Preoperative (Neoadjuvant) Chemotherapy Plus Radiation Therapy for Esophageal Ca
May 1, 1999
Preoperative (Neoadjuvant) Chemotherapy Plus Radiation Therapy for Esophageal Carcinoma
abstract & commentary
Synopsis: Randomized studies evaluating the use of neoadjuvant chemotherapy plus radiation therapy have not consistently shown an advantage for patients receiving preoperative therapy. The incorporation of paclitaxel into the chemotherapy program together with carboplatin and continuous infusions of 5-fluorouracil all given concurrently with radiation therapy yielded a remarkable 46% pathologic complete response rate in patients coming to surgery. Eleven patients (30%) had only microscopic residual disease in the primary tumor site. Follow-up is too short to evaluate the impact on disease-free and overall survival.
Source: Meluch AA, et al. Cancer J Sci Am 1999;5: 84-91.
Esophageal cancer is expected to develop in 12,500 people in 1999 and there will be an estimated 12,200 deaths this year from the disease. It is clear that few patients with esophageal cancer survive. One hint about possible explanations for the refractoriness to therapy was described above; the early spread of potentially drug-resistant tumor cells to bone. Efforts to improve treatment outcome have included the use of adjuvant chemotherapy after definitive surgical resection of tumor. However, adjuvant chemotherapy has not improved treatment outcome in this disease.
Several studies have attempted to use neoadjuvant therapy in patients with esophageal cancer, therapy given before definitive primary surgery. The rationale for neoadjuvant therapy is strong. Preoperative systemic therapy treats metastases early and increases the likelihood that the tumor can be fully resected. However, patients who fail to respond to neoadjuvant therapy or who develop progressive or metastatic disease during preoperative treatment can be spared the morbidity of major thoracic surgery. The evidence seems to support the use of chemotherapy together with radiation therapy in order to achieve local control of the tumor and to suppress metastatic sites of disease because chemotherapy alone without radiation therapy does not improve resectability or survival.1
By contrast, the use of chemotherapy and radiation therapy concurrently is beginning to suggest some benefit. Walsh and colleagues randomly assigned 113 patients to receive two courses of preoperative cisplatin plus 5-fluorouracil plus concurrent radiation therapy followed by surgical resection or surgical resection alone.2 Neoadjuvant therapy induced pathologic complete responses in 25% of patients, documented at the time of resection. The median survival (16 months for combined modality vs 11 months for surgery alone) and the three-year survival (37% for combined modality vs 7% for surgery alone) both favored the use of neoadjuvant chemotherapy and radiation therapy. A similar study by Urba and colleagues using cisplatin, 5-fluorouracil and vinblastine chemotherapy and hyperfractionated radiation therapy induced a similar pathologic complete response rate, 28%.3 However, survival was not significantly influenced by the neoadjuvant treatment. Three-year survival on the neoadjuvant arm is 32% compared to 15% on the surgery alone arm; however, the P value for this two-fold improvement is only 0.08 on this 100-patient study.
It is clear that if there is a benefit for neoadjuvant therapy, it requires concomitant chemotherapy and radiation therapy. Bosset and colleagues undertook a study in 282 patients in which the preoperative therapy was less intensive; single-agent cisplatin separated in time from a lower total dose of radiation therapy.4 This study demonstrated no benefit to preoperative therapy.
In this context, Meluch and colleagues at the Minnie Pearl Cancer Research Network undertook a phase II study to work paclitaxel into the neoadjuvant therapy mix. Paclitaxel has a 32% response rate used as a single agent in esophageal squamous cell or adenocarcinoma,5 and synergism between paclitaxel and cisplatin has been noted in certain tumor types. Thus, in a phase II study with patients entered from 13 community-based centers, 49 patients were treated with paclitaxel 200 mg/m2 by 1-hour infusion on days 1 and 22, carboplatin given at an area under the curve of 6 on days 1 and 22, 5-fluorouracil 225 mg/m2 daily by 24-hour continuous infusion for days 1-42, with concomitant radiation therapy at a total dose of 45 Gy given in 25 fractions over five weeks in 1.8 Gy fractions.
Forty-five patients (92%) completed the six weeks of neoadjuvant treatment; three had not yet completed treatment at the time of the report and one patient had rapid tumor progression despite therapy. Of the 45 patients who had completed therapy, 40 reached the time for surgical resection and five were awaiting surgery. Thirty-four of the 40 patients had definitive resection; two had tumor progression before surgery could be performed, two had poor performance status, and two surgeons refused to operate on patients in clinical complete remission.
At the time of preoperative restaging, 38 of 40 patients (95%) had an objective response to neoadjuvant chemotherapy and radiation therapy. Among the 34 resected patients, 17 (50%) had a pathologic complete response and 17 (50%) had a partial response; twelve of the partial responders had only microscopic residual disease. If one adds to the denominator the three patients who progressed before surgery, the pathologic complete response rate was 17 of 37 (46%).
Follow-up time has been too short to evaluate response durations and survival.
The toxicity of the therapy was substantial. Sixty-five percent of patients had grade III/IV hematologic toxicity; hospitalization for febrile neutropenia was necessary in 10 patients (22%). Grade III/IV esophagitis was noted in 14 patients (31%). Twenty-three patients (51%) lost at least 10 pounds during the six-week neoadjuvant treatment period. Although there were no treatment-related deaths, three patients died perioperatively (9%), two from respiratory complications and sepsis and one from bowel infarction.
Commentary
Esophageal carcinoma is a fatal illness for nearly all those who contract it. The incidence trends are level but adenocarcinoma is increasing in frequency as squamous cell carcinoma is declining. Barrett’s esophagus in white men is the major predisposing factor. The efforts to develop neoadjuvant therapy cannot yet be said to have made a substantial impact on the disease; however, hints of activity provide a glimmer of hope.
One prospective randomized trial of concurrent chemotherapy and radiation therapy is clearly positive, and another shows a trend toward doubling of the three-year survival (up to nearly 1/3 of patients surviving) that is not yet statistically significant. Even in the Bosset et al study that failed to show a survival advantage for neoadjuvant therapy and did not involve using chemotherapy and radiation therapy together, there was a significant improvement in local control and survival for patients surviving surgery. Surgical mortality was, however, increased by the therapy. On to this playing field comes a novel neoadjuvant intervention that incorporates paclitaxel and carboplatin and administers 5-fluorouracil by continuous infusion during the radiation therapy to maximize radiation sensitization effects. The pathologic complete response rate has increased from about 25% to nearly 50%. The toxicity of the therapy, while severe, is generally manageable in general oncology practice. Prospective randomized studies that use survival as the major end point will need to be performed to assess this therapy. However, it appears highly active with nearly all patients experiencing tumor regressions.
It is difficult to compare patients on all of these studies. It is even difficult to know whether patients on a single randomized study are appropriately matched because of the tendency of clinical staging to understage patients. The use of endoscopic ultrasonography improves the accuracy of staging both in the primary tumor and the draining of lymph nodes. In addition, laparoscopy and thoracoscopy may improve the detection of nodal and hepatic metastases. It is hoped that future studies of neoadjuvant therapy use more thorough clinical staging.
It is difficult to know what the surgeons were thinking when they refused to operate on patients who were in clinical complete remission. No evidence supports the concept that clinical complete responses can be durable in esophageal cancer. An esophagus that has been rendered macroscopically free of disease may still have microscopic disease that would be left behind without definitive resection. Thus, it is hoped that future prospective studies manage to enlist the full cooperation of the surgeons upon whom so much of the evaluation and management depends.
References
1. Kelsen DP, et al. N Engl J Med 1998;339:1979-1984.
2. Walsh TN, et al. N Engl J Med 1996;335:462-467.
3. Urba S, et al. Proc Am Soc Clin Oncol 1997;16:277.
4. Bosset J, et al. N Engl J Med 1997;337:161-167.
5. Ajani JA, et al. J Natl Cancer Inst 1994;86:1086-1091.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content