythmias.
In our study the objective response rate and median survival with carboplatin and paclitaxel were 43% and 9 months, respectively, which are comparable to the reported response rates with cisplatin-based regimens. However, this outpatient-based regimen was easier to administer and was better tolerated, with fewer side-effects and hospitalization, which were required in 25% of the patients [19]. Non-metastatic esophageal cancer is potentially curable in a proportion of patients through multimodality treatment involving chemoradiotherapy [20]. Both distant metastases and local relapses remain a major cause of mortality in patients treated with conventional systemic therapies such as 5-FU and cisplatin. Clinical trials aimed at introducing more active agents and combinations for systemic treatment in localized disease are ongoing. Paclitaxel, for example, is being developed in combination with a platinum compound administered concurrently with radiation therapy [21], as part of an induction therapy [22] prior to either chemoradiotherapy or surgery, or in the adjuvant setting [23]. The induction chemotherapy approach followed by chemoradiotherapy offers the advantage of evaluating the efficacy of drug combinations in early disease as well as allowing patients to receive conventional therapy. In our study, carboplatin and paclitaxel administration prior to localized therapy was feasible without increased toxicity of the local treatments. It is uncertain, however, whether response to induction chemotherapy would further decrease the risk of local or distant relapse following local treatment. After a relatively long follow-up of 41 months, 27% of patients with non-metastatic disease in this trial were free of any evidence of relapse. In a recently reported phase II trial of induction carboplatin and paclitaxel prior to esophagectomy, major clinical responses and resectability were achieved in 61% and 77% of the patients, respectively [22]. Complete pathologic response was seen in 12% of the patients, but that was less than would be expected with concurrent chemoradiotherapy. However, when considering that a systemic treatment efficacy is a primary motive with induction chemotherapy, frequent clinical and complete pathologic responses may indicate control of micrometastatic disease. A recently reported phase II trial evaluated adjuvant cisplatin and paclitaxel in 59 patients with resected esophageal and gastroesophageal junction tumors. The regimen appeared to be active, with a 2-year survival rate of 60%. Grade 3 or 4 toxicities were observed in 56% of the patients. Eventually, randomized trails will be needed to evaluate the role of taxane-based adjuvant or neo-adjuvant chemotherapy in localized esophageal cancer.
In conclusion, the carboplatin and paclitaxel combination is a reasonable treatment for patients with metastatic esophageal cancer due to the improved toxicity profile as compared with other cisplatin-based regimens. Future trials focusing on the integration of novel targeted therapies such as epidermal growth factor receptor blockers or cyclooxygenase-2 inhibitors with conventional chemotherapy might further improve systemic efficacy in patients with this disease. Given the moderate activity and safety profile of carboplatin and paclitaxel, we believe this combination would be a reasonable chemotherapy regimen for these investigations.
![]() |
Acknowledgements |
---|
![]() |
FOOTNOTES |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2. Iizuka T, Kakegawa T, Ide H et al. Phase II evaluation of cisplatin and 5-fluorouracil in advanced squamous cell carcinoma of the esophagus: a Japanese Esophageal Oncology Group Trial. Jpn J Clin Oncol 1992; 22: 172176.[ISI][Medline]
3. Polee MB, Kok TC, Siersema PD et al. Phase II study of the combination cisplatin, etoposide, 5-fluorouracil and folinic acid in patients with advanced squamous cell carcinoma of the esophagus. Anticancer Drugs 2001; 12: 513517.[CrossRef][ISI][Medline]
4. Ajani JA, Ilson DH, Daugherty K et al. Activity of taxol in patients with squamous cell carcinoma and adenocarcinoma of the esophagus. J Natl Cancer Inst 1994; 86: 10861091.[Abstract]
5. Jekunen AP, Christen RD, Shalinsky DR, Howell SB. Synergistic interaction between cisplatin and taxol in human ovarian carcinoma cells in vitro. Br J Cancer 1994; 69: 299306.[ISI][Medline]
6. Rowinsky EK, Gilbert MR, McGuire WP et al. Sequences of taxol and cisplatin: a phase I and pharmacologic study. J Clin Oncol 1991; 9: 16921703.[Abstract]
7. Calvert AH, Newell DR, Gumbrell LA et al. Carboplatin dosage: prospective evaluation of a simple formula based on renal function. J Clin Oncol 1989; 7: 17481756.[Abstract]
8. Kearns CM, Belani CP, Erkmen K et al. Pharmacokinetics of paclitaxel and carboplatin in combination. Semin Oncol 1995; 22: 17.
9. Creaven P, Raghavan D, Pendyala L et al. Phase I study of paclitaxel and carboplatin: implications for trials in head and neck cancer. Semin Oncol 1995; 22: 1316.
10. Fleming T. One-sample multiple testing procedure for Phase II clinical trials. Biometrics 1982; 38: 143151.[ISI][Medline]
11. Casella G. Refining binomal confidence intervals. Can J Stat 1987; 14: 113129.[ISI]
12. Mehta C, Patel N. StatXact 5: Statistical Software for Exact Nonparameteric Inference, User Manual. Cambridge, MA: Cytel Software Corporation 1999.
13. Lee E. Statistical Methods for Survival Data Analysis, 2nd edition. New York, NY: Wiley & Sons 1992; pp. 7778.
14. Bleiberg H, Conroy T, Paillot B et al. Randomised phase II study of cisplatin and 5-fluorouracil (5-FU) versus cisplatin alone in advanced squamous cell oesophageal cancer. Eur J Cancer 1997; 33: 12161220.[CrossRef][Medline]
15. Spiridonidis CH, Laufman LR, Jones JJ et al. A phase II evaluation of high dose cisplatin and etoposide in patients with advanced esophageal adenocarcinoma. Cancer 1996; 78: 20702077.[CrossRef][ISI][Medline]
16. Wadler S, Haynes H, Beitler JJ et al. Phase II clinical trial with 5-fluorouracil, recombinant interferon-alpha-2b, and cisplatin for patients with metastatic or regionally advanced carcinoma of the esophagus. Cancer 1996; 78: 3034.[CrossRef][ISI][Medline]
17. Warner E, Jensen JL, Cripps C et al. Outpatient 5-fluorouracil, folinic acid and cisplatin in patients with advanced esophageal carcinoma. Acta Oncol 1999; 38: 255259.[CrossRef][ISI][Medline]
18. Ilson DH, Ajani J, Bhalla K et al. Phase II trial of paclitaxel, fluorouracil, and cisplatin in patients with advanced carcinoma of the esophagus. J Clin Oncol 1998; 16: 18261834.[Abstract]
19. Ilson DH, Forastiere A, Arquette M et al. A phase II trial of paclitaxel and cisplatin in patients with advanced carcinoma of the esophagus. Cancer J 2000; 6: 316323.[ISI][Medline]
20. Herskovic A, Martz K, al-Sarraf M et al. Combined chemotherapy and radiotherapy compared with radiotherapy alone in patients with cancer of the esophagus. N Engl J Med 1992; 326: 15931598.[Abstract]
21. Wang H, Nguyen N, Ryu J et al. Phase II trial of concurrent paclitaxel/carboplatin/radiation followed by paclitaxel/carboplatin consolidation or surgery for esophageal cancer. In ASCO Proceedings, Orlando, FL, 2002. Alexandria, VA: American Society of Clinical Oncology (Abstr 2278).
22. Keresztes R, Port J, Ferrara CA, Altorki NK. Taxol and carboplatin, an effective preoperative regimen for carcinoma of the esophagus: results of a phase II trial. In ASCO Proceedings, Orlando, FL, 2002. Alexandria, VA: American Society of Clinical Oncology (Abst 636).
23. Armanios M, Xu R, Forastiere A et al. Phase II adjuvant chemotherapy for resected adenocarcinoma of the esophagus, gastro-esophageal (GE) junction and cardia (E8296): A trial of the Eastern Cooperative Oncology Group. In ASCO Proceedings, Chicago, IL, 2003. Alexandria, VA: American Society of Clinical Oncology (Abstr 1190).