If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Correspondence to: Department of Thoracic Surgery, The First Affiliated Hospital of Sun Yat-sen University, 58 Zhongshan 2nd Road, Guangzhou 510080, China.
Neoadjuvant camrelizumab plus chemotherapy is effective for ESCC.
•
Pathological response of PT was inconsistent with metastatic LN in 30% cases.
•
Higher ratio of CD8+ to FoxP3+ TILs correlated with better pathological response.
•
Tumour regression patterns of both PT and metastatic LN were summarised.
•
PT pathological response & ypN and circular muscle penetration were independent prognostic factors.
Abstract
Purpose
We aimed to investigate the pathological changes, clinicopathological correlation and prognostic factors of neoadjuvant programmed cell death 1 (PD-1) blockade carrelizumab combined with carboplatin and nab-paclitaxel (CCNP) which we have proved its effectiveness in previous research for resectable esophageal squamous cell carcinoma (ESCC).
Methods
108 patients of resectable ESCC, with a mean follow-up of 13 m (ranging 1–30 m), treated with neoadjuvant CCNP from March 2020 to October 2022 in the First Affiliated Hospital of Sun Yat-sen University were enrolled.
Results
One year overall survival (OS) and disease-free survival (DFS) were 96.4% and 84.7% respectively. Pathological complete response or major pathological response (pCR/MPR) of the primary tumour (T-pCR/T-MPR) and the metastatic lymph node (N-pCR/N-MPR) were 58.3% and 47.5%. Pathological response of both primary tumours (PT) and lymph nodes (LN) metastasis correlated with DFS. LN pathological response was consistent with PT in 70.0% and inconsistent in 30.0% metastatic cases. Higher ratio of CD8+ to FoxP3+ tumour-infiltrating lymphocytes (TILs), earlier ypT stage and PT invasion not beyond circular muscle correlated with better pathological response. Four types of regression patterns of PT and two types of metastatic LN regression were found. A total of 18 (16.7%) out of 108 developed recurrence with a mean time of 6.9 ± 5.3 months. PT pathological response plus ypN and PT invasion beyond circular muscle or not were independent prognostic factors of DFS.
Conclusions
This study suggested that camrelizumab plus chemotherapy had a high rate of T-pCR/T-MPR for resectable ESCC. T-pCR/T-MPR plus ypN0 and tumour invasion not beyond circular muscle predicted better DFS.
Cancer incidence, mortality, and burden in china: a time-trend analysis and comparison with the united states and united kingdom based on the global epidemiological data released in 2020.
The global, regional, and national burden of oesophageal cancer and its attributable risk factors in 195 countries and territories, 1990–2017: a systematic analysis for the global burden of disease study 2017.
Most EC patients are already in the locally advanced stage when first diagnosed and the overall 5-year survival for EC remains less than 20% in the past [
]. In recent years, neoadjuvant chemotherapy (NCT) or neoadjuvant chemoradiotherapy (NCRT) combined with radical resection have gradually become the standard treatment for locally advanced EC [
Survival of neoadjuvant and adjuvant therapy compared with surgery alone for resectable esophageal squamous cell carcinoma: a systemic review and network meta-analysis.
A systematic review and network meta-analysis of neoadjuvant therapy combined with surgery for patients with resectable esophageal squamous cell carcinoma.
Neoadjuvant chemoradiotherapy plus surgery versus surgery alone for oesophageal or junctional cancer (cross): long-term results of a randomised controlled trial.
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.
Surgery alone versus chemoradiotherapy followed by surgery for stage i and ii esophageal cancer: final analysis of randomized controlled phase iii trial FFCD 9901.
A randomized clinical trial of neoadjuvant chemotherapy versus neoadjuvant chemoradiotherapy for cancer of the oesophagus or gastro-oesophageal junction.
Pathological response and survival with neoadjuvant therapy in melanoma: a pooled analysis from the international neoadjuvant melanoma consortium (INMC).
Personalized response-directed surgery and adjuvant therapy after neoadjuvant ipilimumab and nivolumab in high-risk stage iii melanoma: the PRADO trial.
Nie R., Chen F., Provencio M., et al. Predictive value of radiological response, pathological response and relapse-free survival for overall survival in neoadjuvant immunotherapy trials: pooled analysis of 29 clinical trials. Eur J Cancer this issue (pages to be provided by Editorial Office), 2023.
]. And clinical trials have shown the safety and efficacy of the PD-1 inhibitor for patients with advanced or metastatic esophageal squamous cell carcinoma (ESCC) [
Camrelizumab versus investigator's choice of chemotherapy as second-line therapy for advanced or metastatic oesophageal squamous cell carcinoma (escort): a multicentre, randomised, open-label, phase 3 study.
Camrelizumab is an immune checkpoint inhibitor targeting PD-1 developed by a Chinese pharmaceutical company (Hengrui, Jiangsu, China). In the previous study, we proved the safety and feasibility of using the combination of neoadjuvant camrelizumab plus carboplatin and nab-paclitaxel (CCNP) in patients with resectable ESCC [
]. Here, we report on the pathological response and prognostic factors of neoadjuvant anti-PD1 plus chemotherapy in our large series of resectable ESCC.
2. Materials and methods
2.1 Patient population
Patients with locally advanced ESCC who underwent radical surgery after neoadjuvant therapy of CCNP in the First Affiliated Hospital of Sun Yat-sen University from March 2020 to October 2022 were included. Detailed patients’ inclusion criteria, exclusion criteria and therapeutic schedule were provided in Supplementary Material item 1. The study was approved by the Treatment and IEC for Clinical Research and Animal Trials of the First Affiliated Hospital of Sun Yat-sen University.
OS was defined as the time from the date of surgery to the date of death for any cause. Disease-free survival (DFS) was defined as the time from the date of surgery to the date of primary recurrence or metastasis. Disease recurrence included locoregional recurrence (LRR) and distant recurrence (DR) after surgery. LRR was defined as relapse within the oesophagus or regional lymph nodes (LN), excluding the supraclavicular nodes. DR was defined as metastatic disease in distant sites beyond the locoregional sites [
All pathological specimens were evaluated by two experienced pathologists. The ypTNM staging was evaluated according to the Esophageal Cancer Staging System, version 8 [
]. Tumour grade, depth of invasion, lymph node involvement, distal and proximal resection margins, lymphovascular invasion (LVI) and nerve invasion were evaluated. Immune-related pathologic response criteria (irPRC) were applied to assess pathological response using the percentage of immune-related residual viable tumour (%irRVT). %irRVT = viable tumour area/total tumour bed area, whereby the total tumour bed = regression bed + residual viable tumour (RVT) + necrosis [
Pathologic features of response to neoadjuvant anti-pd-1 in resected non-small-cell lung carcinoma: a proposal for quantitative immune-related pathologic response criteria (IRPRC).
]. PCR was defined as absence of viable tumour cells in the specimens, irRVT = 0%; major pathological response (MPR) was defined as viable tumour cells >0% and ≤10%, 0% < irRVT ≤10% [
Pathological response after neoadjuvant chemotherapy in resectable non-small-cell lung cancers: proposal for the use of major pathological response as a surrogate endpoint.
]; pathological partial response (pPR) was defined as viable tumour cells>10% but ≤70%, 10% < irRVT ≤ 70%; pathological non response (pNR) was defined as viable tumour cells >70%, irRVT > 70%. LN regression grade was defined based on the overall irRVT of all involved LN due to the discrepancy of response among multiple LN in the same case.
Immunohistochemical (IHC) staining of FAP, CD8 and FoxP3 was performed in primary tumours and metastatic LN on 15 cases. Detailed staining procedures and antibody information were provided in Supplementary Material item 1 and 2. For the IHC results, positive cells were counted and averaged in 3 HPF (400×).
2.3 Statistical analysis
Categorical variables were expressed as frequency number and percentage. Chi-square test or Fisher’s exact test was used for the comparison of disordered variables between groups, and Mann-Whitney U test was used for the comparison of ordered variables. Univariate and multivariate logistic regression was used to analyse the effect of clinicopathological features on pathological response. Survival probability was computed by the Kaplan-Meier estimator. The log-rank test was used to compare patient survival times between groups. Univariable and multivariate Cox proportional hazards regression model was used to examine the association between clinicopathological features and prognosis. The variables found significant in univariate analysis (P value < 0.25) were evaluated by multivariable analysis after stepwise forwards method. A P-value of less than 0.05 was taken to be significant. The statistical analyses and graphing were performed using SPSS (version 26.0) and GraphPad Prism (version 9.0).
3. Results
3.1 General clinical data and overall prognosis
108 patients were enrolled, including 87 males (80.6%) and 21 females (19.4%), with a median age of 60 years (42–79 years) (Table 1). The follow-up ranged from 1 to 30 months (median 13 m). A total of 18 (16.7%) patients developed primary recurrence or metastasis with a mean time of 6.9 ± 5.3 months, among which 7 (38.8%) cases were LRR and 11 (61.1%) cases were DR. The sites of metastasis included cervical LN, mediastinal LN, abdominal LN, mediastinum, lung, liver and brain. Three (2.8%) patients died of metastasis 2, 7 and 11 months after surgery.
Table 1Clinicopathological data of 108 cases of ESCC treated with neoadjuvant CCNP.
Clinicopathological features
Total (%)
T-pCR/T-MPR (%)
T-pPR/T-pNR (%)
P
Non-recurrence (%)
Recurrence (%)
P
108
63
45
90
18
Age
0.896
0.491
≤60 years old
56 (51.9)
33 (52.4)
23 (51.1)
48 (53.3)
8 (44.4)
>60 years old
52 (48.1)
30 (47.6)
22 (48.9)
42 (46.7)
10 (55.6)
Gender
0.711
1.000
Male
87 (80.6)
50 (79.4)
37 (82.2)
72 (80.0)
15 (83.3)
Female
21 (19.4)
13 (20.6)
8 (17.8)
18 (20.0)
3 (16.7)
Tumour location
0.224
0.718
Upper
22 (20.4)
15 (23.8)
7 (15.6)
19 (21.1)
3 (16.7)
Middle
35 (32.4)
21 (33.3)
14 (31.1)
29 (32.2)
6 (33.3)
Lower
35 (47.2)
27 (42.9)
24 (53.3)
42 (46.7)
9 (50.0)
Tumour grade
0.965
0.168
G1
4 (3.7)
2 (3.2)
2 (4.4)
4 (4.4)
0 (0.0)
G2
64 (59.3)
38 60.3)
26 (57.8)
55 (61.1)
9 (50.0)
G3
40 (37.0)
23 (36.5)
17 (37.8)
31 (34.4)
9 (50.0)
PT pathological response
0.001
T-pCR
37 (34.3)
36 (40.0)
1 (5.6)
T-MPR
26 (24.1)
22 (24.4)
4 (22.2)
T-pPR
27 (25.0)
20 (22.2)
7 (38.9)
T-pNR
18 (16.7)
12 (13.3)
6 (33.3)
ypT
<0.001
0.001
ypT0
37 (34.3)
37 (58.7)
0 (0.0)
36 (40.0)
1 (5.6)
ypT1
23 (21.3)
16 (25.4)
7 (15.6)
21 (23.3)
2 (11.1)
ypT2
25 (23.1)
9 (14.3)
16 (35.6)
17 (18.9)
8 (44.4)
ypT3
22 (20.4)
1 (1.6)
21 (46.7)
16 (17.8)
66 (33.3)
ypT4
1 (0.9)
0 (0.0)
1 (2.2)
0 (0.0)
1 (5.6)
ypN
0.083
0.015
ypN0
83 (76.9)
52 (82.5)
31 (68.9)
73 (81.1)
10 (55.6)
ypN1
21 (19.4)
10 (15.9)
11 (24.4)
15 (16.7)
6 (33.3)
ypN2
2 (1.9)
1 (1.6)
1 (2.2)
1 (1.1)
1 (5.6)
ypN3
2 (1.9)
0 (0.0)
2 (4.4)
1 (1.1)
1 (5.6)
Muscularis propria invasion
<0.001
<0.001
Not beyond circular muscle
71 (65.7)
57 (90.5)
14 (31.1)
67 (74.4)
4 (22.2)
Beyond circular muscle
37 (34.3)
6 (9.5)
31 (68.9)
23 (25.6)
14 (77.8)
Regression of lymph nodes
0.458
0.778
No
80 (74.1)
45 (71.4)
35 (77.8)
66 (73.3)
14 (77.8)
Yes
28 (25.9)
18 (28.6)
10 (22.2)
24 (26.7)
4 (22.2)
LVI
0.002
0.089
No
101 (93.5)
63 (100.0)
38 (84.4)
86 (95.6)
15 (83.3)
Yes
7 (6.5)
0 (0.0)
7 (15.6)
4 (4.4)
3 (16.7)
Nerve invasion
<0.001
0.478
No
91 (84.3)
63 (100.0)
28 (62.2)
77 (85.6)
14 (77.8)
Yes
17 (15.7)
0 (0.0)
17 (37.8)
13 (14.4)
4 (22.2)
T-, in primary tumours; pCR, pathological complete response; MPR, major pathological response; pPR, pathological partial response; pNR, pathological non-response; PT, primary tumours; LVI, lymphovascular invasion; ESCC, esophageal squamous cell carcinoma; CCNP, camrelizumab plus carboplatin and nab-paclitaxel.
3.2 Overall pathological response and morphological changes after CCNP treatment
Tumour regression morphological changes of ESCC after CCNP treatment included interstitial fibrosis, muscular fascicle disarrangement, foam cells reaction, giant cell granuloma, acellular keratin pearl, ghost cell and massive lymphocyte infiltration (Fig. 1). Primary tumours (PT) pathologic response of T-pCR, T-MPR, T-pPR and T-pNR was 37 (34.3%), 26 (24.1%), 27 (25.0%) and 18 (16.7%) respectively (Fig. 2A). The total proportion of T-pCR and T-MPR was 58.3%.
Fig. 1Pathologic response to neoadjuvant camrelizumab plus carboplatin and nab-paclitaxel (CCNP) for resectable esophageal squamous cell carcinoma (ESCC). (A, B, C, D) A representative example of the histopathology of T-pCR, T-MPR, T-pPR, T-pNR. Original magnification: ×10 (pictures) and ×100 (insets).
Pathological response in LN was also evaluated by the same criteria as in PT. Morphological changes of LN were similar to those of the PT (Fig. 3). Totally, 40 cases including 15 ypN0 which had evidence of tumour regression and 25 ypN+ were assessed. LN pathologic response of N-pCR, N-MPR, N-pPR and N-pNR was 15 (37.5%), 4 (10.0%), 4 (10.0%) and 17 (42.5%), respectively (Fig. 2B). The total proportion of N-pCR and N-MPR was 47.5%. The pathological response of LN was consistent with PT in 28 (70.0%) cases, while was inconsistent in 12 (30.0%) cases among which 8 (66.7%) cases showed worse LN pathological response and 4 (33.3%) showed better LN pathological response compared with PT (Table 2).
Fig. 3Pathologic response of lymph nodes. (A) N-pCR after neoadjuvant camrelizumab plus carboplatin and nab-paclitaxel (CCNP) therapy. Some giant cell granulomas, keratin pearls and interstitial fibrosis were left. (B) Lymph node metastasis with N-MPR. (C) Lymph node metastasis with N-pPR. (D) Lymph node metastasis with N-pNR. Original magnification: ×10 (pictures) and ×100 (insets).
Among 18 recurrent cases, T-pCR, T-MPR, T-pPR and T-pNR was 1 (5.5%), 4 (22.2%), 7 (38.8%) and 6 (33.3%), respectively. Nine cases developed LN metastasis including eight cases were N-pPR/N-pNR and only 1N-pCR. PT pathologic response in five out of nine cases was inconsistent with LN pathological response (Table 3).
Table 3Clinicopathological data of 18 recurrent cases.
3.3 Clinicopathologic features correlated with pathological response of primary tumour
There were significant differences in ypT (P < 0.001), PT invasion beyond circular muscle or not (P < 0.001), LVI (P = 0.002) and nerve invasion (P < 0.001) between T-pCR/ T-MPR and T-pPR/T-pNR (Table 1). Earlier ypT stage, PT invasion not beyond circular muscle, less LVI and nerve invasion tended to have better PT pathological response. Univariate logistic analysis showed that ypT (P < 0.001) and PT invasion beyond circular muscle or not (P < 0.001) correlated with PT pathological response, respectively. And multivariate logistic analysis showed that ypT (P < 0.001) was an independent protective factor of PT pathological response (Table 4). Patients with earlier ypT stage had better PT pathological response.
Table 4Univariate and multivariate logistic analysis of PT pathological response of ESCC treated with neoadjuvant CCNP.
Clinicopathological features
T-pCR/T-MPR
T-pPR/T-pNR
OR
95% CI
P
Univariate
Tumour location
1.696
0.628–4.579
0.297
Upper
15
7
Middle + lower
48
38
Tumour grade
0.705
0.096–5.201
0.732
G1
2
2
G2+G3
61
43
ypT
28.771
10.049–82.375
<0.001
T0 + 1
53
7
T2 + 3 + 4
10
38
ypN
2.135
0.863–5.284
0.101
N0
52
31
N1 + 2 + 3
11
14
Muscularis propria invasion
21.036
7.349–60.211
<0.001
Not beyond circular muscle
57
14
Beyond circular muscle
6
31
Regression of lymph nodes
0.714
0.293–1.740
0.459
No
45
35
Yes
18
10
Multivariate
ypT
28.771
10.049–82.375
<0.001
T0 + 1
53
7
T2 + 3 + 4
10
38
ypN
N0
52
31
N1 + 2 + 3
11
14
Muscularis propria invasion
Not beyond circular muscle
57
14
Beyond circular muscle
6
31
T-, in primary tumours; pCR, pathological complete response; MPR, major pathological response; pPR, pathological partial response; pNR, pathological non-response; PT, primary tumours; ESCC, esophageal squamous cell carcinoma; CCNP, camrelizumab plus carboplatin and nab-paclitaxel.
3.4 IHC analysis of immune microenvironment and its correlation with pathological response
Cancer-associated fibroblasts (CAFs) marker FAP, CD8+ and FoxP3+ tumour-infiltrating lymphocytes (TILs) were tested in 15 cases, including 15 PT and 20 LN. The intra-/peritumoral border was defined at low magnification (40×). Three HPF (400×) were then selected for both intratumoral and peritumoral sites of each specimen. The integrated density value of each HPF was calculated with ImageJ and LOG10 of the average value was used for statistics. (Fig. 4A). In the PT of non-pCR cases numbers of FAP+ CAFS, CD8+ and FoxP3+ TILs in peritumoral area were more than in intratumoral sites (Fig. 4B). CD8+ TILs in the PT of T-pCR /T-MPR were more than T-pPR/T-pNR (Fig. 4C). And the ratio of CD8+ to FoxP3+ TILs of T-pCR/T-MPR was higher than T-pPR/T-pNR (Fig. 4D). In the LN, FAP+ CAFS, CD8+ and FoxP3+ TILs in peritumoral tissues were more than in intratumoral sites (Fig. 4E). FoxP3+ TILs of N-pNR were more than N-pCR (Fig. 4F) and the ratio of CD8+ to FoxP3+ TILs of N-pCR was higher than N-pNR (Fig. 4G).
Fig. 4FAP+, CD8+ and FoxP3+ cells in primary tumours (PT) and lymph nodes (LN). (A) FAP, CD8 and FoxP3 immunohistochemical (IHC) staining in PT. (B) In the PT of non-pCR, cell numbers of FAP+ CAFS, CD8+ and FoxP3+ tumour-infiltrating lymphocytes (TILs) in peritumoral tissues were more than in intratumoral sites. (C) Cell numbers of CD8+ TILs in PT of T-pCR /T-MPR were more than T-pPR/T-pNR. (D) The ratio of CD8+ to FoxP3+ TILs in the PT of T-pCR/T-MPR was higher than T-pPR/T-pNR. (E) Cell numbers of FAP+ CAFS, CD8+ and FoxP3+ TILs in peritumoral tissues were more than in intratumoral sites in the LN. (F) Cell numbers of FoxP3+ TILs in the LN of N-pNR were more than N-pCR. (G) The ratio of CD8+ to FoxP3+ TILs in the LN of N-pCR was higher than N-pNR.
The pretreatment pathological tumour stage before neoadjuvant CCNP therapy was graded as prepT based on the regression changes and the presence of residual tumours in the resection specimen. A total of 61 patients were staged as prepT3–4 (Fig. 5A). Analysis of residual tumours showed that mucosa had the highest rate of residual tumours (86.9%), followed by submucosa (80.3%), muscularis propria (75.4%), and adventitia was the lowest (37.7%) (Fig. 5B).
Fig. 5Tumour regression patterns and analysis of residual tumour cells in esophageal squamous cell carcinoma (ESCC) after camrelizumab plus carboplatin and nab-paclitaxel (CCNP) therapy. (A) 61 patients with non-pCR status of primary lesion and staged as prepT3–4. (B) Location of residual tumour within the oesophagal wall of these 61 patients. (C) Regression patterns of the oesophagal wall and lymph nodes (LN) after neoadjuvant CCNP therapy (modified from the article of Joel Shapiro and Han Tang). The dashed line areas represent the areas of original (i.e. pretreatment) tumour extension. The red solid areas represent the areas with residual tumour cells after neoadjuvant therapy. The blue dots represent TILs, and the reticulated black lines represent cancer-associated fibroblasts (CAFs). (D) Different primary tumours (PT) regression patterns in 61 prepT3–4 patients. (E) Different LN regression patterns in 25 metastatic LN. (F) Regression patterns of PT and LN in recurrent and non-recurrent cases. (G) Heat map of the residual tumour location of different PT and LN pathological response.
] on tumour regression patterns, four patterns of Type Ⅰ, Type Ⅱ, Type Ⅲ and Type Ⅳ could be classified in PT, and two patterns of Type A and Type B could be classified in metastatic LN. Detailed type definitions were provided in Supplementary Material item 3. We found that CAFs accumulated in the peritumoral area of the residual tumour, TILs accumulate more extensively in type Ⅰ than in type Ⅱ, Ⅲ and Ⅳ in the tumour regression bed (Fig. 5C). Cases from type Ⅰ to Ⅳ were 25 (41.0%), 7 (11.5%), 4 (6.6%), and 25 (41.0%) respectively (Fig. 5D). In 25 metastatic LN, there were four (16.0%) cases of Type A, 9 (36.0%) cases of Type B and 12 (48.0%) cases of no regression (Fig. 5E). Out of 61, 16 cases relapsed, including 3 cases of Type Ⅰ, 2 cases of Type Ⅱ, 1 case of Type Ⅲ and 10 cases of Type Ⅳ which was the highest (62.5%). Three cases with LN regression relapsed, all of which were Type B (Fig. 5F).
In T-MPR cases, the residual tumour cells mainly presented in mucosa and submucosa area, with a small proportion cases in muscularis propria and adventitia. In T-pPR and T-pNR cases, residual tumour cells mostly presented in all four layers only a few cases with adventitia expired. The pathological response grade of the PT and LN could be inconsistent (Fig. 5G).
3.6 Patients’ survival and clinicopathologic features correlated with DFS
Considering recurrence or not, there were differences in PT pathological response (P = 0.001), ypT (P = 0.001), ypN (P = 0.015) and PT invasion beyond circular muscle or not (P < 0.001) between two groups (Table 1).
Ten patients followed up for less than 6 months without recurrence or death were excluded. The 1-year OS was 96.4% (Fig. 6A) and 1-year DFS was 84.7% (Fig. 6B). One-year DFS of patients with T-pCR, T-MPR, T-pPR and T-pNR were 96.8%, 84.6%, 82.5% and 61.8% respectively. There was no significant difference of DFS between T-pCR and T-MPR as well as between T-pPR and T-pNR (Fig. 6C). Therefore, patients were divided into T-pCR/T-MPR and T-pPR /T-pNR. DFS of T-pCR/T-MPR group was much better than those of T-pPR/T-pNR (Fig. 6D).
Fig. 6Kaplan-Meier of overall survival and disease-free survival. (A.B) Overall survival (OS) and Disease-free survival (DFS). (C) DFS of different primary tumours (PT) pathological response. Group T-pCR had the best DFS. (D) Group T-pCR/T-MPR had better DFS than group T-pPR/T-pNR. (E) DFS was significantly different between N-pCR/N-MPR and N-pPR/N-pNR. (F) Consistent pathological response of PT and lymph nodes (LN) had better DFS than inconsistent cases, though the difference was not statistically significant. (G) DFS was the best in T-pCR/T-MPR plus ypN0. (H) DFS of PT Invasion not beyond circular muscle was better than PT invasion beyond circular muscle.
The grade of LN pathological response correlated with DFS in 40 LN metastatic patients. LN with pathological response of N-pPR/N-pNR had worse DFS compared with N-pCR/N-MPR (Fig. 6E). Cases with consistent pathological response of PT and LN tended to have better DFS than inconsistent cases, though the difference was not statistically significant (P = 0.097) (Fig. 6F).
We combined the PT pathological response with ypN and found that T-pCR/T-MPR plus ypN0 had the best DFS. And the DFS of T-pPR/T-pNR was the worst regardless whether there was LN metastasis or not. DFS of T-pCR/T-MPR with LN metastasis was in the middle (Fig. 6G). In addition, we found that the DFS of PT invasion beyond circular muscle was worse than not beyond (Fig. 6H).
Univariate Cox analysis showed that PT pathological response (P = 0.004), ypT (P = 0.002), ypN (P = 0.023), PT invasion beyond circular muscle or not (P < 0.001) and PT pathological response combined with ypN (P = 0.003) were respectively correlated with DFS (Table 5). Multivariate Cox analysis showed that PT pathological response combined with ypN (P = 0.001) and PT invasion beyond circular muscle or not (P = 0.004) were the independent factors of DFS (Table 6). Thus, patients with T-pCR/T-MPR, earlier ypT, ypN0, PT invasion not beyond circular muscle and T-pCR/T-MPR plus ypN0 predicted better DFS and lower recurrence rate. T-pCR/T-MPR plus ypN0 and PT invasion not beyond circular muscle were the independent protective factor.
Table 5Univariate COX analysis of DFS of ESCC treated with neoadjuvant CCNP.
Clinicopathological features
HR
95% CI
P
Univariate
Tumour location
1.085
0.313–3.759
0.897
Upper
Middle+ lower
Tumour grade
22.074
0.002–220413.888
0.510
G1
G2 + G3
PT pathological response
4.601
1.620–13.068
0.004
T-pCR/T-MPR
T-pPR/T-pNR
ypT
6.815
1.968–23.596
0.002
T0 + 1
T2 + 3 + 4
ypN
2.954
1.165–7.490
0.023
N0
N1 + 2 + 3
Muscularis propria invasion
8.064
2.643–24.606
<0.001
Not beyond circular muscle
Beyond circular muscle
Regression of lymph nodes
0.836
0.274–2.545
0.752
No
Yes
LVI
3.234
0.933–11.203
0.064
No
Yes
Nerve invasion
1.930
0.632–5.890
0.248
No
Yes
PT pathological response and ypN
20.824
2.756–157.344
0.003
T-pCR/T-MPR plus ypN0
Others
T-, in primary tumours; pCR, pathological complete response; MPR, major pathological response; pPR, pathological partial response; pNR, pathological non-response; PT, primary tumours; LVI, lymphovascular invasion; DFS, disease-free survival; ESCC, esophageal squamous cell carcinoma; CCNP, camrelizumab plus carboplatin and nab-paclitaxel.
]. A trial conducted in China called ESCORT explored the efficacy and safety of the PD-1 inhibitor camrelizumab versus chemotherapy in advanced/metastatic ESCC and showed that camrelizumab could significantly prolong survival when compared with chemotherapy [
Camrelizumab versus investigator's choice of chemotherapy as second-line therapy for advanced or metastatic oesophageal squamous cell carcinoma (escort): a multicentre, randomised, open-label, phase 3 study.
]. And the ESCORT-1st study showed that adding camrelizumab to conventional chemotherapy could significantly improve overall survival and progression-free survival among patients with advanced/metastatic ESCC compared with placebo and chemotherapy alone [
Effect of camrelizumab vs placebo added to chemotherapy on survival and progression-free survival in patients with advanced or metastatic esophageal squamous cell carcinoma: the ESCORT-1st randomized clinical trial.
]. All these studies suggest that camrelizumab has a huge potential in EC.
After curative resection for ESCC, postoperative complications, pT3, pT4, and pN+ were independent poor prognostic factors in multivariate analyses in pStage 0, I, and II patients [
]. However, only a handful of articles investigated the relationship between clinicopathological factors and the prognosis of EC after neoadjuvant immunotherapy. In this study, we observed the pathological changes, tumour regression patterns, immune microenvironment, the effect of clinicopathological features on pathological response and the factors affecting disease recurrence of neoadjuvant CCNP for resectable ESCC.
Our study was conducted on patients with clinical stage T2–4aN0–3M0. And its total proportion of T-pCR and T-MPR was 58.3%, in which pCR rate was 34.3%. To compare with, landmark trials such as NEOCRTEC5010 trial [
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.
Surgery alone versus chemoradiotherapy followed by surgery for stage i and ii esophageal cancer: final analysis of randomized controlled phase iii trial FFCD 9901.
], in which chemoradiotherapy was used with pCR rates of 43.2%, 33.3%, and 29%, respectively, indicate that the pCR rate of this CCNP treatment regimen is similar to that of previous NCRT. In addition, 1-year DFS and OS of patients in our study were 84.7% and 96.4%, which was higher than traditional therapies (about 80% and 85%) [
Neoadjuvant chemoradiotherapy plus surgery versus surgery alone for oesophageal or junctional cancer (cross): long-term results of a randomised controlled trial.
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.
Surgery alone versus chemoradiotherapy followed by surgery for stage i and ii esophageal cancer: final analysis of randomized controlled phase iii trial FFCD 9901.
]. However, cross-comparisons between different trials still need to be more cautious. Because the patients enrolled in different trials had different clinical stages and the end points of pathological response may also be defined differently.
Previous articles indicated that tumour microenvironment had potential correlation with pathological response and clinical outcomes [
Pd-l1 expression, cd8+ and cd4+ lymphocyte rate are predictive of pathological complete response after neoadjuvant chemoradiotherapy for squamous cell cancer of the thoracic esophagus.
Immunoscore signatures in surgical specimens and tumor-infiltrating lymphocytes in pretreatment biopsy predict treatment efficacy and survival in esophageal cancer.
] on tumour regression patterns, we added some connections between tumour regression patterns and the tumour microenvironment.
The previous article showed that pCR patients with locally advanced EC after neoadjuvant treatment significantly improved OS and DFS, and hence pCR was considered as predictor of long-term survival [
Complete pathological response after neoadjuvant treatment in locally advanced esophageal cancer predicts long term survival: a retrospective cohort study.
]. However, our study showed that T-MPR has similar DFS as T-pCR and both had a better prognosis than T-pPR and T-pNR. What’s more, patients with T-pCR/T-MPR could get better DFS if they did not have LN metastasis. However, one case of T-MPR plus ypN0 whose PT invaded beyond circular muscle relapsed. Our results also suggested that in addition to pathological response, the depth of muscle invasion also affect the prognosis of ESCC.
This study still has some limitations. First, patients in the study were all from one hospital. Second, the follow-up time was not long enough so that we have not been able to get more comprehensive prognostic data.
In conclusion, the results of this study suggest that CCNP is a promising neoadjuvant treatment for resectable ESCC. T-pCR/T-MPR plus ypN0 and tumour invasion not beyond circular muscle predict better prognosis.
Funding
This research was supported by the National Natural Science Foundation of China No. 81772701 and No. 82072824.
CRediT authorship contribution statement
Honglei Wang and Zeying Jiang collected and analysed the clinical and pathological data, conducted the IHC experiment, interpreted the results and drafted the original manuscript. Honglei Wang managed data and performed the statistical analysis. Qihua Wang, Tong Wu, Fangzhou Guo and Zhengyuan Xu collected the follow-up information. Chao Cheng and Weixiong Yang performed the surgery, collected and analysed the clinical data. Wenfang Chen and Shicong Yang confirmed the pathological diagnosis and evaluated pathological response. Shiting Feng, Xiaoyan Wang and Shuling Chen provided the imaging material. Wenfang Chen and Chao Cheng conceived and designed the study, interpreted the results, and revised the manuscript. Wenfang Chen provided the overall direction and the financial support of the series of studies and supervised the study.
Conflict of interest statement
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
We thank the Nephropathy Laboratory of the First Affiliated Hospital of Sun Yat-sen University for providing platform for our IHC experiments and thank Xiaoyang Chen for drawing the diagram of the tumour regression patterns.
Cancer incidence, mortality, and burden in china: a time-trend analysis and comparison with the united states and united kingdom based on the global epidemiological data released in 2020.
The global, regional, and national burden of oesophageal cancer and its attributable risk factors in 195 countries and territories, 1990–2017: a systematic analysis for the global burden of disease study 2017.
Survival of neoadjuvant and adjuvant therapy compared with surgery alone for resectable esophageal squamous cell carcinoma: a systemic review and network meta-analysis.
A systematic review and network meta-analysis of neoadjuvant therapy combined with surgery for patients with resectable esophageal squamous cell carcinoma.
Neoadjuvant chemoradiotherapy plus surgery versus surgery alone for oesophageal or junctional cancer (cross): long-term results of a randomised controlled trial.
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.
Surgery alone versus chemoradiotherapy followed by surgery for stage i and ii esophageal cancer: final analysis of randomized controlled phase iii trial FFCD 9901.
A randomized clinical trial of neoadjuvant chemotherapy versus neoadjuvant chemoradiotherapy for cancer of the oesophagus or gastro-oesophageal junction.
Pathological response and survival with neoadjuvant therapy in melanoma: a pooled analysis from the international neoadjuvant melanoma consortium (INMC).
Personalized response-directed surgery and adjuvant therapy after neoadjuvant ipilimumab and nivolumab in high-risk stage iii melanoma: the PRADO trial.
Nie R., Chen F., Provencio M., et al. Predictive value of radiological response, pathological response and relapse-free survival for overall survival in neoadjuvant immunotherapy trials: pooled analysis of 29 clinical trials. Eur J Cancer this issue (pages to be provided by Editorial Office), 2023.
Camrelizumab versus investigator's choice of chemotherapy as second-line therapy for advanced or metastatic oesophageal squamous cell carcinoma (escort): a multicentre, randomised, open-label, phase 3 study.
Pathologic features of response to neoadjuvant anti-pd-1 in resected non-small-cell lung carcinoma: a proposal for quantitative immune-related pathologic response criteria (IRPRC).
Pathological response after neoadjuvant chemotherapy in resectable non-small-cell lung cancers: proposal for the use of major pathological response as a surrogate endpoint.
Effect of camrelizumab vs placebo added to chemotherapy on survival and progression-free survival in patients with advanced or metastatic esophageal squamous cell carcinoma: the ESCORT-1st randomized clinical trial.
Pd-l1 expression, cd8+ and cd4+ lymphocyte rate are predictive of pathological complete response after neoadjuvant chemoradiotherapy for squamous cell cancer of the thoracic esophagus.
Immunoscore signatures in surgical specimens and tumor-infiltrating lymphocytes in pretreatment biopsy predict treatment efficacy and survival in esophageal cancer.
Complete pathological response after neoadjuvant treatment in locally advanced esophageal cancer predicts long term survival: a retrospective cohort study.
An increasing number of clinical trials are being conducted exploring the efficacy of neoadjuvant immune checkpoint inhibitors. Surrogate end-points for overall survival (OS) are urgently needed.
00143-0&id=gr1.jpg){kind=link}
00143-0&id=gr2.jpg){kind=link}
00143-0&id=gr3.jpg){kind=link}
00143-0&id=gr4.jpg){kind=link}
00143-0&id=gr5.jpg){kind=link}
00143-0&id=gr6.jpg){kind=link}