Fig. 3.1
American Joint Committee on Cancer 2010 esophageal cancer staging system
3.5 Prognostic Factors
- (a)
According to SEER Data, the 5-year survival of patients with esophageal cancer is approximately 20 %, and patients with esophageal cancers with lymph node-negative disease, lymph node metastases, and systemic metastases have a 5-year survival rate of 40 %, 21 %, and 4 %, respectively.
- (b)
Resection status, age, and histologic subtype have been prognostic for patients undergoing surgery alone [23]:
- (i)
Survival data from the Intergroup 0113 trial, which randomized esophageal cancer patients to surgery or neoadjuvant cisplatin and 5-FU followed by surgery, reported 5-year survival of patients undergoing an R1 or R2 resection that was significantly inferior to those patients with an R0 resection [24].
- (ii)
The CROSS study, which randomized patients to neoadjuvant chemoradiation followed by surgery or surgery only, showed that there was a near doubling of overall survival for patients with squamous cell carcinoma versus adenocarcinoma [25].
- (i)
- (c)
Tumor size has also been reported to be prognostic for both adenocarcinoma and squamous cell carcinoma:
- (i)
In one series 5-year survival decreased from 77 to 23 % for patients with resected squamous cell carcinomas measuring less than 1 cm compared to those greater than 3 cm [26].
- (ii)
Patients with adenocarcinomas greater than 2 cm have also been shown to have significantly worse 5-year survival compared to patients with tumors <2 cm [27].
- (i)
3.6 Management
- (a)
Surgery alone
- (i)
Surgery for esophageal cancer often involves a subtotal or total esophagectomy, via a transthoracic or transhiatal approach, with nodal dissection:
- (i)
It has been suggested that exposure of the chest cavity with a transthoracic approach can facilitate a more complete resection and therefore improved disease-related outcomes.
- (ii)
The question of optimal surgical approach was examined in a Dutch trial which randomized 220 patients to transthoracic or transhiatal resection [28]:
- 1.
Although the rate of locoregional recurrence was similar between the two surgical approaches (31 % in the transthoracic arm versus 32 % in the transhiatal arm), there appeared to be improved survival at a median follow-up of 4.7 years with the transthoracic approach (40 % vs. 30 %, p = 0.012).
- 2.
Higher rates of perioperative morbidity, pulmonary complications, and lengths of hospital stays were seen in patients undergoing resection by the transthoracic approach.
- 1.
- (i)
- (ii)
Independent of surgical technique, local recurrence rates range from 32 to 45 % in randomized trials containing a surgery-alone arm (Table 3.1), providing a rationale for multimodality treatment of this malignancy.Table 3.1Comparison of surgery alone arms in randomized studies of surgery with or without preoperative chemoradiationTrialYearPatients, totalPatients, surgicalMedian survival (months)2-year survival3-year survivalWalsh et al. [60]1996110551126 %6 %Urba et al. [41]20011005018NA15 %Bosset et al. [38]19972821391940 %35 %Kelsen et al. [36]19984402271637 %23 %MRC [61]20028024021334 %NABurmeister et al. [62]200525612819NA31 %Van Hagen et al. [42]201036618824NA48 %Mariette et al. [39]20141959844NANA
- (i)
- (b)
Postoperative therapy
- (i)
One of the advantages of adjuvant therapy is that the pathological stage of the malignancy is known; thus, patients with either early stage or metastatic disease, who may not benefit from adjuvant therapy, can be identified.
- (ii)
Pathologic staging and knowledge of surgical findings are helpful in radiation therapy planning.
- (iii)
Studies investigating the efficacy of adjuvant radiation therapy following surgery have not consistently demonstrated an improvement in either local control or survival:
- (i)
French investigators
- 1.
Randomized 221 patients with squamous cell carcinoma of the mid- to distal esophagus to either resection, with or without postoperative radiation therapy alone.
- 2.
Irradiated patients received 45–55 Gy within 3 months of surgery [29].
- 3.
There was no improvement in survival for patients randomized to adjuvant radiation therapy even with a reduction in local recurrence rates from 35 to 10 % [29].
- 1.
- (ii)
Hong Kong investigators
- 1.
Evaluated the outcome of 130 patients undergoing surgery and adjuvant radiation therapy alone versus surgery only in patients undergoing curative or palliative resection [30].
- 2.
Similar to the French trial, local recurrence was decreased (31 % vs. 15 %, p = 0.06) in the radiation group.
- 3.
Median survival, however, was significantly worse in the adjuvant group (median OS 8.7 vs. 15.2 months, p = 0.02), possibly due to morbidity associated with the large dose-per-fraction of 3.5 Gy and high overall dose of 49 Gy in patients treated with curative intent.
- 1.
- (iii)
Chinese investigators
- 1.
Study of 549 patients that reported a near doubling of survival at 5 years for lymph node-positive patients receiving adjuvant radiation therapy versus those having surgery alone (17.6–34.1 %) [31].
- 2.
As expected, patients with three or more involved lymph nodes had worse 5-year survival (14.4 %) compared to patients with 1–2 (30.6 %) or no lymph nodes involved (58.1 %).
- 1.
- (i)
- (iv)
Results of the Intergroup 0116 study [32] support an approach of combined chemotherapy and radiation therapy following resection of gastroesophageal junction (GEJ) adenocarcinomas:
- (i)
This study included patients with gastric or GEJ adenocarcinomas undergoing an R0 resection.
- (ii)
Five hundred fifty-six patients were randomized to no adjuvant therapy versus adjuvant therapy with 5-FU and leucovorin before, during, and after radiation therapy.
- (iii)
With a median follow-up greater than 10 years, the HR for survival was 1.32 favoring postoperative chemoradiation [33].
- (iv)
This benefit was observed across all stages and tumor locations.
- (i)
- (v)
Therefore, adjuvant treatment for GEJ adenocarcinomas is advised for resected T2–T4 N0 or any node-positive patients not receiving neoadjuvant therapy.
- (i)
- (c)
Preoperative therapy
- (i)
Preoperative treatment with either radiation or chemotherapy has a number of potential advantages:
- (i)
For radiation specifically, preoperative treatment often employs smaller radiation fields with less treatment-related morbidity compared to postoperative treatment.
- (ii)
Resection of the treated esophagus may also limit long-term complications as one of the primary tissues at risk, the esophagus itself, is removed.
- (iii)
Neoadjuvant chemotherapy may allow for elimination of micrometastatic disease and determination of tumor chemosensitivity.
- (iv)
Overall, an increased likelihood of resection due to downstaging and avoidance of surgery in patients with progression through treatment underscore the rationale for preoperative therapy.
- (v)
Preoperative therapy is associted with higher rates of compliance and ability to deliver intended therapy as compared to the adjuvant setting.
- (i)
- (ii)
Preoperative chemotherapy
- (i)
Similar to the conflicting results of adjuvant radiation therapy (described previously), the outcomes of randomized trials of preoperative chemotherapy alone have not consistently shown a survival benefit:
- 1.
Medical Research Council (MRC) OEO2 trial
- (a)
Largest trial including 802 patients with adenocarcinoma or squamous cell carcinoma of the esophagus randomized to cisplatin and 5-FU for two cycles prior to surgery versus surgery alone [34].
- (b)
Long-term follow-up at 6 years showed significantly improved 5-year overall survival in the preoperative chemotherapy arm (23 %) versus the surgery-alone (17 %) arm [35].
- (a)
- 2.
Intergroup 0113 trial
- (a)
Four hundred forty patients received either cisplatin with 5-FU before and following resection or resection alone.
- (b)
- (c)
Potential caveats to this study include that only approximately 60 % of patients in either arm underwent an R0 resection, and in patients undergoing R1 resection, the only long-term survivors received adjuvant radiation therapy [24].
- (a)
- 3.
MAGIC trial
- (a)
Perioperative combination of epirubicin, cisplatin, and 5-FU (ECF) was evaluated [37].
- (b)
Although designed for patients with gastric cancer, the study eligibility was later expanded to include patients with distal esophageal and GEJ cancers, and ultimately one-fourth of patients accrued on this study had esophageal or GEJ tumors.
- (c)
There were no pathologic complete responders to the neoadjuvant chemotherapy component.
- (d)
The 5-year survival was significantly improved for patients randomized to perioperative chemotherapy (36 %) compared to patients undergoing surgery only (23 %, p = 0.009). This survival benefit was seen in all primary sites – esophageal, GEJ, or stomach.
- (a)
- 4.
Results of the phase III studies of preoperative chemotherapy are summarized in Table 3.2.Table 3.2Results of preoperative chemotherapy vs. surgery-alone phase III trialsStudyMedian F/U (years)PathArmsNumber of patientspCR2-yearsurvivalSurvival differenceMRC6.0SCC+adeno5-FU-CDDP/surgsurg4004024 %–43 %34 %p = 0.004Kelsen et al. [36]Intergroup4.6SCC+adeno5-FU-CDDP/surgsurg2132272.5 %–23 % (3 years)26 % (3 years)NSCunningham et al. [37]MRC4.0AdenoEPI-CDDP-5-FU/surgsurg2502530 %–36 %23 %p = 0.009
- 1.
- (i)
- (iii)
Preoperative chemoradiation:
- (i)
Preoperative radiation in addition to chemotherapy has been investigated given the limited complete pathological response rate of the primary tumor and discrepancy in survival outcomes with chemotherapy alone:
- 1.
EORTC (European Organisation for Research and Treatment of Cancer) trial:
- (a)
Randomized 282 patients with squamous cell carcinoma.
- (b)
Demonstrated a median survival of 18.6 months with or without neoadjuvant chemoradiation, albeit with improvement in disease-free survival and cancer-related deaths with the use of neoadjuvant therapy [38].
- (c)
In this study, cisplatin alone was administered concurrently with 37 Gy in a split course at 3.7 Gy per fraction.
- (d)
Postoperative mortality was worse in the patients randomized to preoperative chemoradiation (12 %) versus surgery alone (4 %). This has been hypothesized to be due to the increased fraction size and may potentially account for the lack of overall survival benefit seen [38].
- (a)
- 2.
FFCD (La Fédération Francophone de Cancérologie Digestive) 9901 trial:
- (a)
Randomized 195 patients with clinically staged I–II squamous cell or adenocarcinoma.
- (b)
Stopped early due to crossing a prespecified boundary for futility in terms of improved survival [39].
- (c)
Although the pathologic complete response rate (33.3 %) was high, there was no difference in R0 resection rate of 93.8 % in the chemoradiation group versus 92.1 % in the surgery group.
- (d)
Postoperative mortality was significantly worse with neoadjuvant treatment (11.1 % vs. 3.4 %), with a 3-year overall survival of 47.5 % in the neoadjuvant group compared to 53 % with surgery alone, potentially negating any treatment-related survival benefit.
- (a)
- 3.
CALGB (Cancer and Leukemia Group B) 9781 trial:
- (a)
Randomized 56 patients to 50.4 Gy with concurrent cisplatin/5-FU and surgery or surgery only.
- (b)
The 5-year overall survival was 16 % in the surgery-alone arm compared to 39 % with neoadjuvant chemoradiation [40].
- (a)
- 4.
University of Michigan:
- (a)
One hundred patients with either adenocarcinoma or squamous cell carcinoma.
- (b)
Three-year survival was improved from 16 to 30 % with the addition of preoperative radiation with 5-FU, cisplatin, and vinblastine to surgery alone, although this did not reach statistical significance [41].
- (a)
- 5.
CROSS (Chemoradiotherapy for Oesophageal Cancer Followed by Surgery Study):
- (a)
Largest study of combined modality therapy, randomizing 368 patients to 41.4 Gy with weekly paclitaxel (50 mg/m2) and carboplatin (AUC = 2) followed by surgery versus resection alone [42].
- (b)
Pathologic complete response was seen in 47/161 patients (29 %) of the chemoradiation group.
- (c)
148/161 of neoadjuvantly treated patients (92 %) underwent R0 resection versus 69 % in the surgery-alone group. Similarly, locoregional failure rates were significantly lower in the chemoradiation group versus the surgery-alone group (22 vs. 38 %, p < 0.0001) [25].
- (d)
With a median follow-up of 84.1 months in surviving patients, an overall survival benefit to neoadjuvant chemoradiation (median 48.6 vs. 24 months) was reported [25]. This benefit was greater in patients with squamous cell carcinoma (median survival 81.6 months vs. 21.1 months), as compared to adenocarcinoma (median survival 43.2 vs. 27.1 months).
- (e)
Preoperative chemoradiation did not increase the toxicity of surgery as in hospital mortality was 4 % in each group, and rates of anastomotic leak (30 % vs 22 %) and mediastinitis (6 % vs. 3 %) were worse in the surgery-alone group [42].
- (a)
- 6.
Table 3.3 summarizes the results of the prospective phase III randomized trials evaluating the role of preoperative chemoradiation.Table 3.3Results of preoperative combined chemoradiation vs. surgery-alone phase III trialsStudyMedian F/U (years)PathArmsNumber of patientspCR3-year survivalSurvival differenceUrba et al. [41]Michigan8.2SCC+adeno5-FU-CDDP-Vinb/45 Gy/surgsurg505028 %–30 %16 %p = 0.15Bosset et al. [38]EORTC4.6SCC
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
Get Clinical Tree app for offline access
- 1.
- (i)
- (i)