Annals of Oncology

Volume 34, Issue 2, February 2023, Pages 163-172

Original Article

Neoadjuvant chemoradiotherapy versus

neoadjuvant chemotherapy followed by
minimally invasive esophagectomy for
locally advanced esophageal squamous
cell carcinoma: a prospective
multicenter randomized clinical trial
H. Tang 1 2 †, H. Wang 1 2 †, Y. Fang 1 2 †, J.Y. Zhu 2 3 †, J. Yin 1 2, Y.X. Shen 1 2, Z.C. Zeng 2 3,
D.X. Jiang 2 4, Y.Y. Hou 2 4, M. Du 5, C.H. Lian 6, Q. Zhao 6, H.J. Jiang 7, L. Gong 8, Z.G. Li 9,
J. Liu 10, D.Y. Xie 11, W.F. Li 12, C. Chen 13, B. Zheng 13…L.J. Tan 1 2

Show more

Outline Share Cite

https://doi.org/10.1016/j.annonc.2022.10.508
Get rights and content

Under a Creative Commons license open access

Highlights

• The CMISG1701 trial assessed the safety and efficacy of nCRT

versus nCT followed by MIE for locally advanced bulky ESCC.

• The nCRT followed by MIE strategy could not improve

:
 • The nCRT followed by MIE strategy could not improve
survival significantly compared with the nCT strategy.

• The best strategy of neoadjuvant therapy for locally

advanced bulky ESCC remains a pending issue.

Background
Neoadjuvant therapy is recommended for locally advanced esophageal cancer,
but the optimal strategy remains unclear. We aimed to evaluate the safety and
efficacy of neoadjuvant chemoradiotherapy (nCRT) versus neoadjuvant
chemotherapy (nCT) followed by minimally invasive esophagectomy (MIE) for
locally advanced esophageal squamous cell carcinoma (ESCC).

Patients and methods

Eligible patients staged as cT3-4aN0-1M0 ESCC were randomly assigned (1 : 1)
to the nCRT or nCT group stratified by age, cN stage, and centers. The
chemotherapy, based on paclitaxel and cisplatin, was administered to both
groups, while concurrent radiotherapy was added for the nCRT group; then MIE
was carried out. The primary endpoint was 3-year overall survival. This study is
registered with ClinicalTrials.gov (NCT03001596).

Results
A total of 264 patients were eligible for the intention-to-treat analysis. By 30
November 2021, 121 deaths had occurred. The median follow-up was 43.9
months (interquartile range 36.6-49.3 months). The overall survival in the
intention-to-treat population was comparable between the nCRT and nCT
strategies [hazard ratio (HR) 0.82, 95% confidence interval (CI) 0.58-1.18; P =
0.28], with a 3-year survival rate of 64.1% (95% CI 56.4% to 72.9%) versus 54.9%
(95% CI 47.0% to 64.2%), respectively. There were also no differences in
progression-free survival (HR 0.83, 95% CI 0.59-1.16; P = 0.27) and recurrence-
free survival (HR 1.07, 95% CI 0.71-1.60; P = 0.75), although the pathological
complete response in the nCRT group (31/112, 27.7%) was significantly higher
:
 than that in the nCT group (3/104, 2.9%; P < 0.001). Besides, a trend of lower risk
of recurrence was observed in the nCRT group (P = 0.063), while the recurrence
pattern was similar (P = 0.802).

Conclusions
NCRT followed by MIE was not associated with significantly better overall
survival than nCT among patients with cT3-4aN0-1M0 ESCC. The results
underscore the pending issue of the best strategy of neoadjuvant therapy for
locally advanced bulky ESCC.

Previous Next

Key words
neoadjuvant chemoradiotherapy; neoadjuvant chemotherapy; esophageal
squamous cell carcinoma; minimally invasive esophagectomy; survival

Introduction
Esophageal cancer (EC) is the sixth leading cause of cancer deaths in the world,
and esophageal squamous cell carcinoma (ESCC) accounts for >80% of all
cases.1,2 As a result of obscure symptoms, most patients are diagnosed at an
advanced stage, and neoadjuvant therapy, both neoadjuvant chemoradiotherapy
(nCRT) and neoadjuvant chemotherapy (nCT), is recommended for these
patients because it achieved improved survival compared with surgery alone in
several studies.3, 4, 5 However, controversy still exists over which neoadjuvant
strategy is better and no agreement has been reached about the standard
strategy. At present, only three clinical trials are available that directly compare
nCRT with nCT for EC6, 7, 8, 9; notably, the cases in these studies are all or
predominantly cases of adenocarcinoma located in the distal esophagus or
esophagogastric junction, and therefore whether the results could be
generalized to patients with ESCC mainly located in the middle or upper one-
:
 third of esophagus remains to be validated. By contrast, it is reported that nCRT
was significantly associated with an increased risk of perioperative morbidity or
mortality for patients with ESCC,10,11 which may impose restrictions on the
application of nCRT. However, the esophagectomy carried out in these trials was
an open surgical procedure, which could cause huge trauma as the result of
severe postoperative complications. Recently, minimally invasive
esophagectomy (MIE) become the mainstream for ESCC worldwide as it has the
advantages of less trauma, quicker recovery, and equal oncologic survival12,13;
however, whether it could be beneficial for patients who underwent
neoadjuvant therapy has never been tested in any reported prospective
randomized clinical trials. Therefore, we take neoadjuvant therapy followed by
MIE as a treatment strategy in this trial to reduce trauma. We assumed that
nCRT may have better response than nCT for bulky tumor, because of the
radiation. Our retrospective data also showed that patients with locally
advanced bulky tumor (cT3-4aN0-1) who underwent nCRT followed by MIE had
similar mortality and morbidity as those who underwent nCT, as well as better
3-year overall survival.14 To prove our assumption and critically evaluate the
safety and efficacy of nCRT compared with nCT followed by MIE for locally
advanced cT3-4aN0-1M0 ESCC, the prospective, multicenter, randomized
clinical trial (Chinese MIE Interest Study Group, CMISG1701) was carried out.
Previously, we reported the short-term outcomes of the trial, showing that there
were no differences in the incidence and severity of all kinds of postoperative
complications, including pulmonary/cardiovascular-related complications,
anastomotic leak, recurrent nerve paralysis, among others. Here, we report the
primary endpoint of overall survival time for the first time.

Methods

Trial oversight
The CMISG1701 trial was a prospective, randomized, multicenter study carried
out in 10 high-volume hospitals in China from January 2017 to December 2018.
The study protocol is given in Supplementary File S1, available
at https://doi.org/10.1016/j.annonc.2022.10.508 , and approved by the Ethics
Committee of the Zhongshan Hospital (B2016-177R) and the institutional
:
 review board at each institution. The study was carried out in accordance with
the Declaration of Helsinki and the Good Clinical Practice Guideline. All patients
provided written informed consent before enrollment. This trial has been
registered and released in ClinicalTrials.gov (NCT03001596) and followed the
Consolidated Standards of Reporting Trials (CONSORT) reporting guideline. The
primary outcome was 3-year overall survival. The secondary endpoints were 3-
year progression-free survival, 3-year recurrence-free survival, postoperative
pathologic stage, postoperative complications, mortality, and quality of life.

Patients
Eligible patients were aged 18-75 years and had histologically confirmed ESCC
staged as cT3-4aN0-1M0 according to the International Union Against Cancer
Tumor, Node, Metastasis (TNM) Classification (8th edition). Patients also should
have adequate hematologic, kidney, liver, and pulmonary function as well as an
Eastern Cooperative Oncology Group (ECOG) performance status score of ≤2 to
tolerate the treatment. Besides, no history of other cancer or
radiotherapy/chemotherapy was required.

Randomization and masking

Eligible patients were centrally randomly assigned in a 1 : 1 ratio to either the
nCRT group or the nCT group stratified by age, clinical N stage, and trial centers.
Randomization was assigned by the computer-generated random system in the
Biomedical Statistics Center, Fudan University.

Treatment

nCRT
On days 1, 8, 15, and 22, paclitaxel (50 mg/m2) and cisplatin (25 mg/m2 of body
surface area) were administered intravenously. A total radiotherapy dose of 40
Gy was administered in 20 fractions of 2 Gy, 5 fractions per week, starting on
the first day of chemotherapy. Three-dimensional conformal radiotherapy of
involved field irradiation was applied to patients with EC. The clinical target
volume included the primary tumor plus a 3-cm craniocaudal and 0.7-1-cm
:
 lateral margin as well as potential metastatic lymph nodes plus a 0.7-1-cm
craniocaudal and lateral margin based on the enhanced thoracic computed
tomography (CT) or positron emission tomography (PET)/CT scan. The planning
target volume was defined as clinical target volume plus a 1-1.5-cm margin in
the craniocaudal direction and 1-cm margin in the lateral direction. The
planning target volume was the irradiation area. The radiation criteria were
established by all the centers, and they all had routine discussion to guarantee
the radiation quality. Besides, at the beginning of the trial, the first several
patients enrolled in every center were discussed together to have high-quality
assurance.

nCT
It consisted of two cycles of preoperative chemotherapy. The regimen was
intravenous paclitaxel (175 mg/m2) and cisplatin (75 mg/m2) on day 1. The
second cycle was initiated after 3 weeks on day 22.

MIE
After 4-8 weeks of neoadjuvant therapy, MIE would be carried out. The
procedure is referred in-depth in a previous article.15 In our study, all
participating centers had an experience of at least 100 MIEs and had completed
the blind revision of the surgery video. To achieve an accurate ypTNM stage, the
extent of lymphadenectomy demanded resecting radically. Dissected lymph
nodes were classified according to lymph node stations adopted by the Japanese
Classification.16,17 For lower and middle third tumors, extensive two-field
lymphadenectomy was carried out, and for upper third tumors, three-field
lymphadenectomy, including bilateral cervical lymph node dissection, was
mandatory.

Assessments
There were three assessment phases, including pretherapeutic, therapeutic, and
follow-up assessments. They all included a physical examination, demography,
discomfort symptoms, medical history, vital signs, body weight, and laboratory
tests. Specifically, the pretherapeutic evaluation also had upper endoscopy with
:
 biopsies, pulmonary function evaluation, endoscopic ultrasound of the neck,
and an enhanced CT scan of the thorax and abdomen. PET/CT was
recommended for every patient for additional improvement of cstage (before
neoadjuvant therapy). All aforementioned evaluation and scans should have
been finished within 2 weeks before therapy. During the therapeutic phase,
blood routine and biochemistry were examined weekly, which assessed the
toxicity. After 4-8 weeks of neoadjuvant therapy, re-evaluation was carried out,
and the ycstage (after neoadjuvant therapy) was confirmed by an enhanced CT
scan of the thorax and abdomen or combined with PET/CT. After discharge from
hospital, follow-up began. The first follow-up visit was carried out 1 month after
surgery. From then on, follow-up visits were carried out every 3 months in the
first 2 years and every 6 months from the third year until the end of follow-up
(at least 3 years). For all patients, follow-up assessment was carried out until the
end of the trial or death. The detailed examination items included standard
laboratory tests (blood routine and tumor biomarker), a CT scan of thorax, an
ultrasound of the neck, and abdomen endoscopy and PET/CT if necessary.

Sample size
The sample size calculations were based on overall survival. The 3-year overall
survival rate in a previous report was ∼77% in the nCRT group and 50% in the
nCT group, without differences in mortality.14 We assumed an increase in 3-year
overall survival of 20% in the nCRT group. Therefore, the total sample size was
calculated to be 264, which was based on the intention of showing a difference
in the primary endpoint of 20% between treatment groups with a type I error of
5% and a power of 90%, as well as a 15% dropout rate before surgery or loss to
follow-up according to Power Analysis and Sample Size (PASS) software. Thus
132 patients were enrolled in each group according to the 1 : 1 randomized
allocation.

Statistical analysis
Data were analyzed according to the intention-to-treat principle in all
randomized patients. Overall survival was defined as the time from
randomization to death of any causes. Progression-free survival was defined as
the time from randomization to the first recurrence/progression (local, regional
:
 or distant) or death. Recurrence-free survival was defined as the time from the
date of surgery to the date of first recurrence (local, regional, or distant) or
death. The last follow-up date was 30 November 2021, and the median follow-
up time was 43.9 months. We used Cox proportional hazard models for
univariate and multivariate analyses of factors with potential prognostic
relevance for survival. Associations between categorical variables were tested
with Fisher’s exact test and the chi-square test. All statistical analysis was
undertaken using SPSS, version 23.0 (IBM Corp.) and the significance level was
set at 0.05.

Results
From 1 January 2017 to 31 December 2019, 264 patients were enrolled in the
trial, all of whom were randomly assigned to the two treatment groups
(Figure 1). Baseline characteristics were well-balanced between the treatment
groups (Table 1). The flowchart of the trial is presented in Figure 1. A total of 111
patients (111/132, 84.1%) in the nCRT group received the planned
chemoradiotherapeutic regimen, whereas 118 patients (118/132, 89.4%) in the
nCT group completed the entire process as planned; no statistically significance
difference was found (P = 0.204). The adverse event during the neoadjuvant
phase and short-term surgical outcomes, including intraoperative results,
postoperative morbidity and mortality, from this trial have already been
published.18 There were no differences between the treatment groups in the
intraoperative parameters and in the frequency/severity of postoperative
complications, although grade ≥3 adverse events were significantly more
frequent in the nCRT group (P = 0.03).
:
 Download: Download high-res image (669KB)
Download: Download full-size image

Figure 1. Consolidated Standards of Reporting Trials (CONSORT) diagram.

ESCC, esophageal squamous cell carcinoma.

Table 1. Patient demographic characteristics

Characteristics nCRT (n = 132), n (%) nCT (n = 132), n (%) P value

Sex 0.293

Male 116 (87.9) 110 (83.3)

Female 16 (12.1) 22 (16.7)

:
 Age, years 0.802

≤60 54 (40.9) 52 (39.4)

≥61 78 (59.1) 80 (60.6)

ECOG PS 0.746

0 110 (83.3) 108 (81.8)

1 22 (16.7) 24 (18.2)

Tumor location 0.475

Upper 16 (12.1) 13 (9.8)

Middle 89 (67.4) 84 (63.6)

Lower 27 (20.5) 35 (26.5)

Clinical T stagea 0.589

cT3 95 (72) 91 (68.9)

cT4a 37 (28) 41 (31.1)

Clinical N stagea 0.322

cN0 37 (28) 30 (22.7)

cN1 95 (72) 102 (77.3)

nCRT, neoadjuvant chemoradiotherapy; nCT, neoadjuvant chemotherapy; ECOG PS,

Eastern Cooperative Oncology performance status.

a
Tumor stage classified according to the American Joint Committee on Cancer, 7th
Edition.

Pathology
The R0 resection rates were similar between the two treatment groups (109/112,
97.3%, versus 100/104, 96.2%; P = 0.92) as non-R0 (R1/2) was defined as finding a
vital tumor at ≤1 mm from the proximal, distal, or circumferential resection
:
 margin. The pathological complete response (pCR) was achieved in 31/112
(27.7%) patients in the nCRT group versus 3/104 (2.9%) in the nCT group (P <
0.001). Similarly, patients in the nCRT group had better primary tumor
regression grade (residual tumor, 0%: 40/112, 35.7%, versus 4/104, 3.8%; P <
0.001), fewer lymph nodes involved (ypN0: 74/112, 66.1%, versus 48/104, 46.2%;
P = 0.03), less lymphovascular/perineural invasion (12/112, 10.7%, versus 27/104,
26%; P = 0.004), and decreased ypTNM stage (stage I: 58/112, 51.8%, versus
21/104, 20.2%; P < 0.001) than those in the nCT group. A median of 20 and 24
lymph nodes were dissected in the nCRT and nCT groups, respectively (P =
0.001). Notably, among the patients with tumor regression grade 1 of primary
tumor, 9 (22.5%) and 1 (25%) had at least one metastatic lymph node in the nCRT
and nCT groups, respectively (Table 2).

Table 2. Pathologic outcome

Outcome nCRT group (n = 112), nCT group (n = 104), P

n (%) n (%) value

R0 resection 109 (97.3) 100 (96.2) 0.921

Histological response of primary <0.001

tumor

TRG1 (residual tumor 0%) 40 (35.7) 4 (3.8)

TRG2 (residual tumor 1%-10%) 31 (27.7) 10 (9.6)

TRG3 (residual tumor 11%-50%) 19 (17) 17 (16.3)

TRG4 (residual tumor >50%) 22 (19.6) 73 (70.2)

ypT stage <0.001

ypT0 40 (35.7) 4 (3.8)

ypT1 17 (15.2) 15 (14.4)

ypT2 23 (20.5) 23 (22.1)

ypT3 22 (19.6) 48 (46.2)

ypT4 10 (8.9) 14 (13.4)

:
 Lymph nodes involved 0.030

ypN0 74 (66.1) 48 (46.2)

ypN1 26 (23.2) 36 (34.6)

ypN2 9 (8.0) 14 (13.5)

ypN3 3 (2.7) 6 (5.8)

LVI + PNI 0.004

Negative 100 (89.3) 77 (74)

Positive 12 (10.7) 27 (26)

Lymph node harvested, median 20 24 0.001

ypStage <0.001

ypStage I 58 (51.8) 21 (20.2)

Including ypT0N0M0, pCR 31 (27.7) 3 (5.3) <0.001

ypStage II 11 (9.8) 21 (20.2)

ypStage III 34 (30.4) 49 (47.1)

ypStage IV 9 (8.0) 13 (12.5)

ypT0N+M0 9 (8.0) 1 (0.96) 0.013

LVI, lymphovascular invasion; nCT, neoadjuvant chemotherapy; nCRT, neoadjuvant

chemoradiotherapy; pCR, pathological complete response; PNI, perineural invasion;
TRG, tumor regression grade.

Survival
The median follow-up of the survivors was 45.0 months (interquartile range
37.1-50.4 months) in the nCRT group and 42.4 months (interquartile range 36.3-
48.7 months) in the nCT group. Median overall survival was not reached (NR)
[95% confidence interval (CI) 40.0-NR months] in the nCRT group, and was 43.2
months (95% CI 32.2-NR months) in the nCT group. The 3-year overall survival
rate was 64.1% (95% CI 56.4% to 72.9%) in the nCRT group and 54.9% (95% CI 47%
:
 to 64.2%) in the nCT group. Slightly improved survival was observed in the nCRT
group, but the difference did not reach statistical significance (hazard ratio [HR]
0.82; 95% CI 0.58-1.18; P = 0.28; Figure 2). Median progression-free survival was
46.5 months (95% CI 26.3-NR months) and 34.1 months (95% CI 18.1-NR months)
in the nCRT and nCT groups, respectively. The 3-year progression-free survival
rate was 54.3% (95% CI 46.3% to 63.6%) and 49.8% (95% CI 41.9% to 59.2%) in the
nCRT and nCT groups, respectively, and no difference was detected between the
two groups (HR 0.83; 95% CI 0.59-1.16; P = 0.27; Figure 3). Among the 209
patients undergoing R0 resection, the median recurrence-free survival was 49.2
months (95% CI 38.3-NR months) in the nCRT group, compared with 50.2
months (95% CI 30.2-NR months) in the nCT group, and the recurrence-free
survival was also comparable (HR 1.07; 95% CI 0.71-1.60; P = 0.75; Figure 3),
with 3-year survival rate of 57.7% (95% CI 49.1% to 67.8%) and 59.9% (95% CI 51%
to 70.4%) in the nCRT and nCT groups, respectively. Notably, these two curves
coincided with each other completely, and there was no survival difference at
all. In other words, there were no differences in survival time between the two
treatment strategies. We found that female sex, lower clinical T stage, and better
ECOG performance status tended to have a more favorable overall survival
compared with male sex, higher clinical T-stage, and worse ECOG performance
status (Table 3). To assess whether certain patient groups had an increased
likelihood of improved survival with nCT or nCRT, a Cox regression analysis with
adjustment for baseline variables was carried out. The results showed that
neither of the two treatment options seem to offer any advantage to a specific
group of patients as specified by their different baseline characteristics
(Figure 4).
:
 Download: Download high-res image (307KB)
Download: Download full-size image

Figure 2. Kaplan–Meier estimates of overall survival. CI, confidence interval;

HR, hazard ratio.

Download: Download high-res image (420KB)

Download: Download full-size image

Figure 3. Kaplan–Meier estimates of progression-free survival and

recurrence-free survival. CI, confidence interval; HR, hazard ratio.

Table 3. The association between pretreatment characteristics and overall

:
 survival

Characteristics n Univariate analysis Multivariate analysis

Crude hazard ratio P Adjusted hazard ratio P

(95% CI)a value (95% CI)b value

Sex 0.006 0.005

Male 226 1 1

Female 380 0.388 (0.197-0.766) 0.380 (0.192-0.750)

Age, years 0.523

≤60 106 1

≥61 158 1.127 (0.781-1.627)

ECOG PS 0.015 0.031

0 218 1 1

1 46 1.682 (1.108-2.555) 1.588 (1.044-2.417)

Tumor location 0.502

Upper/middle 202 1

Lower 62 1.15 (0.765-1.73)

cT stage 0.063 0.054

cT3 186 1 1

cT4a 78 0.701 (0.482-1.019) 1.449 (0.994-2.112)

cN stage 0.215

cN0 67 1

cN1 197 1.312 (0.854-2.014)

CI, confidence interval; ECOG PS, Eastern Cooperative Oncology performance status.

a
Crude hazard ratios and 95% CIs were obtained using univariate Cox proportional
:
 hazard regression models.

b
Adjusted hazard ratios and 95% CIs were obtained using multivariate Cox
proportional hazard regression models.

Download: Download high-res image (510KB)

Download: Download full-size image

Figure 4. Overall survival, subgroup analysis. Forest plot showing HR under the
univariate and multivariate analyses for death with 95% CI according to baseline
characteristics. CI, confidence interval; HR, hazard ratio; nCRT, neoadjuvant
chemoradiotherapy; nCT, neoadjuvant chemotherapy.

Recurrence and death

A total of 50 (37.9%) patients in the nCRT group and 65 (49.2%) in the nCT group
:
 experienced a recurrence (P = 0.063; Table 4), with a trend of lower risk of
recurrence in the nCRT group, although the recurrence pattern itself, including
regional, distant recurrence, or both, presented no significant differences (P =
0.802; Table 4).

Table 4. Recurrence pattern and causes of death

Pattern and cause of death nCRT group (n = nCT group (n = P

132), n (%) 132), n (%) value

Recurrence 50 (37.9) 65 (49.2) 0.063

a

Recurrence pattern 0.802

a

Regional/local 23 (17.4) 28 (21.2) 0.436

a

Distant 18 (13.6) 22 (16.7) 0.492

a

Both 9 (6.8) 15 (11.4) 0.199

a

Death 57 (43.2) 64 (48.5) 0.387

a

Causes of death 0.368

a

Esophageal cancer 41 (71.9) 53 (82.8) 0.123

a

Side-effect from 4 (7.0) 2 (3.1) 0.684

neoadjuvant/adjuvant therapy b

Postoperative complication-related 7 (12.3) 4 (6.2) 0.355

death b

Other disease 3 (5.3) 1 (1.6) 0.622

b
:
 Unknown 2 (3.5) 4 (6.2) 0.684
b

nCRT, neoadjuvant chemoradiotherapy; nCT, neoadjuvant chemotherapy.

a
Chi-square test for association.

b
Fisher’s exact test.

At the time of the intention-to-treat analysis, 57 (43.2%) patients in the nCRT

group and 64 (48.5%) patients in the nCT group died. Specifically, in the nCRT
group, 41 (71.9%) patients died of EC, 7 (12.3%) of complication-related reasons,
4 (7%) of neoadjuvant/adjuvant therapy-related side-effect, 3 (5.3%) of other
diseases, and 2 (3.5%) of unknown reasons. In the nCT group, 53 (82.8%), 4
(6.2%), 2 (3.1%), 1 (1.6%), and 4 (6.2%) died of the aforesaid reasons, respectively.
There was no significant difference between the two groups in causes of death
(P = 0.368; Table 4).

Discussion
This work is the first available well-designed multicenter randomized clinical
trial to directly compare overall survival as well as progression-free survival and
recurrence-free survival of nCRT versus nCT followed by MIE for locally
advanced resectable ESCC. We have shown that there was no difference in
survival and morbidity between those who received nCRT and those who
received nCT, although nCRT significantly increased pCR rate, reduced
metastatic lymph node numbers, and reduced ypTNM stage. Moreover, the
recurrence pattern and causes of death were similar in these two treatment
strategies. Therefore nCRT followed by MIE was not associated with significantly
better overall survival than nCT among patients with cT3-4aN0-1M0 ESCC. The
results underscore the pending issue of the best strategy of neoadjuvant therapy
for locally advanced bulky ESCC.
:
 In recent years, several clinical trials confirmed the role of neoadjuvant therapy
for patients with locally advanced EC. The CROSS trial demonstrated that nCRT
could significantly improve the prognosis of patients with EC in the cT1N1M0 or
cT2-3N0-1M0 stage compared with surgery alone, and the efficacy of nCRT was
more excellent especially for ESCC despite the small number.4 Then, in the
Neoadjuvant Chemoradiotherapy Followed by Surgery Versus Surgery Alone for
Locally Advanced Squamous Cell Carcinoma of the Esophagus (NEOCRTEC5010)
clinical trial it was further concluded that nCRT was a safe and efficient
treatment for patients with cT1-4N1M0 or cT4N0M0 ESCC with a large sample
size.5 As for nCT, in the JCOG9907 trial, the overall survival rate was improved
under the nCT strategy compared with adjuvant chemotherapy in patients with
clinical stage II-III (excluding cT4) ESCC.3 Besides, a three-arm phase III
randomized controlled trial (JCOG1109) is ongoing aiming at confirming the
best neoadjuvant treatment regimen for ESCC among three regimens of a
doublet chemotherapeutic regimen (two courses of cisplatin plus 5-
fluorouracil), a triplet chemotherapeutic regimen (three courses of docetaxel,
cisplatin plus 5-fluorouracil), and a chemoradiotherapy regimen (41.4 Gy/23
fractions with two courses of cisplatin plus 5-fluorouracil).19 Nevertheless, the
optimal modality for the treatment of locally advanced resectable ESCC remains
unclear as the previous clinical trials directly comparing nCRT with nCT mainly
enrolled patients with esophageal adenocarcinoma, and limited evidence was
provided for ESCC in these studies. The Neoadjuvant Chemotherapy Versus
Radio chemotherapy for Cancer of the Esophagus or Cardia (NeoRes) trial7 was
the only study that enrolled patients with ESCC, which accounted for 27% of the
whole group, and it showed that there was no statistical significance in overall
survival between the two treatments, although slightly improved survival was
seen in the ESCC subgroups,7 and the results were consistent with ours. Besides,
female sex, better ECOG performance, and lower clinical T stage were
independent favorable prognostic factors as previously described in the NeoRes
trial.20 It remained unclear why female sex was an independent factor,20 and
further exploitation of this scenario might give more insights into the
pathogenesis of the disease.

Although there is no clear definition for locally advanced tumor of ESCC, the
reasons why we just focused on patients with ESCC staged as cT3-4aN0-1M0 in
:
 this study were as follows: first, lymph nodes metastasis in ESCC is frequent and
extensive, and it is difficult to evaluate lymph node status accurately via
preoperative examination, so clinically suspicious N2 stage with multiple
stations involved cannot be included in the study as it may be at an advanced
stage with potentially distant metastasis. Second, for cT1b-2N+ tumor, although
it is always regarded as another type of locally advanced tumor, radical
resection is easily achieved without preoperative radiation. NCRT may add more
value in patients with bulky tumor staged as cT3-4a, possibly combined with
limited lymph node metastasis (N0-1), if compared with nCT only. Third, the
possible rationale for neoadjuvant therapy to ESCC has not been clarified yet.
NCT may lead to strong systemic chemo-responses both for primary tumor,
regional lymph nodes and micro systemic metastasis as a result of improved
survival time. While nCRT has great superiority in tumor regression, it could
also shrink the bulky tumor significantly, thus improving R0 resection rates as a
consequence of possibly prolonged survival time.7,21 Initially, we assumed that
nCRT might have better survival than nCT for this special candidate, which was
supported by our retrospective data.14 However, there is a lack of a
corresponding significant advantage in overall survival despite the downstaging
effect.7,22 A possible explanation for the lack of survival benefit despite better
tumor response could be that the addition of radiotherapy may not increase
local tumor control when thoracoscopic esophagectomy with extensive lymph
node dissection was used and high R0 rate was achieved. Under such a surgical
method, local regional residual tumors were eliminated radically as the
recurrence pattern was similar between the two treatment strategies. There was
also a clear trend toward improved survival in the nCRT group, which could be
because the no-surgery-patients who underwent nCRT had better prognosis
than those who underwent nCT, whereas those who underwent surgery in the
two groups had comparable prognosis (recurrence-free survival), as patients
with cT3-4aN0-1M0 ESCC could progress to the unresectable status (cT4b) if
they showed poor response to the therapeutic regimen, which mostly occurred
in the nCT group [13 (9.85%) versus 3 (2.3%); P = 0.010]. Consequently, the
cancer-specific survival between the two strategies was analyzed and results are
presented in Supplementary Figure S1, available at
https://doi.org/10.1016/j.annonc.2022.10.508 . There was a noticeable trend of
:
 fewer cancer-related deaths in the nCRT group compared with that in the nCT
group; however, the better cancer-specific survival of the nCRT strategy may be
counteracted by the high rate of therapy-related noncancer deaths [11 (19.3%)
versus 6 (9.3%); P = 0.21]. Several studies indicated that nCRT may increase the
severity of postoperative complications and 30/90-day mortality,8,10,11 and in
our study, the postoperative morbidity and mortality were comparable between
the two strategies.18 However, in the long-term outcome, there were more
treatment-related deaths in the nCRT group, which may be due to the persistent
side-effects of the neoadjuvant, as nCRT altered pulmonary physiology, in
particular impairing the diffusion capacity for carbon monoxide.23 Minimally
invasive surgery could reduce the surgical trauma in the acute phase, but could
not eliminate the pulmonary injury of nCRT. Based on our results, as well as
those from previous studies, ∼50% of neoadjuvant-therapy patients had a high
risk of relapse. Therefore systemic therapy with acceptable adverse effects is
required to reduce and avoid recurrence further, such as immunotherapy, and
we are conducting a new clinical trial to explore the safety and efficacy of nCRT
in combination with the programmed cell death protein 1 (PD-1) antibody for
cII-III ESCC (NCT04973306).

Surgery is also an essential element in the neoadjuvant therapeutic strategy and

it is of great importance in the selection of surgical approach to achieve R0
resection and meet the demand of lymphadenectomy. In our study, the Ivor-
Lewis or Mckeown method with the standard extensive two-field (including
bilateral laryngeal recurrent nerve lymph nodes dissection in the upper
mediastinum) or three-field lymphadenectomy was recommended and carried
out in all patients to achieve radical resection. However, this procedure was
neglected in the previous trials. It is also well-known that transhiatal
esophagectomy compromises thoracic lymph node dissection, but was reported
to be optional and acceptable in the previous studies, such as the CROSS trial.
Thus, for those patients undergoing transhiatal esophagectomy, there could
have been early recurrence due to nondissection of thoracic lymph nodes,
especially among the surgery-alone cases. The extent of lymphadenectomy
played a key role in determining the therapeutic effect, and thoracic and
abdominal lymph nodes were dissected thoroughly in our study as a result of
improved survival compared with that of the CROSS trial on the condition of
:
 more advanced tumors.

Another potential advantage of our study is that MIE was suggested and
successfully completed in >95% patients in both groups. The enlarged exposure
by the thoracoscope and laparoscope are helpful to improve radical resection of
the primary tumor as well as lymph node dissection. Besides, the quality control
can be easily carried out for every operation by checking the recorded surgical
video from each center. The consensus on surgical quality plays a key role in the
results of such kinds of clinical trials, from either the oncological or the surgical
point of view. A high risk of postoperative morbidity and mortality of patients
after nCRT was reported in the previous trials,8,10,11 which may partly be due to
the trauma caused by open esophagectomy; therefore MIE was chosen as one
key composition in our trial to explore the safety of the strategy. It revealed that
nCRT followed by MIE could be safe and feasible for locally advanced ESCC as
the severity of postoperative complications and 30-day mortality were not
increased compared with nCT (reported previously).18

This study has several limitations. First, the current sample size of our study
may be underpowered to detect a true but small difference between the two
strategies. A further individual patient-level data meta-analysis of all relevant
RCTs may be needed to identify treatment options with larger survival benefits.
In addition, patients with poorer performance status and older patients were
not recruited; therefore the applicability of this therapy to these patients
requires additional study.

In conclusion, this study demonstrated that nCRT followed by MIE could not
improve overall survival significantly, despite better tumor response and similar
morbidity and mortality, compared with nCT. Consequently, the results
underscore the pending issue of the best strategy of neoadjuvant therapy for
locally advanced bulky ESCC.

Acknowledgements
We thank Fei Liang, PhD (Department of Biostatistics, Zhongshan Hospital,
Fudan University) for his assistance in statistical analysis with no compensation.
:
 Funding
This work was supported by National Natural Science Foundation of China,
China [grant number 81902396]; Science and Technology Commission of
Shanghai Municipality, China [grant number 16411965900]; Zhongshan
Hospital, Fudan University, China [grant numbers 2016ZSLC15 and
2021ZSYQ27]; respectively.

Disclosure
The authors have declared no conflicts of interest.

Data sharing
Deidentified individual participant data (including a data dictionary) that
underlie the results reported in this article will be shared beginning 9 months
and ending 36 months following publication. Investigators who wish to use the
data for individual patient data meta-analysis can direct their proposal to the
corresponding author ([email protected] ). Data will be made available
via an appropriate data archive.

Supplementary data

Download all supplementary files What’s this?

Download: Download Acrobat PDF file (356KB)

Supplementary Material. Kaplan-Meier estimates of cancer-specific survival. HR,
hazard ratios; CI, confidence intervals. 150x95mm (300 x 300 DPI).

Download: Download high-res image (262KB)

Download: Download full-size image (38KB)
:
 Supplementary Figure1.

Recommended articles

References
1 M. Arnold, J. Ferlay, M.I. van Berge Henegouwen, et al.
Global burden of oesophageal and gastric cancer by histology and
subsite in 2018
Gut, 69 (9) (2020), pp. 1564-1571
Crossref View in Scopus Google Scholar

2 H. Sung, J. Ferlay, R.L. Siegel, et al.

Global cancer statistics 2020: GLOBOCAN estimates of incidence
and mortality worldwide for 36 cancers in 185 countries
CA Cancer J Clin, 71 (3) (2021), pp. 209-249
Crossref Google Scholar

3 N. Ando, H. Kato, H. Igaki, et al.

A randomized trial comparing postoperative adjuvant
chemotherapy with cisplatin and 5-fluorouracil versus
preoperative chemotherapy for localized advanced squamous cell
carcinoma of the thoracic esophagus (JCOG9907)
Ann Surg Oncol, 19 (1) (2012), pp. 68-74
Crossref View in Scopus Google Scholar

4 P. van Hagen, M.C. Hulshof, J.J. van Lanschot, et al.

Preoperative chemoradiotherapy for esophageal or junctional
cancer
N Engl J Med, 366 (22) (2012), pp. 2074-2084
Crossref View in Scopus Google Scholar

5 H. Yang, H. Liu, Y. Chen, et al.

Neoadjuvant chemoradiotherapy followed by surgery versus
surgery alone for locally advanced squamous cell carcinoma of
the esophagus (NEOCRTEC5010): a phase III multicenter,
:
 randomized, open-label clinical trial
J Clin Oncol, 36 (27) (2018), pp. 2796-2803
Crossref View in Scopus Google Scholar

6 B.H. Burmeister, J.M. Thomas, E.A. Burmeister, et al.

Is concurrent radiation therapy required in patients receiving
preoperative chemotherapy for adenocarcinoma of the
oesophagus? A randomised phase II trial
Eur J Cancer, 47 (3) (2011), pp. 354-360
View PDF View article View in Scopus Google Scholar

7 F. Klevebro, G. Alexandersson von Dobeln, N. Wang, et al.

A randomized clinical trial of neoadjuvant chemotherapy versus
neoadjuvant chemoradiotherapy for cancer of the oesophagus or
gastro-oesophageal junction
Ann Oncol, 27 (4) (2016), pp. 660-667
View PDF View article Crossref View in Scopus
Google Scholar

8 F. Klevebro, G. Johnsen, E. Johnson, et al.

Morbidity and mortality after surgery for cancer of the
oesophagus and gastro-oesophageal junction: a randomized
clinical trial of neoadjuvant chemotherapy vs. neoadjuvant
chemoradiation
Eur J Surg Oncol, 41 (7) (2015), pp. 920-926
View PDF View article View in Scopus Google Scholar

9 M. Stahl, M.K. Walz, M. Stuschke, et al.

Phase III comparison of preoperative chemotherapy compared
with chemoradiotherapy in patients with locally advanced
adenocarcinoma of the esophagogastric junction
J Clin Oncol, 27 (6) (2009), pp. 851-856
View in Scopus Google Scholar

10 K. Kumagai, I. Rouvelas, J.A. Tsai, et al.

Meta-analysis of postoperative morbidity and perioperative
:
 mortality in patients receiving neoadjuvant chemotherapy or
chemoradiotherapy for resectable oesophageal and gastro-
oesophageal junctional cancers
Br J Surg, 101 (4) (2014), pp. 321-338
Crossref View in Scopus Google Scholar

11 C. Mariette, L. Dahan, F. Mornex, et al.

Surgery alone versus chemoradiotherapy followed by surgery for
stage I and II esophageal cancer: final analysis of randomized
controlled phase III trial FFCD 9901
J Clin Oncol, 32 (23) (2014), pp. 2416-2422
View in Scopus Google Scholar

12 S.S. Biere, M.I. van Berge Henegouwen, K.W. Maas, et al.

Minimally invasive versus open oesophagectomy for patients
with oesophageal cancer: a multicentre, open-label, randomised
controlled trial
Lancet, 379 (9829) (2012), pp. 1887-1892
View PDF View article View in Scopus Google Scholar

13 J. Straatman, N. van der Wielen, M.A. Cuesta, et al.

Minimally invasive versus open esophageal resection: three-year
follow-up of the previously reported randomized controlled trial:
the TIME Trial
Ann Surg, 266 (2) (2017), pp. 232-236
View in Scopus Google Scholar

14 H. Tang, H. Zheng, L. Tan, et al.

Neoadjuvant chemoradiotherapy followed by minimally invasive
esophagectomy: is it a superior approach for locally advanced
resectable esophageal squamous cell carcinoma?
J Thorac Dis, 10 (2) (2018), pp. 963-972
View in Scopus Google Scholar

15 H. Wang, Y. Shen, M. Feng, et al.

Outcomes, quality of life, and survival after esophagectomy for
:
 squamous cell carcinoma: a propensity score-matched
comparison of operative approaches
J Thorac Cardiovasc Surg, 149 (4) (2015), pp. 1006-1014
discussion 1014-1005.e1004
Google Scholar

16 Japan Esophageal Society

Japanese classification of esophageal cancer, 11th Edition: part II
and III
Esophagus, 14 (1) (2017), pp. 37-65
Google Scholar

17 Japan Esophageal Society

Japanese classification of esophageal cancer, 11th Edition: part I
Esophagus, 14 (1) (2017), pp. 1-36
Google Scholar

18 H. Wang, H. Tang, Y. Fang, et al.

Morbidity and mortality of patients who underwent minimally
invasive esophagectomy after neoadjuvant chemoradiotherapy vs
neoadjuvant chemotherapy for locally advanced esophageal
squamous cell carcinoma: a randomized clinical trial
JAMA Surg, 156 (5) (2021), pp. 444-451
Crossref View in Scopus Google Scholar

19 K. Nakamura, K. Kato, H. Igaki, et al.

Three-arm phase III trial comparing cisplatin plus 5-FU (CF)
versus docetaxel, cisplatin plus 5-FU (DCF) versus radiotherapy
with CF (CF-RT) as preoperative therapy for locally advanced
esophageal cancer (JCOG1109, NExT study)
Jpn J Clin Oncol, 43 (7) (2013), pp. 752-755
Crossref View in Scopus Google Scholar

20 G.A. von Dobeln, F. Klevebro, A.B. Jacobsen, et al.

Neoadjuvant chemotherapy versus neoadjuvant
chemoradiotherapy for cancer of the esophagus or
:
 gastroesophageal junction: long-term results of a randomized
clinical trial
Dis Esophagus, 32 (2) (2019), p. doy078
View in Scopus Google Scholar

21 S.K. Kamarajah, A.W. Phillips, L. Ferri, et al.

Neoadjuvant chemoradiotherapy or chemotherapy alone for
oesophageal cancer: population-based cohort study
Br J Surg, 108 (4) (2021), pp. 403-411
Crossref View in Scopus Google Scholar

22 N. Fan, Z. Wang, C. Zhou, et al.

Comparison of outcomes between neoadjuvant
chemoradiotherapy and neoadjuvant chemotherapy in patients
with locally advanced esophageal cancer: a network meta-
analysis
EClinicalMedicine, 42 (2021), Article 101183
View PDF View article View in Scopus Google Scholar

23 J.A. Elliott, L. O’Byrne, G. Foley, et al.

Effect of neoadjuvant chemoradiation on preoperative
pulmonary physiology, postoperative respiratory complications
and quality of life in patients with oesophageal cancer
Br J Surg, 106 (10) (2019), pp. 1341-1351
Crossref View in Scopus Google Scholar

Cited by (67)

Neoadjuvant chemotherapy combined with immunotherapy versus

neoadjuvant chemoradiotherapy in patients with locally advanced
esophageal squamous cell carcinoma
2024, Journal of Thoracic and Cardiovascular Surgery

Show abstract
: