The impact of taxane-based preoperative chemotherapy in gastroesophageal signet ring cell adenocarcinomas

The benefit of preoperative chemotherapy in resectable gastroesophageal adenocarcinomas was not observed in signet ring cell subtype. However, the potential interest of taxane-based preoperative chemotherapy on this subtype is still an unresolved issue. Nineteen patients with localized signet ring cell adenocarcinomas received taxane-based regimens, and 17 patients underwent surgery. Complete resection was achieved in 80 %, and median overall survival was 40.8 months (95 % confidence interval (CI), 20.2—not reached). Even though one patient achieved a complete pathological response, seven patients had an upstaging of their tumors at surgery. The potential benefits of taxane-based chemotherapy seem to be limited to a reduced number of patients. Electronic supplementary material The online version of this article (doi:10.1186/s13045-015-0148-y) contains supplementary material, which is available to authorized users.


Background
Signet ring cell (SRC) adenocarcinoma is a particular histological subtype of gastroesophageal adenocarcinomas (GEA) displaying a worse prognosis [1]. Even though the perioperative chemotherapy (PCT) in resectable GEA demonstrated a significant benefit in terms of overall survival (OS) compared to surgery alone [2,3], this benefit seems to be limited to non-SRC histology [4]. This observation prompted physicians to perform surgery without preoperative chemotherapy in SRC GEA patients with a resectable disease.
Taxanes are potent microtubule-stabilizing agents with demonstrated antitumor activity in advanced GEA and with encouraging results in resectable GEA, as reported in several phase II trials [5][6][7][8][9][10]. The potential interest of taxane-based PCT on SRC GEA is still an unresolved issue.

Results
Between January 2005 and December 2012, 19 patients with localized SRC GEA received taxane-based PCT from six French hospitals. (Additional file 1) Patients' median age was 64 years (range, 41-81 years). The majority of tumors (58 %) were located in the stomach and were predominantly stage III (42 %) and II (42 %) ( Table 1).
Seventeen patients (89 %) underwent surgery. One patient presented an unexpected death (cardiac failure) after three DCF cycles and before surgery, and another patient refused surgery after eight PET cycles. Total gastrectomy was performed in eight patients (47 %) and esophagogastrectomy via abdominal and right thoracic approaches (Lewis-Santi) in seven patients (41 %). Postoperative adverse events were observed in three patients with favorable recovery (Table 1).
All 17 patients who underwent surgery had a curativeintent resection. Pathological information about surgical margins was available in 15 patients, and the pathological complete resection (R0) was achieved in 12 patients (80 %). One patient presented a complete pathological response (pCR). This patient had a T2 disease with lymph node enlargement at diagnosis. In seven patients, more advanced disease was found at surgery compared to initial staging. Two patients presented intraoperative peritoneal metastases, and five patients had T4 disease (Table 1).
After a median follow up of 26.

Conclusion
Even though our study has obvious limitations as a retrospective analysis and regarding the limited number of patients, this is the largest cohort of SRC GEA patients   [4]. Future efforts should be focused on developing predictive biomarkers to identify SRC GEA patients potentially sensitive to taxanes.

Future perspectives
Targeted agents have shown promising results in advanced GEA. [11] Among them, trastuzumab (in HER2 positive patients) and ramucirumab have been approved in advanced GEA. However, most SRC GEAs are HER2 negative, and ramucirumab, an antiangiogenic mAb selectively targeting VEGFR2, will hardly be developed in perioperative setting due to negative experience of bevacizumab in gastrointestinal adenocarcinomas [12,13]. Among novel molecules in development in GEA, checkpoint inhibitors are probably the most promising. Pembrolizumab, an antiPD1 mAb was administered as monotherapy in 39 GEA patients with PD-L1 expression. Most patients have received ≥2 prior chemotherapies. An encouraging overall response rate of 22 % and the 6-month OS rate of 69 % were observed [14]. The expression of PD-L1 in SRC GEA is present in about 23 %, and a growing body of evidence suggests that taxane induces immunogenic cell death sustaining the potential interest to combine taxane and antiPD1 in clinical trials including SRC GEA patients [15,16].
Pembrolizumab and other checkpoint inhibitors should be evaluated in prospective preoperative trials in GEA patients including SRC histology, probably in association with taxane-based chemotherapy. Future exhaustive molecular analysis in SRC GEA is needed to find targets for novel molecules in this chemorefractory disease.

Competing interest
The authors declare that they have no competing interests.
Authors' contributions SK conceived the study, participated in its design, acquisition and interpretation of data, and coordination, and helped to draft the manuscript. FFi participated in study design, acquisition of data and statistical analysis, and helped to draft the manuscript. SPB participated in statistical analysis and helped to draft manuscript. FG participated in acquisition. ZL participated in acquisition of data. MJ participated in acquisition of data. FFe participated in acquisition of data. FB participated in its design and statistical analysis and helped to draft the manuscript. CM participated in acquisition and interpretation of data and helped to draft the manuscript. CB conceived the study, participated in its design and interpretation of data, and helped to draft the manuscript. All authors read and approved the final manuscript.