AIM: To investigate the role of perioperative chemoradiotherapy (CRT) in the treatment of locally advanced thoracic esophageal squamous cell carcinoma (ESCC).

METHODS: Using preoperative computed tomography (CT)-based staging criteria, 238 patients with ESCC (stage II-III) were enrolled in this prospective study between January 1997 and June 2004. With informed consent, patients were randomized into 3 groups: preoperative CRT (80 cases), postoperative CRT (78 cases) and surgery alone (S) (80 cases). The 1-, 3-, 5- and 10-year survival were followed up. Progression-free survival (PFS) was chosen as the primary endpoint by treatment arm measured from study entry until documented progression of disease or death from any cause. The secondary endpoint was overall survival (OS) determined as the time (in months) between the date of therapy and the date of death. Other objectives were surgical and adjuvant therapy complications.

RESULTS: With median follow-up of 45 mo for all the enrolled patients, significant differences in the 1-, 3-, 5-, 10-year OS (91.3%, 63.5%, 43.5%, 24.5% vs 91%, 62.8%, 42.3%, 24.4% vs 87.5%, 51.3%, 33.8%, 12.5%, P = 0.0176) and PFS (89.3%, 61.3%, 37.5%, 18.1% vs 89.1%, 61.1%, 37.2%, 17.8% vs 84.5%, 49.3%, 25.9%, 6.2%, P = 0.0151) were detected among the 3 arms. There were no significant differences in OS and PFS between the preoperative CRT and postoperative CRT arm (P > 0.05). For the patients who had radical resection, significant differences in median PFS (48 mo vs 61 mo vs 39.5 mo, P = 0.0331) and median OS (56.5 mo vs 72 mo vs 41.5 mo, P = 0.0153) were detected among the 3 arms, but there were no significant differences in OS and PFS between the preoperative CRT and postoperative CRT arm (P > 0.05). The local recurrence rates in the preoperative CRT, postoperative CRT group and S group were 11.3%, 14.1% and 35%, respectively (P < 0.05). No significant differences were detected among the 3 groups when comparing complications but tended to be in favor of the postoperative CRT and S groups (P > 0.05). Toxicities of CRT in the preoperative or postoperative CRT arms were mostly moderate, and could be quickly alleviated by adequate therapy.

CONCLUSION: Rational application of preoperative or postoperative CRT can provide a benefit in PFS and OS in patients with locally advanced ESCC.

Esophageal squamous cell carcinoma (ESCC) is one of the common malignancies and the seventh leading cause of cancer-related deaths in the world[1]. The 5-year survival rate is only 15%-20% in patients with locally advanced esophageal cancer[2]. Treatment failure mainly results from recurrence or metastasis. A standardized comprehensive treatment is still under development. Radiotherapy can control local-regional esophageal cancer and chemotherapy has both local and systemic antineoplastic activity. With increasing enthusiasm for multidisciplinary treatment modalities to improve outcome, a commonly employed treatment approach for esophageal cancer is chemoradiotherapy (CRT) in addition to surgical resection[3]. Preoperative CRT had been applied to patients with esophageal carcinoma in an effort to reduce the relapse rate and improve survival. Many studies have demonstrated the effectiveness of neoadjuvant CRT[4-9]. In contrast to data available on preoperative CRT, although a number of trials compared adjuvant chemotherapy or radiotherapy with surgery alone, there are almost no randomized trials comparing adjuvant CRT with surgery alone. The only exception is a trial that compared surgery alone with surgery and adjuvant CRT for patients with resectable cancers of the stomach and gastroesophageal junction[10]. Although the results suggest that resection followed by concurrent CRT should be considered for patients with adenocarcinoma of the stomach and the gastroesophageal junction, the question remains whether postoperative CRT can improve overall survival in patients with ESCC.

To our knowledge, as yet no study is available to determine the best sequencing of CRT and surgery. To resolve this deficiency, the present prospective study was conducted in our center from January 1997 to June 2004.

By June 30 2009, 228 out of all cases were followed up by means of telephone or outpatient service, and 10 cases were lost. There were 3.8% treatment-related deaths in preoperative CRT but no treatment-related deaths in the postoperative CRT and S group, and 51.3% of patients died from the local tumor recurrence and distant metastasis.

Median PFS was 46.5 mo (95% CI: 37.4-66.3) in the preoperative CRT group, 45 mo (95% CI: 35.8-67) in the postoperative CRT group, and 32.5 mo (95% CI: 25.4-42.3) in the S group. The 1-year PFS in preoperative CRT, postoperative CRT and S groups was 89.3%, 89.1% and 84.5%, respectively, with no significant differences (χ² = 0.64, P = 0.4123). However, the 3-, 5- and 10-year PFS of the preoperative CRT (61.3%, 37.5%, 18.1%, respectively) and postoperative CRT groups (61.1%, 37.2%, 17.8%, respectively) were significantly different (χ² = 4.16, P = 0.0319; χ² = 4.14, P = 0.0321; χ² = 5.38, P = 0.0203; respectively) from those of the S group (49.3%, 25.9%, 6.2%, respectively). There was no significant difference in PFS between the neoadjuvant and adjuvant therapy groups (χ² = 0.14, P = 0.7060) (Figure 1A). Median OS was 53 mo (95% CI: 43-64.3) in the preoperative CRT group, 48 mo (95% CI: 38.5-72.5) in the postoperative CRT group, and 36 mo (95% CI: 29.7-47.3) in the S group. The 1-year OS in the preoperative CRT, postoperative CRT and S groups was 91.3%, 91% and 87.5%, respectively, which were not significantly different (χ² = 0.72, P = 0.3970). However, the 3-, 5- and 10-year survival rates of the preoperative CRT (63.5%, 43.5%, 24.5%, respectively) and postoperative CRT groups (62.8%, 42.3%, 24.4%, respectively) were significantly different (χ² = 3.98, P = 0.0453; χ² = 4.76, P = 0.0402; χ² = 4.27, P = 0.0389; respectively) from those of the S group (51.3%, 33.8%, 12.5%, respectively). There was no significant difference in OS between the neoadjuvant and adjuvant therapy groups (χ² = 0.46, P = 0.4978) (Figure 1B).

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Kaplan-Meier curves. A: Postoperative progression-free survival. Logrank test: χ² = 8.39, P = 0.0151; “+”: Loss to follow-up. Preoperative vs surgery: χ² = 6.81, P = 0.0091; Postoperative vs surgery: χ² = 5.38, P = 0.0203; Preoperative vs postoperative: χ² = 0.14, P = 0.7060; B: Postoperative overall survival. Logrank: χ² = 8.07, P = 0.0176; “+”: Loss to follow-up. Preoperative vs surgery: χ² = 7.85, P = 0.0051; Postoperative vs surgery: χ² = 5.33, P = 0.0209; Preoperative vs postoperative: χ² = 0.46, P = 0.4978; C: Postoperative progression-free survival of radical resection. Logrank: χ² = 6.82, P = 0.0331; “+”: Loss to follow-up. Preoperative vs surgery: χ² = 6.16, P = 0.0130; Postoperative vs surgery: χ² = 4.02, P = 0.0449; Preoperative vs postoperative: χ² = 0.22, P = 0.6416; D: Postoperative overall survival of radical resection. Logrank: χ² = 8.36, P = 0.0153; “+”: Loss to follow-up. Preoperative vs surgery: χ² = 7.65, P = 0.0057; Postoperative vs surgery: χ² = 4.78, P = 0.0288; Preoperative vs postoperative: χ² = 0.16, P = 0.6873.

For the patients who experienced radical resection, median PFS of the preoperative CRT group, postoperative CRT group and S group was 48 mo (95% CI: 40-67.0), 61 mo (95% CI: 44.0-75.4), and 39.5 mo (95% CI: 30-60), respectively. A significant difference in PFS was detected among the 3 arms (χ² = 6.82, P = 0.0331), however, there was no significant difference in PFS between neoadjuvant and adjuvant therapy groups (χ² = 0.22, P = 0.6416) (Figure 1C). There was a significant difference in OS among the 3 arms (χ² = 8.36, P = 0.0153). The median OS of the preoperative CRT group, postoperative CRT group and S group was 56.5 mo (95% CI: 44-72), 72 mo (95% CI: 49.6-88.4), and 41.5 mo (95% CI: 31.6-57.7), respectively. However, there was no significant difference in OS between neoadjuvant and adjuvant therapy groups (χ² = 0.16, P = 0.6873) (Figure 1D).

The local recurrence rates of the preoperative CRT group, postoperative CRT group and S group were 11.3%, 14.1% and 35% respectively, and showed a statistically significant difference (P = 0.011). There was no significant difference among the 3 groups when comparing perioperative complications but there was a tendency in favor of the postoperative CRT and S groups (P = 0.179).

For the 158 patients receiving CRT, incidence of grade 3 or greater leucopenia, thrombocytopenia, anemia and vomiting were 11.4% (18/158), 6.3% (10/158), 1.3% (2/158) and 6.3% (10/158), respectively. Postoperative pathologic staging and perioperative complications, and the causes of death are presented in Table 2.

Patients with resectable ESCC should receive multimodal treatment to prolong OS[11,12]. There is still controversy about how to improve prognosis and how to reduce local recurrence and distant metastasis. Both chemotherapy and radiotherapy may be active against different tumor cell populations, the chemotherapy may be effective against micrometastases while radiation counteracts spatial metastases. In the Western world, preoperative chemotherapy and CRT can increase OS by 4.4% and 6.4%, respectively, however, treatment-related mortality increases by 1.7% with neoadjuvant chemotherapy and by 3.4% with CRT, compared with surgery alone[13]. There were a few nonrandomized trials on postoperative CRT in treating esophageal carcinoma[14-19], however, the results had discrepancies. Regretfully, the most optimal sequence of CRT in relation to surgical resection is unclear. Thus our study is a unique randomized controlled study to evaluate the outcome of preoperative CRT in patients with local advanced thoracic ESCC, and it includes a long-term follow-up.

The present study showed a benefit in PFS and OS with neoadjuvant or adjuvant therapy. Meanwhile, our study indicated that preoperative CRT reduced the rate of local recurrence, and the survival benefit resulted from improved local cancer control brought about by the adjuvant arm. Although there was no significant difference found in 1-year PFS and OS, both preoperative CRT and postoperative CRT showed a significant advantage in longtime PFS and OS. A meta-analysis had suggested that preoperative CRT may improve survival and locoregional control but was associated with higher toxicity and increased mortality[3]. Our results also confirm this point. Rice and coworkers found that 31 patients treated with postoperative adjuvant CRT had improved survival, in a retrospective review[20]. Together with our present study, it further confirms that postoperative CRT may be an alternative option especially for locally advanced thoracic ESCC. Moreover, the results indicated that postoperative CRT may have almost the same long-term of efficacy as preoperative CRT, although the latter showed a nonsignificant trend of higher survival compared to the former.

To throw further light on the benefit in PFS and OS with neoadjuvant or adjuvant therapy for ESCC, we subsequently analyzed the data in patients who had undergone radical resection. The results also showed the same advantages as the above when including the palliative resection and esophageal bypass patients. The results also showed no significant difference in survival rates when comparing the preoperative CRT and postoperative CRT arms. It further clarifies that adjuvant CRT can provide almost the same long-term of efficacy as neoadjuvant CRT.

In addition, there was no suggestion that treatment-related mortality was increased by the use of postoperative CRT, which was accomplished with manageable toxicity. Most patients had less than grade 2 hematological toxicities. However, our results showed an increase in postoperative deaths and a trend for relatively higher perioperative complications in the neoadjuvant CRT arm. The involved reasons could be as follows: surgeons may undertake a challenging esophagectomy resulting in surgical difficulty and postoperative complications when performed after neoadjuvant CRT. For example, radiation might contribute to the failure of an anastomotic leak and postoperative acute lung injury. The results also showed that preoperative CRT can facilitate complete resection by downstaging tumors when compared to postoperative CRT and surgery alone. Thus, whether or not the survival benefit of neoadjuvant CRT can be negated by an increase in postoperative deaths should of concern. In addition, loss to follow-up was low (4.2%), with only 3 patients receiving preoperative CRT and 3 patients receiving postoperative CRT and 4 patients receiving surgery alone, thus our study is robust.

In conclusion, long-term survival is maximized by the use of CRT followed by surgery for locally advanced esophageal cancer. However, patients are more likely to develop toxicity. As therapies improve, it is likely that the toxicity may be reduced and neoadjuvant CRT may provide a more marked benefit in esophageal cancer. Meanwhile, postoperative CRT can also be safely administered and considered as the multimodal treatment of choice for locally advanced ESCC. In appropriately selected patients, either pre- or postoperative CRT is a viable strategy. Further comparison of pre- and postoperative CRT in treating esophageal cancer is required for verification through multicenter and large sample randomized clinical trials.