Recent advances in preoperative management of esophageal adenocarcinoma

Esophageal cancer is an aggressive malignancy with increasing incidence, and the prognosis of patients treated by surgery alone remains dismal. Preoperative treatment can modestly prolong overall survival. Preoperative chemotherapy or chemoradiation is the standard of care for resectable esophageal cancer (greater than clinical stage I and less than clinical stage IV). One of the challenges is to predict complete response in the surgical specimen from preoperative therapy and to avoid surgery in some patients but also predict ineffectiveness of preoperative therapy if the tumor is resistant and avoid such therapies altogether. In-depth understanding of the molecular biology could lead to personalized therapy, and in the future, clinical trials designed according to molecular features are expected. Here, we summarize preoperative treatment for esophageal adenocarcinoma and their potential.


Introduction
Esophageal cancer (EC) is estimated to be the eighth most common cause of cancer in the world (456,000 cases) and the sixth most common cause of cancer death (400,000 deaths) 1 . EC has two common histologic types: adenocarcinoma (EAC) and squamous cell carcinoma (ESCC). EAC is becoming prevalent worldwide, especially in North America and Western Europe 2 . Esophagectomy is the most effective treatment for loco-regional control, but the 5-year survival rate after esophagectomy for locally advanced EC without preoperative treatment is less than 30% 3 . For early-stage EC, endoscopical resection or esophagectomy without preoperative therapy is one of the available options 4 . For the metastatic EAC, two-drug cytotoxic regimens, mainly a combination of a fluoropyrimidine and a platinum compound is recommended, and if EAC overexpresses HER2, trastuzumab should be added to chemotherapy 4 . In case of locally advanced EC, despite radical resection, local-regional and distant recurrence develop in 33% and 20% of patients after resection, respectively 5 . The preexisting occult micrometastases or unresected occult local disease is responsible for relapses. Interestingly, in one study, at the time of operation, 88% of patients with EC were already found to have micro-metastases in rib marrow aspirated during esophagectomy 6 . To overcome relapses after surgery, preoperative or postoperative treatments have been developed 4 . Importantly, preoperative therapy can modestly prolong overall survival (OS) and increase the R0 resection rate. R0 resection is associated with a longer survival 7 . Moreover, if preoperative therapy leads to a pathological complete response (pCR), longer OS may be expected 8,9 . Table 1 summarizes the preoperative therapy trials for EC conducted so far. Finally, in the Fédération Nationales des Centres de Lutte Contre le Cancer/Fédération Francophone de Cancérologie Digestive (FNCLCC/FFCD) trial, 224 patients were randomly assigned to two treatment groups: 113 to surgery plus perioperative chemotherapy (two or three preoperative and three or four postoperative cycles of CF) and 111 to surgery alone 13 . In this trial, 75% of patients had EAC. Compared with the surgery-alone group, the perioperative chemotherapy group had had a favorable OS (5-year rate: 38% versus 24%; HR 0.69; 95% CI 0.50 to 0.95; P = 0.02). Moreover, perioperative chemotherapy significantly increased the R0 resection rate (84% versus 73%; P = 0.04). These trial results were considered acceptable, and perioperative chemotherapy became standard therapy in Europe. Recently, the MRC-OEO5 trial compared two chemotherapy regimens: two cycles of FP and four cycles of ECX (epirubicin, cisplatin, and capecitabine) 14 . The ECX group had a higher R0 resection rate and pCR; however, there was no OS benefit for ECF compared with FP (3-year rate: 42% versus 39%; HR 0.92; 95% CI 0.79 to 1.08; P = 0.30). Furthermore, chemotherapy toxicity was higher in the ECX group 14 . This trial suggests that preoperative chemotherapy with more drugs and longer duration is not worthwhile and the addition of epirubicin does not provide any advantage.
The first and only study conducted in the US was the RTOG trial 8911 (USA Intergroup 113), which demonstrated no advantage from the addition of preoperative chemotherapy to surgery 15,16 . Thus, enthusiasm for preoperative chemotherapy has been low in the US and preoperative chemoradiation has been favored.

Preoperative chemoradiation
A prior meta-analysis proposed that preoperative chemoradiation may be beneficial; however, meta-analyses are only hypothesis-generating 17

Chemotherapy and chemoradiation
A recent meta-analysis compared preoperative chemoradiotherapy (n = 1,078) and chemotherapy (n = 1,141) for both EAC and ESCC, showing better OS of preoperative chemoradiation but not to a significant degree (HR 0.88; 95% CI 0.76 to 1.01; P = 0.07) 17 . To date, three randomized trials have compared preoperative chemoradiotherapy with chemotherapy, but none showed a benefit in OS of preoperative chemoradiation. The Preoperative Chemotherapy or Radiochemotherapy in Esophago-gastric Adenocarcinoma (POET) trial failed to recruit a sufficient number to document an OS advantage for preoperative chemoradiation, and the trial had to be terminated early 22 . The pCR rate was higher with preoperative chemoradiation compared with preoperative chemotherapy (15.6% versus 2.0%) 22 . Another phase II trial also showed no benefit from preoperative chemoradiation for patients with EAC; the median OS was 32 months from preoperative chemoradiation compared with 29 months from preoperative chemotherapy (P = 0.83) 23 . Another trial compared neoadjuvant chemotherapy with chemoradiotherapy in resectable cancer of the esophagus and gastric cardia patients (n = 181 with 73% EAC), and although chemoradiation resulted in a higher pCR rate (28% versus 9%), higher R0 resection rate (87% versus 74%), and a lower rate of lymph nodal metastases (35% versus 62%), there was no OS benefit (3-year OS: 49% versus 47%; P = 0.77) 24 .
Our group retrospectively reviewed sequential phase II/III trials performed at the University of Texas M.D. Anderson Cancer Center, showing that compared with preoperative chemotherapy, preoperative chemoradiation exhibited a longer OS (P = 0.046) and disease-free survival (P = 0.015) and the higher pCR rate (P <0.001) 25 .

Induction chemotherapy followed by preoperative chemoradiation
Our group had proposed a strategy of induction chemotherapy before preoperative chemoradiation 26 . To document whether there is any benefit to the addition of induction chemotherapy, we reported a randomized phase II trial that compared induction chemotherapy followed by preoperative chemoradiation with preoperative chemoradiation only. One hundred twenty-six patients with localized EAC were randomly assigned to one of two groups. The median OS rates with and without induction chemotherapy were 43.6 and 45.6 months, respectively. The pCR rates were 13% in the no induction chemotherapy group and 26% in the induction chemotherapy group (P = 0.094), concluding that induction chemotherapy did not appear to benefit these patients 27 . However, subgroup analysis showed that induction chemotherapy had a considerable benefit for only those patients who had a well-moderate differentiated tumor 28 .

Preoperative treatment with molecular targeting drug
Currently, there is no evidence that the addition of a targeted drug benefits to patients with localized EC. In patients with advanced EAC or gastric cancer, the ToGA (Trastuzumab for Gastric Cancer) trial showed that the addition of HER2 inhibitor, trastuzumab, to chemotherapy had modest benefit 29 . However, the benefit of trastuzumab in the neoadjuvant setting is not established. In Japan, a phase II trial is evaluating S-1 plus cisplatin with or without trastuzumab in the neoadjuvant setting for HER2-positive gastric or esophagogastric junction adenocarcinoma 30 . Epidermal growth factor receptor (EGFR) inhibitors have been evaluated in this setting 31-34 on the basis of a tantalizing phase II study which added cetuximab to chemoradiation and produced a pCR rate of 27% 33 . However, two phase III trials that added an EGFR inhibitor to dCRT (definitive chemoradiation) failed to show survival benefit 31,35 . In addition, bevacizumab or erlotinib was evaluated with preoperative chemoradiation but did not demonstrate survival benefit or improvement in the pCR rate 36 .

Future perspective
Approximately 25% of patients who undergo preoperative chemoradiation achieve a pCR 18 . If one could predict the possibility of pCR with a high level of certainty, then novel strategies to preserve the esophagus could be implemented. However, there are no useful clinical variables including positron emission tomographic (PET) changes and there are no reliable biomarkers for such a prediction at the moment. A clinical CR defined as endoscopic biopsies without cancer cells and PET scan with physiologic uptake provides an OS benefit 37 but does not correlate with pCR 38 . Therefore, we recommend that all patients eligible for surgery proceed to surgery after recovering from chemoradiation.
Metabolic imaging with 18-fluorodeoxy-glucose positron emission computed tomography was assessed in various circumstances. Our group reported that PET parameters could correlate with prolonged OS but could not predict pCR 39 . Interestingly, the Municon I and II trials evaluated treatment modifications according to PET responses for patients with EAC 40,41 . In the Municon II trial, PET non-responders from preoperative chemotherapy had chemoradiation before surgery; however, non-responders still had an unfavorable prognosis, suggesting that primary resistance cannot be overcome by simply changing therapy. Thus, in-depth study of the tumor biology is warranted and may contribute to personalize therapy.

Conclusions
This review described an understanding of preoperative therapy for EC. The benefits of preoperative treatment, and preoperative chemoradiation in particular, have been established. Currently, preoperative chemoradiation is preferred over preoperative chemotherapy in the US. Head-to-head comparison of preoperative chemotherapy versus preoperative chemoradiation is not completed as it is a subject of ongoing trials. A further challenge is to identify patients who are destined to achieve a pCR. CTCs or ctDNA might prove useful in surveillance after therapy and occasionally for selection of therapy.
Author contributions KH helped to conceive the study and prepared the first draft of the manuscript. DK and JA helped to conceive the study. All authors contributed to the preparation of the manuscript and were involved in the revision of the draft and have agreed to the final content.

Competing interests
The authors declare that they have no competing interests. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.