Introduction
Esophageal cancer is a major global health hazard. An estimated 482β300 new esophageal cancer cases and 406β800 deaths occurred in 2008 worldwide [
1]. The projected numbers of new esophageal cancer cases and deaths in the United States were 17β460 and 15β070 in 2012, respectively [
2]. In China, esophageal cancer is the fourth common malignant tumor with high morbidity and malignancy. More than half of patients with esophageal cancer were diagnosed with locally advanced disease when they presented with dysphagia [
3].
Surgery is the standard treatment for resectable esophageal cancer; however, the 5-year survival rate is only 15%-34% [
4,
5] despite changes in surgical techniques and extension of surgical resection over the past 20 years. Most patients who undergo radical resection for esophageal cancer eventually relapse and die as a result of disease progress [
6]. Given the difficulties in administering chemotherapy or radiotherapy soon after a surgical procedure, the high perioperative morbidity, and the disappointing results of trials of adjuvant chemotherapy [
7] or radiotherapy [
8], neoadjuvant treatment is being extensively studied.
Role of neoadjuvant chemoradiotherapy (nCRT) based on clinical trials
Definitive concurrent chemoradiotherapy (CCRT) vs. nCRT followed by surgery
Radiation Therapy Oncology Group (RTOG) 85-01 [
9-
11] trial demonstrated that combining chemotherapy with radiotherapy is superior to radiotherapy alone in locally advanced esophageal cancer, thereby establishing CCRT as a standard treatment. In this trial, patients were randomly assigned to receive CCRT (5-FU/cisplatin/50βGy) or radiotherapy alone (64βGy). Patients in the CCRT group showed a significant improvement in both median survival rate (14.1 months vs. 9.3 months) and 5-year overall survival (OS) rate (27% vs. 0%,
P<0.0001). The 8-year survival rate was 22%, and the projected 10-year survival rate was 20%. The incidence of local failure was high even though CCRT showed good performance (47% vs. 65%).
Given the high incidence of local failure after conventional CCRT, RTOG 94-05 trial [
12] was performed to compare a higher radiation dose of 64.8βGy with the standard dose of 50.4βGy using the same chemotherapy regimen (5-FU/cisplatin). The study was closed after interim analysis showed no probability of superiority in the high-dose group. No significant difference was observed in median survival (13.0 months vs. 18.1 months,
P = NS), two-year survival (31% vs. 40%,
P = NS), and local failure (56% vs. 52%,
P = NS) between two arms. In 2011, Kachnic [
13] reported a longitudinal quality of life (QoL) analysis of this trial. Total mean QoL was low in the high-dose arm (
P = 0.02) and remained low 8 months and 12 months after the start of CCRT (
P = NS). Consequently, CCRT with 5-FU and cisplatin at a radiation dose of 50.4 Gy was established as the standard of care for patients with esophageal cancer.
Despite a dramatic improvement in survival, CCRT was associated with persistent disease or local recurrence in>50% of patients. Two randomized trials [
14,
15] addressed the necessity of surgery following combined modality therapy. All patients with potentially resectable esophageal squamous cell carcinoma (ESCC) received induction chemotherapy (5-FU/leucovorin/etoposide/cisplatin) followed by CCRT (etoposide/cisplatin/40βGy) in a German study [
14]. Patients were then randomized to surgery vs. continuing CCRT. The 2-year local control rate was significantly improved in patients undergoing surgery (64.3% vs. 40.7%,
P = 0.03). No significant survival benefit was seen (median survival, 16.4 months vs. 14.9 months; 3-year survival rate, 31.3% vs. 24.4%;
P = NS). Overall treatment-related mortality was significantly higher in patients undergoing surgery (12.8% vs. 3.5%;
P = 0.03). The Federation Francophone de Cancerologie Digestive (FFCD) 9102 [
15] hypothesized that CCRT and trimodality treatment (CCRT followed by surgery) would yield equivalent outcomes in patients responding to nCRT. Patients who had at least a partial response after nCRT were then randomized to either surgery or additional CCRT. No significant difference in 2-year survival (33.6% vs. 39.8%;
P = NS) or median survival (18 months vs. 19 months;
P = NS) was seen between the groups. The death rate at 3 months following treatment was 9% in the surgery group vs. 1% in the non-surgical group. Thus, if patients respond to chemoradiotherapy in ESCC, surgery may not be needed after nCRT.
These two trials yield the only low-level evidence comparing definitive CCRT with trimodality treatment. They show the difficulty of translating the local control benefit obtained with trimodality treatment into a significant survival benefit in patients with esophageal cancer. Suntharalingam
et al. [
16] found that contemporarily treated patients had a decreased risk of death (hazard ratio (HR), 0.32) if treated with trimodality treatment compared with CCRT alone. Early assessment of responders and non-responders to nCRT is attracting clinical interest regarding the appropriate use of trimodality therapy.
Surgery alone vs. nCRT followed by surgery
Randomized trials comparing surgery alone with nCRT followed by surgery in patients with clinically resectable cancer show conflicting results.
A population-based study showed that long-term survival after esophagectomy without neoadjuvant therapy has not improved since 2000. The 5-year survival rate increased from 19.7% in 1987-1991 to 30.7% in 1997-2000 but remained at 30.5% in 2001-2005 [
5]. Several studies and meta-analyses have shown a significant survival benefit of using nCRT plus surgery vs. surgery alone [
17-
21]. Walsh [
17] reported the first randomized study to evaluate the role of nCRT combined with surgery. nCRT (cisplatin/5-FU/40βGy) followed by surgery was associated with a longer median survival (16 months vs. 11 months,
P = 0.01) and a higher 3-year survival rate (32% vs. 6%,
P = 0.01) compared with surgery alone.
A meta-analysis performed by Gebski [
18] revealed that patients with both ESCC and adenocarcinoma benefit from preoperative chemoradiotherapy. Ten randomized comparisons of nCRT vs. surgery alone (1209 patients) were identified with local operable esophageal cancer. A significant survival benefit for nCRT was found regardless of histological tumor types. The HR for all-cause mortality with nCRT vs. surgery alone was 0.81 (
P = 0.002), corresponding to a 13% absolute difference in survival at 2 years. However, a recent meta-analysis [
19] involving 12 randomized controlled trials suggested that compared with surgery alone, trimodality treatment was associated with improved 1-, 3-, and 5-year survival but not with increased postoperative morbidity and mortality in patients with esophageal carcinoma.
The CALGB9781 trial [
20] compared trimodality therapy (cisplatin/fluorouracil/radiotherapy followed by surgery) with surgery alone for esophageal cancer. Results showed a favorable pathologic complete response (pCR) rate of 40% and a 4.5 year median survival in the trimodality therapy group. However, the clinical trial enrolled only 30 patients in the treatment arm.
The CROSS trial [
21,
22] was a multicenter phase III randomized study in which patients with resectable esophageal tumors received nCRT (paclitaxel/carboplatin/41.4βGy) followed by surgery vs. surgery alone. The reported R0 resection rate was 92% (148/161) in the trimodality arm vs. 69% (111/161) in the surgery alone arm. The pCR rate was 29% for trimodality arm. The OS was significantly better in trimodality group (HR 0.657,
P = 0.003). The median survival was 49.4 months in the trimodality arm vs. 24.0 months in the surgery alone arm. The respective OS rates at 1, 2, 3, and 5 years were 82%, 67%, 58%, and 47% in the trimodality arm compared with 70%, 50%, 44%, and 34% in the surgery alone arm, respectively. Given the acceptable toxicity and improved survival, weekly administrations of carboplatin and paclitaxel with concurrent radiotherapy followed by surgery may be considered as a reasonable choice for patients with resectable esophageal or esophagogastric junction cancer.
Compared with the results of the CROSS study, the results of an interim analysis of another phase-III randomized controlled study (FFCD 9901) [
23] showed that nCRT with cisplatin and fluorouracil did not improve OS but enhanced postoperative mortality rate for patients with localized esophageal cancer compared with surgery alone. Full publications of these data are awaited.
Neoadjuvant chemotherapy followed by surgery vs. nCRT followed by surgery
Radiotherapy, as a regional treatment modality, plays a vital role in neoadjuvant therapy in esophageal cancer patients because of its synergistic effect with chemotherapy. In the trial of Stahl
et al. [
24], patients with locally advanced adenocarcinoma of the lower esophagus or gastric cardia were randomly allocated to induction chemotherapy followed by surgery or sequential chemotherapy, chemoradiotherapy, and surgery. The R0 resection rate was almost equal between the two groups (69.5% vs. 71.5%,
P = NS). Nevertheless, pCR (15.6% vs. 2.0%,
P = 0.03) and the rate of tumor-free lymph nodes (ypN0) (64.4% vs. 36.7%,
P = 0.01) significantly increased by chemoradiotherapy compared with chemotherapy alone. Preoperative radiotherapy marginally improved the 3-year survival rate (27.7% vs. 47.4%,
P = 0.07). Postoperative mortality did not increase in the chemoradiotherapy group (10.2% vs. 3.8%,
P = NS). Interestingly, survival benefit was achieved although the postoperative mortality was more than doubled by adding radiotherapy. The improvement in both local tumor-free and OS indicates that preoperative chemoradiotherapy is most valuable in curing patients with localized esophagogastric adenocarcinoma.
Burmeister
et al. [
25] designed a randomized phase-II trial to compare preoperative chemotherapy (cisplatin/5-FU) and chemoradiotherapy (cisplatin/5-FU/35βGy) for resectable adenocarcinoma of the esophagus and gastroesophageal junction. The pCR was higher (31% vs. 8%,
P = 0.01) and R1 resection rate was lower (0% vs. 11%,
P = 0.04), thereby favoring recipients of chemoradiotherapy. No differences existed in the median progression-free survival (14 months vs. 26 months,
P = 0.37) and OS (29 months vs. 32 months,
P = 0.83) between chemotherapy and chemoradiotherapy.
An updated meta-analysis [
26] provided strong evidence of the survival benefit of nCRT or chemotherapy over surgery alone in patients with esophageal carcinoma. In this meta-analysis, the HRs for all-cause mortality for nCRT and neoadjuvant chemotherapy vs. surgery alone were 0.78 (
P<0.0001) and 0.87 (
P = 0.005), respectively. However, a clear advantage of nCRT over neoadjuvant chemotherapy was not established. The HR for the overall indirect comparison of all-cause mortality for nCRT vs. neoadjuvant chemotherapy was 0.88 (
P = 0.07).
Neoadjuvant radiotherapy followed by surgery vs. nCRT followed by surgery
A study [
27] suggested that the nCRT group had more satisfactory tumor reduction than neoadjuvant radiotherapy alone. Neoadjuvant radiotherapy and/or chemotherapy were well tolerated by patients and did not increase postoperative complications. The results demonstrated that patients subjected to trimodality treatment presented complete or near-complete response rate in 62.5% of patients, whereas patients submitted to surgery followed by radiotherapy presented 33.3%. The results obtained in this study with neoadjuvant therapy suggest significant benefits for patients within the philosophy of “multimodal” treatment. Neoadjuvant therapy with chemotherapy and radiotherapy can improve local effectiveness and complete resection rates as well as a reduction in clinical staging in the preoperative period, which benefits patients.
nCRT regimens
According to the new 2013 National Comprehensive Cancer Network guidelines of esophageal cancer [
28], the recommended chemotherapy regimens with category 1 evidence include paclitaxel/carboplatin, cisplatin/fluoropyrimidine, and oxaliplatin/fluorouracil. The recommended radiation dose is 41.4-50.4 Gy (1.8-2 Gy/day).
Many trials evaluated other chemotherapy regimens, including 5-FU/oxaliplatin, cisplatin/irinotecan/cetuximab, and paclitaxel/platinum/5-FU [
29-
37]. Combinations with targeted drugs such as bevacizumab, cetuximab, and erlotinib were also evaluated [
38-
41]. However, none of these trials showed a significant improvement in survival over historical results with 5-FU and cisplatin (Table 1).
Relationship between pCR after nCRT and survival
Increasing evidence shows that patients with pCR have benefits in terms of survival regardless of the approach used to achieve it [
42-
45]. pCR seems to be a significant prognostic factor for improved OS after surgery in patients undergoing nCRT for esophagus and gastroesophageal junction cancers. In the study of Urba
et al. [
42], patients with pCR have a median survival of 50 months and a 3-year survival rate of 64%, compared with patients with residual tumor in the surgical specimen with a median survival of 12 months and a 3-year survival rate of 19% (
P = 0.01). Berger
et al. [
43] correlated OS with pCR and reported that the five-year survival of esophageal cancer patients who achieved pCR after preoperative chemoradiotherapy was almost 50%. In the study of Scheer
et al. [
44], data from 22 articles suggested that patients with pCR are two to three times more likely to survive than those with residual tumor in the esophagus at the time of resection. These data also suggest that 33%-36% patients survive when pCR is achieved than otherwise.
A review [
46] of studies analyzing preoperative chemoradiation for esophageal carcinoma showed no significant difference in pCR between adenocarcinoma and SCC. Patients with esophageal adenocarcinoma require a combined higher chemoradiation dose than SCC patients to achieve pCR. However, when the tumor responds, the pCR rate more rapidly increases with increased chemoradiotherapy dosage in adenocarcinoma patients than in SCC patients.
This result indicates that pCR is a powerful early predictor of the efficacy of neoadjuvant treatments for patients with esophageal and gastroesophageal cancer. Although pCR is more favorable for survival, the locoregional recurrence in the pCR group was lower than that in the non-pCR group without any statistical difference (13% vs. 31%;
P = 0.095) [
47]. Achieving pCR is not equal to cure or complete locoregional disease control.
Prediction markers
A new era of individualized treatment has arrived, with medical therapies tailored to individual patients based on their own gene signatures. Schauer
et al. [
48] noted significant differences between the gene profiles of responders and non-responders to nCRT for esophageal cancer. Many trials and reviews report promising predictive markers, although these results require further exploration by more well-designed multicenter trials.
Among a vast number of potential predictive markers, the most promising is 18-fluorodeoxyglucose positron emission tomography (
18F-FDG PET) imaging. Several studies [
49,
50] that evaluated tumor response with FDG PET at the completion of neoadjuvant therapy have yielded encouraging results. Mamede
et al. designed a study [
49] showing that FDG PET tumor segmentation-derived indices of metabolic activity play a definite role in evaluating response to nCRT and progression-free survival in esophageal cancer patients. Percentage changes in metabolic variables before and after therapy>32.3% and standard uptake value (SUV) before therapy>5.5 proved to be reliable predictors of pathologic response. The mean value of SUV after therapy>3.55 and the maximum SUV after therapy>4.35 were the best predictors of disease progression during follow-up, with the latter having the best prognostic value. Tan
et al. [
50] showed that spatial-temporal
18F-FDG PET/computed tomography (CT) features were useful in predicting pathologic tumor response to nCRT in esophageal cancer. Compared with SUV intensity features and tumor size features, changes were more predictive than pre- or post-CRT assessment alone.
However, results of the prospective multi-center SAKK 75/02 trial [
51] showed that for an individual patient,<40% decrease in FDG tumor uptake after induction chemotherapy predicted histopathological non-response after CRT completion, with a sensitivity of 68% and a specificity of 52% (positive predictive value, 58%; negative predictive value, 63%).
18F-FDG PET/CT did not predict non-response after induction chemotherapy with sufficient clinical accuracy to justify withdrawal of subsequent CRT and selection of patients to proceed directly to surgery. Reviewers also have different opinions [
52,
53]. Thus, more well-designed multicenter trials and standardizations from international centers are required to further explore the predictive value of
18F-FDG PET/CT in the neoadjuvant treatment of esophageal cancer.
Many other potential predictive markers include
18F-FLT PET/CT imaging [
54], three-dimensional (3D) CT volumetry [
55], and immunohistochemical analysis [
56,
57]. They all require more well-designed clinical trials to confirm their predicting roles in esophageal cancer treatment.
Ongoing trials
Clinical trials of different phases of nCRT are ongoing. Buduhan is conducting a phase II/III trial (NCT01404156) to compare neoadjuvant chemotherapy with nCRT in patients with surgically resectable esophageal cancer. In China, Sun Yat-sen University is recruiting participants for a phase-III multicenter randomized controlled study of nCRT followed by surgery vs. surgery for locally advanced ESCC. New chemotherapeutics and targeted drugs such as pemetrexed, gefitinib, cetuximab, panitumumab, and bevacizumab are research hotspots of ongoing clinical trials.
Future directions
The prognosis for patients with esophageal cancer remains poor despite recent advances in multimodality therapies. Trimodality based on nCRT combined with surgery plays an important role in managing locally advanced esophageal cancer. The available data suggest that nCRT may modestly improve outcomes in patients who are candidates for surgery. Current unresolved issues are still under study. Large, well-designed, multi-center trials are required to further establish the role of nCRT. Newer chemotherapeutic or targeted agents combined with novel technologies such as 3D conformal therapy, intensity-modulated radiation therapy, and image-guided radiotherapy have the potential to form a new paradigm for managing esophageal cancer. Local failure remains a significant problem for patients without surgery. Determining whether surgery is necessary after nCRT and identifying the subsets of patients more likely to benefit from the addition of surgery remains a clinical challenge. Moreover, the identification of molecular prognostic markers that allows individualized treatments among patients is of clinical interest that deserves immediate attention.
Acknowledgements
This work was supported by National Natural Science Foundation of China (Grant No. 81101700).
Compliance with ethics guidelines
Jing Liu, JinboYue, Ligang Xing and Jinming Yu declare that they have no conflict of interest. This manuscript is a review article and does not involve a research protocol requiring approval by the relevant institutional review board or ethics committee.
Higher Education Press and Springer-Verlag Berlin Heidelberg