Abstract
Background
The National Comprehensive Cancer Network's Rectal Cancer Guideline Panel recommends American Joint Committee of Cancer and College of American Pathologists (AJCC/CAP) tumor regression grading (TRG) system to evaluate pathologic response to neoadjuvant chemoradiotherapy for locally advanced rectal cancer (LARC). Yet, the clinical significance of the AJCC/CAP TRG system has not been fully defined.
Materials and Methods
This was a multicenter, retrospectively recruited, and prospectively maintained cohort study. Patients with LARC from one institution formed the discovery set, and cases from external independent institutions formed a validation set to verify the findings from discovery set. Overall survival (OS), disease‐free survival (DFS), local recurrence‐free survival (LRFS), and distant metastasis‐free survival (DMFS) were assessed by Kaplan‐Meier analysis, log‐rank test, and Cox regression model.
Results
The discovery set (940 cases) found, and the validation set (2,156 cases) further confirmed, that inferior AJCC/CAP TRG categories were closely /ccorrelated with unfavorable survival (OS, DFS, LRFS, and DMFS) and higher risk of disease progression (death, accumulative relapse, local recurrence, and distant metastasis) (all p < .05). Significantly, pairwise comparison revealed that any two of four TRG categories had the distinguished survival and risk of disease progression. After propensity score matching, AJCC/CAP TRG0 category (pathological complete response) patients treated with or without adjuvant chemotherapy displayed similar survival of OS, DFS, LRFS, and DMFS (all p > .05). For AJCC/CAP TRG1–3 cases, adjuvant chemotherapy treatment significantly improved 3‐year OS (90.2% vs. 84.6%, p < .001). Multivariate analysis demonstrated the AJCC/CAP TRG system was an independent prognostic surrogate.
Conclusion
AJCC/CAP TRG system, an accurate prognostic surrogate, appears ideal for further strategizing adjuvant chemotherapy for LARC.
Implications for Practice
The National Comprehensive Cancer Network recommends the American Joint Committee of Cancer and College of American Pathologists (AJCC/CAP) tumor regression grading (TRG) four‐category system to evaluate the pathologic response to neoadjuvant treatment for patients with locally advanced rectal cancer; however, the clinical significance of the AJCC/CAP TRG system has not yet been clearly addressed. This study found, for the first time, that any two of four AJCC/CAP TRG categories had the distinguished long‐term survival outcome. Importantly, adjuvant chemotherapy may improve the 3‐year overall survival for AJCC/CAP TRG1–3 category patients but not for AJCC/CAP TRG0 category patients. Thus, AJCC/CAP TRG system, an accurate surrogate of long‐term survival outcome, is useful in guiding adjuvant chemotherapy management for rectal cancer.
Keywords: Tumor regression grade system, Locally advanced rectal cancer, Neoadjuvant treatments, Adjuvant chemotherapy, Survival outcome
Short abstract
The aim of this large cohort study was to define the clinical significance of the AJCC/CAP tumor regression grading system for locally advanced rectal cancer, which could potentially be used to select the patients who would benefit from more intensive adjuvant chemotherapy as well as to protect patients from excessive treatment.
Introduction
Neoadjuvant chemoradiotherapy (nCRT) followed by total mesorectal excision (TME) is the standard therapeutic strategy for patients with locally advanced rectal cancer (LARC) [1, 2]. After the nCRT treatment, patients show varying degree of tumor downsizing and downstaging, and 15%–38% of cases achieve a pathological complete response (pCR) [3, 4]. Patients with a pCR have a more favorable long‐term outcome than those without complete regression [3, 5]. In contrast, the nonresponders have an inferior prognosis compared with partial or complete responders [6].
Several tumor regression grading (TRG) systems have been established to stratify tumor pathologic response to nCRT for LARC. Mandard et al. developed a five‐category TRG system, a semiquantitative evaluation of residual tumor cells versus inflammatory fibrosis, to assess tumor response to chemoradiotherapy for esophageal cancer [7]. Later studies showed that the Mandard system was effective in stratifying tumor response to nCRT and predicting prognosis for LARC [8, 9]. The five‐category Dowrak/Rödel system ranked their TRG numerical scheme in a manner opposite to the Mandard system [6, 10, 11]. The Dowrak/Rödel system delineated the percentage of fibroinflammatory tissue in the tumor zone (0, <25%, 25%–50%, >50%, 100%), rather than offering a descriptive definition. Eventually, the Mandard and Dowrak/Rödel five‐category TRG systems were merged into three‐category (TRG0 + 1, TRG2 + 3, TRG4) [11], which has been shown to be as effective as their original five‐category systems [12]. The American Joint Committee on Cancer (AJCC) and the College of American Pathologists (CAP) proposed an adaptation of the Mandard system and developed a four‐category TRG system by scoring the residual tumor cells [13, 14].
Trakarnsanga et al. compared the concordance indices and the prognostic value of six different TRG systems. They concluded that the AJCC/CAP TRG system was the most accurate and should be adopted as the standard TRG system in clinical pathology practice [12]. Later, Mace et al. suggested that the AJCC/CAP TRG system might be an independent prognosticator of overall survival (OS), disease‐free survival (DFS), and cumulative recurrence [14]. Because of the low local recurrence rate and small cohort size, this study did not find the correlation between AJCC/CAP TRG criteria with local or distant recurrence, particularly for intermediate subgroups (TRG1 and TRG2). Currently, the National Comprehensive Cancer Network (NCCN)‘s Rectal Cancer Guideline Panel recommends the AJCC/CAP TRG criteria for evaluation of pathologic response in LARC, but its clinical significance has not yet been fully defined.
The aim of this large cohort study was to define the clinical significance of the AJCC/CAP TRG system for LARC, which could potentially be used to select the patients who would benefit from more intensive adjuvant chemotherapy as well as protect patients from excessive treatment.
Materials and Methods
Study Population
This was a multicenter, retrospectively designed, and data set prospectively maintained cohort study including patients from Guangdong Institute of Gastroenterology, the Sixth Affiliated Hospital of Sun Yat‐sen University in Guangzhou; Beijing Hospital in Beijing; Sun Yat‐sen University Cancer Center in Guangzhou; Sichuan Cancer Center, University of Electronic Science and Technology of China in Chengdu; Sir Run Run Shaw Hospital, College of Medicine Zhejiang University in Hangzhou; Peking University Cancer Hospital in Beijing; and Nanfang Hospital of Southern Medical University in Guangzhou. All patients with histologically confirmed locally advanced rectal adenocarcinoma within 15 cm from the anal verge, who were treated between 2006 and 2018, were included in this study. Only patients who had pelvic magnetic resonance imaging and/or transrectal ultrasonography defined II–III stage were included. Exclusion criteria were as follows: tumors other than adenocarcinoma, incomplete pathologic data, distant metastases at diagnosis, failure to complete preoperative chemoradiotherapy, failure to undergo proctectomy, prior history of malignancy, R2 resection (defined as incomplete local resection), and stage I or IV. The following information of the patient were collected: age, gender, tumor length, distance between tumor and anal margin, preoperative TNM stage, preoperative neoadjuvant radiotherapy dose and fractions, starting and ending time of preoperative neoadjuvant chemotherapy, neoadjuvant chemotherapy cycle and scheme, operation time and method, postoperative pathology, starting and ending time of postoperative adjuvant chemotherapy, adjuvant chemotherapy scheme, and course number. The Clinical Ethics Review Committee at the Sixth Affiliated Hospital of Sun Yat‐sen University approved this study.
Treatment
Neoadjuvant treatment included external beam radiation of 50.4 Gy in 25 fractions and concurrent 5‐fluorouracil chemotherapy, given orally or intravenously. Surgery with curative intent was performed according to the TME principles 4–8 weeks after the completion of chemoradiation. Standard FOLFOX‐ or CAPOX‐based adjuvant chemotherapy was administered to most cases, and the cycles of the adjuvant chemotherapy to be given were at the treating physician's discretion.
AJCC/CAP TRG Category Evaluation
Prior to initiating the study, principal investigator (X. F.) underwent a 1‐year training in the scoring of the AJCC/CAP TRG categories at the gastroenterology section, department of pathology, University of Texas MD Anderson Cancer Center. All resection specimens were examined and embedded for hematoxylin and eosin staining at the local pathology department of each institution. A central pathologic laboratory was established to review and score the AJCC/CAP TRG category at Guangdong Institute of Gastroenterology, the Sixth Affiliated Hospital of Sun Yat‐sen University. The AJCC/CAP TRG category was determined independently by two expert gastroenterology pathologists (X.‐J. F. and Y. H.), who were blinded to patient characteristics and outcomes. Discrepancies in conclusions between these two pathologists would be arbitrated by a third pathologist. The AJCC/CAP TRG system was adopted to evaluate the pathologic response to nCRT according to the volume of residual tumor cells. The four categories of AJCC/CAP TRG system were classified as following: grade 0 (complete response), no remaining viable cancer cells; grade 1 (moderate response), only small cluster or single cancer cells remaining; grade 2 (minimal response), residual cancer remaining, but with predominant fibrosis; grade 3 (poor response), minimal or no tumor kill with extensive residual cancer [13].
Follow‐Up
All of the time‐to‐event endpoints from the TME surgery date were recorded. After TME surgery, patients were followed at 3‐month intervals during the first 3 years and at 6‐month intervals thereafter. OS was defined as time to death, or when censored at the latest date if patients were still alive. DFS was defined as time to the date of relapse, or the date of death or when censored at the latest date. Local recurrence‐free survival (LRFS) and distant metastasis‐free survival (DMFS) were defined as time to the date of local recurrence or distant metastasis, respectively, or the date of death or when censored at the latest date [15].
Statistical Analysis
The primary endpoint was OS, and the secondary endpoints were LRFS, DMFS, and DFS. The χ2 test was performed to compare each AJCC/CAP TRG level with categorical clinicopathologic variables. Normally distributed continuous variables were shown as mean ± SD and were compared between AJCC/CAP TRG categories by analysis of variance. Unequal variances or nonnormally distributed variables were expressed as median and range and assessed with the Kruskal‐Wallis test. The Kaplan‐Meier method and log‐rank tests were performed to assess the survival probability differences between AJCC/CAP TRG categories or other clinicopathologic variables (ypT stage, ypN stage) with patient outcomes (OS, DFS, LRFS, and DMFS). The multivariate Cox proportional hazards model was employed to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs) for patient outcomes. Variables with p < .05 in univariate analysis were subjected to multivariate Cox regression analysis. All p values quoted were two‐sided, and p < .05 indicated a statistically significant difference. Statistical analysis was performed with IBM SPSS v20 (SPSS, Inc, Chicago, IL).
Results
Patient Characteristics
Overall, 3,096 patients with LARC, including the discovery cohort (940) and validation cohort (2,156), who underwent nCRT and TME curative surgery were recruited. Patients demographics and clinical characteristics were shown in Table 1. The median age was 55 years. There were 2,095 men (67.7%) and 1,001 women (32.3%). The median tumor distance to the anal verge was 5.0 cm. Clinical T1 (cT1), T2 (cT2), T3 (cT3), and T4 (cT4) stage was observed in 8, 110, 2006, and 972 patients, respectively. In addition, clinical N0 (cN0), N1 (cN1), and N2 (cN2) stage was noted in 656, 1352 and 1,088 cases.
Table 1.
Patient characteristics
| Characteristics | Discovery cohort | Validation cohort | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total (n = 940) | TRG 0 (n = 189) | TRG 1 (n = 258) | TRG 2 (n = 330) | TRG 3 (n = 163) | p value | Total (n = 2,156) | TRG 0 (n = 471) | TRG 1 (n = 539) | TRG 2 (n = 927) | TRG 3 (n = 219) | p value | |
| Age, yr | .621 | .067 | ||||||||||
| ≤55 | 477 | 100 (21.0) | 137 (28.7) | 161 (33.8) | 79 (16.6) | 1,003 | 239 (23.8) | 241 (24.0) | 434 (43.3) | 89 (8.9) | ||
| >55 | 463 | 89 (19.2) | 121 (26.1) | 169 (36.5) | 84 (18.1) | 1,153 | 232 (20.1) | 298 (25.8) | 493 (42.8) | 130 (11.3) | ||
| Sex | .234 | .061 | ||||||||||
| Male | 673 | 131 (19.5) | 185 (27.5) | 230 (34.2) | 127 (18.9) | 1,422 | 301 (21.2) | 341 (24.0) | 621 (43.7) | 159 (11.2) | ||
| Female | 267 | 58 (21.7) | 73 (27.3) | 100 (37.5) | 36 (13.5) | 734 | 170 (23.2) | 198 (27.0) | 306 (41.7) | 60 (8.2) | ||
| Clinical T stage | .11 | .336 | ||||||||||
| cT1 | 1 | 0 (0.0) | 0 (0.0) | 1 (100.0) | 0 (0.0) | 7 | 1 (14.3) | 3 (42.9) | 3 (42.9) | 0 (0.0) | ||
| cT2 | 37 | 14 (37.8) | 11 (29.7) | 11 (29.7) | 1 (2.7) | 73 | 24 (32.9) | 18 (24.7) | 26 (35.6) | 5 (6.8) | ||
| cT3 | 705 | 138 (19.6) | 195 (27.7) | 242 (34.3) | 130 (18.4) | 1,301 | 275 (21.1) | 326 (25.1) | 556 (42.7) | 144 (11.1) | ||
| cT4 | 197 | 37 (18.8) | 52 (26.4) | 76 (38.6) | 32 (16.2) | 775 | 171 (22.1) | 192 (24.8) | 342 (44.1) | 70 (9.0) | ||
| Clinical N stage | .013 | .002 | ||||||||||
| cN0 | 225 | 43 (19.1) | 57 (25.3) | 93 (41.8) | 31 (13.8) | 431 | 113 (26.2) | 97 (22.5) | 164 (38.1) | 57 (13.2) | ||
| cN1 | 371 | 81 (21.8) | 86 (23.2) | 133 (35.8) | 71 (19.1) | 981 | 217 (22.1) | 234 (23.9) | 429 (43.7) | 101 (10.3) | ||
| cN2 | 344 | 65 (18.9) | 115 (33.4) | 103 (29.9) | 61 (17.7) | 744 | 141 (19.0) | 208 (28.0) | 334 (44.9) | 61 (8.2) | ||
| Location from anal verge, cm | .008 | <.001 | ||||||||||
| 0–5 | 469 | 110 (23.5) | 122 (26.0) | 178 (38.0) | 59 (12.6) | 1,176 | 259 (22.0) | 306 (26.0) | 513 (43.6) | 98 (8.3) | ||
| 5–10 | 399 | 70 (17.5) | 117 (29.3) | 125 (31.3) | 87 (21.8) | 855 | 197 (23.0) | 192 (22.5) | 358 (41.9) | 108 (12.6) | ||
| >10 | 49 | 6 (12.2) | 13 (26.5) | 18 (36.7) | 12 (24.5) | 71 | 10 (14.1) | 14 (19.7) | 37 (52.1) | 10 (14.1) | ||
| Unknown/missing | 23 | 3 (13.0) | 6 (26.1) | 9 (39.1) | 5 (21.7) | 54 | 5 (9.3) | 27 (50.0) | 19 (35.2) | 3 (5.6) | ||
| Tumor differentiation | <.001 | <.001 | ||||||||||
| Well‐differentiated | 284 | 50 (17.6) | 77 (27.1) | 101 (35.6) | 56 (19.7) | 179 | 32 (17.9) | 52 (29.1) | 71 (39.7) | 24 (13.4) | ||
| Moderately differentiated | 527 | 97 (18.4) | 143 (27.1) | 196 (37.2) | 91 (17.3) | 1,347 | 264 (19.6) | 334 (24.8) | 599 (44.5) | 150 (11.1) | ||
| Poorly differentiated | 100 | 24 (24.0) | 30 (30.0) | 30 (30.0) | 16 (16.0) | 352 | 78 (22.2) | 90 (25.6) | 148 (42.0) | 36 (10.2) | ||
| Unknown/missing | 29 | 18 (62.1) | 8 (27.6) | 3 (10.3) | 0 (0.0) | 278 | 97 (34.9) | 63 (22.7) | 109 (39.2) | 9 (3.2) | ||
| Completeness of local resection | .021 | .007 | ||||||||||
| R0 | 936 | 189 (20.2) | 258 (27.6) | 329 (35.1) | 160 (17.1) | 2,143 | 470 (21.9) | 537 (25.1) | 922 (43.0) | 214 (10.0) | ||
| R1 | 4 | 0 (0.0) | 0 (0.0) | 1 (25.0) | 3 (75.0) | 13 | 1 (7.7) | 2 (15.4) | 5 (38.5) | 5 (38.5) | ||
| ypT stage | <.001 | <.001 | ||||||||||
| ypT0 | 199 | 189 (95.0) | 10 (5.0) | 0 (0.0) | 0 (0.0) | 490 | 471 (96.1) | 19 (3.9) | 0 (0.0) | 0 (0.0) | ||
| ypT1 | 72 | 0 (0.0) | 57 (79.2) | 13 (18.1) | 2 (2.8) | 108 | 0 (0.0) | 70 (64.8) | 37 (34.3) | 1 (0.9) | ||
| ypT2 | 212 | 0 (0.0) | 98 (46.2) | 86 (40.6) | 28 (13.2) | 484 | 0 (0.0) | 207 (42.8) | 243 (50.2) | 34 (7.0) | ||
| ypT3 | 429 | 0 (0.0) | 87 (20.3) | 219 (51.0) | 123 (28.7) | 754 | 0 (0.0) | 168 (22.3) | 462 (61.3) | 124 (16.4) | ||
| ypT4 | 28 | 0 (0.0) | 6 (21.4) | 12 (42.9) | 10 (35.7) | 320 | 0 (0.0) | 75 (23.4) | 185 (57.8) | 60 (18.8) | ||
| ypN stage | <.001 | <.001 | ||||||||||
| ypN0 | 748 | 189 (25.3) | 203 (27.1) | 246 (32.9) | 110 (14.7) | 1,641 | 471 (28.7) | 399 (24.3) | 637 (38.8) | 134 (8.2) | ||
| ypN1 | 138 | 0 (0.0) | 45 (32.6) | 56 (40.6) | 37 (26.8) | 410 | <0.001 | 123 (30.0) | 222 (54.1) | 65 (15.9) | ||
| ypN2 | 54 | 0 (0.0) | 10 (18.5) | 28 (51.9) | 16 (29.6) | 105 | <0.001 | 17 (16.2) | 68 (64.8) | 20 (19.0) | ||
| ypTN stage | <.001 | <.001 | ||||||||||
| ypT0N0 | 189 | 189 (100) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 471 | 471 (100) | 0 (0.0) | 0 (0.0) | 0 (0.0) | ||
| I | 258 | 0 (0.0) | 138 (53.5) | 92 (35.7) | 28 (10.9) | 496 | 0 (0.0) | 235 (47.4) | 228 (46.0) | 33 (6.7) | ||
| IIA | 283 | 0 (0.0) | 60 (21.2) | 147 (51.9) | 76 (26.9) | 493 | 0 (0.0) | 123 (24.9) | 302 (61.3) | 68 (13.8) | ||
| IIB | 14 | 0 (0.0) | 5 (35.7) | 4 (28.6) | 5 (35.7) | 180 | 0 (0.0) | 46 (25.6) | 98 (54.4) | 36 (20.0) | ||
| IIC | 4 | 0 (0.0) | 0 (0.0) | 2 (50.0) | 2 (50.0) | 24 | 0 (0.0) | 4 (16.7) | 15 (62.5) | 5 (20.8) | ||
| IIIA | 30 | 0 (0.0) | 24 (80.0) | 6 (20.0) | 0 (0.0) | 108 | 0 (0.0) | 59 (54.6) | 46 (42.6) | 3 (2.8) | ||
| IIIB | 143 | 0 (0.0) | 28 (19.6) | 68 (47.6) | 47 (32.9) | 325 | 0 (0.0) | 58 (17.8) | 204 (62.8) | 63 (19.4) | ||
| IIIC | 19 | 0 (0.0) | 3 (15.8) | 11 (57.9) | 5 (26.3) | 59 | 0 (0.0) | 14 (23.7) | 34 (57.6) | 11 (18.6) | ||
| Vascular invasion | <.001 | <.001 | ||||||||||
| Negative | 912 | 189 (20.7) | 257 (28.2) | 313 (34.3) | 153 (16.8) | 2,090 | 470 (22.5) | 531 (25.4) | 888 (42.5) | 201 (9.6) | ||
| Positive | 28 | 0 (0.0) | 1 (3.6) | 17 (60.7) | 10 (35.7) | 66 | 1 (1.5) | 8 (12.1) | 39 (59.1) | 18 (27.3) | ||
| Neural invasion | <.001 | <.001 | ||||||||||
| Negative | 882 | 189 (21.4) | 251 (28.5) | 297 (33.7) | 145 (16.4) | 2042 | 470 (23.0) | 527 (25.8) | 849 (41.6) | 196 (9.6) | ||
| Positive | 58 | 0 (0.0) | 7 (12.1) | 33 (56.9) | 18 (31.0) | 114 | 1 (0.9) | 12 (10.5) | 78 (68.4) | 23 (20.2) | ||
| Circumferential resection margin, mm | .079 | .01 | ||||||||||
| ≤1 | 9 | 0 (0.0) | 1 (11.1) | 4 (44.4) | 4 (44.4) | 25 | 0 (0.0) | 4(16.0) | 16 (64.0) | 5 (20.0) | ||
| >1 | 931 | 189 (20.3) | 257 (27.6) | 326 (35.0) | 159 (17.1) | 2,131 | 471 (22.1) | 535 (25.1) | 911 (42.7) | 214 (10.0) | ||
Abbreviation: TRG, tumor regression grading.
The interobserver variability of the AJCC/CAP TRG categories evaluation between the two pathologists showed an excellent agreement, with a κ value of 0.93. Among these 3,096 patients, 660 patients (21.3%) were categorized as TRG0 (discovery cohort: 189/940, 20.1%; validation cohort: 471/2156, 21.8%), 797 patients (25.7%) were categorized as TRG1 (discovery cohort: 258/940, 27.4%; validation cohort: 539/2156, 25.0%), 1,257 patients (40.6%) were categorized as TRG2 (discovery cohort: 330/940, 35.1%; validation cohort: 927/2156, 43.0%), and 382 patients (12.3%) were categorized as TRG3 (discovery cohort: 163/940, 17.3%; validation cohort: 219/2156, 10.2%).
In the discovery cohort, a tendency toward statistical significance between ypT (all p < .001; supplemental online Fig. 1A–D) or ypN (all p < .001; supplemental online Fig. 1E–H) with patient outcomes (OS, DFS, LRFS, and DMFS) was noted (supplemental online Table 1). In the enlarged size of validation cohort, the similar finding was confirmed (p < .001; Table 2; supplemental online Fig. 2). In particular, a significant survival difference of OS in both subsets was mainly observed at ypT3–4 versus ypT0–2 (discovery set, HR, 2.665; 95% CI, 1.658–4.283; validation set, HR, 2.874; 95% CI, 2.265–3.647; with all p < .001), as well as ypN1–2 versus ypN0 (discovery set, HR, 3.754; 95% CI, 2.439–5.778; validation set, HR, 2.783; 95% CI, 2.241–3.457; with all p < .001). Similar findings were obtained in DFS, LRFS, and DMFS (supplemental online Fig. 1–2). However, a the similar five‐year survival ratio was detected in other pairwise comparison (all p > .05, supplemental online Table 2). Furthermore, a close correlation between ypT or ypN stage with risk to disease progression was observed at ypT0–2 versus ypT3–4 (p < .001, supplemental online Fig. 3) and ypN0 versus ypN1–2 (p < .001, supplemental online Fig. 4) but not in other pairwise comparison.
Table 2.
Univariate and multivariate analysis of risk of death, disease relapse, local recurrence, and distant metastasis in the validation cohort
| Parameter | Univariate analysis | Multivariate analysis | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 5‐yr OS | 5‐yr DFS | 5‐yr LRFS | 5‐yr DMFS | 5‐yr OS | 5‐yr DFS | 5‐yr LRFS | 5‐yr DMFS | |||||||||
| HR (95% CI) | p value | HR (95% CI) | p value | HR (95% CI) | p value | HR (95% CI) | p value | HR (95% CI) | p value | HR (95% CI) | p value | HR (95% CI) | p value | HR (95% CI) | p value | |
| AJCC/CAP | <0.001 | <.001 | <.001 | <.001 | <.001 | <.001 | <.001 | <.001 | ||||||||
| TRG0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
| TRG1 | 2.060 (1.377–3.081) | <0.001 | 2.020 (1.431–2.851) | .001 | 4.470 (1.861–10.740) | .001 | 1.971 (1.384–2.807) | <.001 | 1.558 (1.029–2.361) | .036 | 1.541 (1.079–2.201) | .017 | 3.543 (1.453–8.636) | .005 | 1.514 (1.050–2.182) | .026 |
| TRG2 | 2.786 (1.926–4.029) | <0.001 | 2.907 (2.125–3.975) | <.001 | 6.600 (2.871–15.175) | <.001 | 2.692 (1.951–3.715) | <.001 | 2.069 (1.406–3.046) | <.001 | 2.078 (1.493–2.891) | <.001 | 4.989 (2.128–11.697) | <.001 | 1.911 (1.360–2.686) | <.001 |
| TRG3 | 3.843 (2.516–5.871) | <0.001 | 4.369 (3.047–6.265) | <.001 | 11.828 (4.923–28.419) | <.001 | 4.067 (2.802–5.904) | <.001 | 2.666 (1.700–4.181) | <.001 | 3.090 (2.102–4.542) | <.001 | 8.433 (3.395–20.949) | <.001 | 2.872 (1.927–4.279) | <.001 |
| Age, yr | ||||||||||||||||
| >55 vs. ≤55 | 1.326 (1.068–1.647) | 0.011 | 0.994 (0.829–1.192) | .050 | 0.716 (0.512–1.000) | .050 | 1.076 (0.889–1.302) | .450 | 1.474 (1.183–1.836) | .001 | 1.079 (0.898–1.297) | .415 | 0.756 (0.539–1.060) | .104 | 1.164 (0.960–1.412) | .123 |
| Sex | ||||||||||||||||
| Female vs. male | 0.973 (0.776–1.221) | 0.815 | 1.072 (0.887–1.295) | .871 | 1.029 (0.726–1.460) | .871 | 1.058 (0.867–1.291) | .576 | ||||||||
| Clinical T stage | 0.231 | .447 | .447 | .141 | ||||||||||||
| cT1 | 1 | 1 | 1 | 1 | ||||||||||||
| cT2 | 549.875 (<0.001–3.910E+31) | 0.852 | 1.029 (0.134–7.912) | .913 | 226.163 (<0.001–3.512E+44) | .913 | 1.102 (0.143–8.476) | .926 | ||||||||
| cT3 | 1,182.509 (<0.001–8.370E+31) | 0.835 | 1.572 (0.221–11.195) | .887 | 1,137.130 (<0.001–1.732E+45) | .887 | 1.536 (0.216–10.944) | .668 | ||||||||
| cT4 | 1,042.654 (<0.001–7.381E+31) | 0.838 | 1.313 (0.184–9.379) | .887 | 1,165.511 (<0.001–1.775E+45) | .887 | 1.227 (0.172–8.774) | .839 | ||||||||
| Clinical N stage | 0.604 | .519 | .519 | .107 | ||||||||||||
| cN0 | 1 | 1 | 1 | 1 | ||||||||||||
| cN1 | 1.116 (0.841–1.479) | 0.448 | 1.033 (0.807–1.322) | .597 | 1.133 (0.714–1.796) | .597 | 1.007 (0.778–1.302) | .960 | ||||||||
| cN2 | 1.158 (0.866–1.548) | 0.322 | 1.269 (0.989–1.627) | .270 | 1.303 (0.814–2.083) | .270 | 1.240 (0.956–1.607) | .105 | ||||||||
| Location from anal verge, cm | 0.030 | .168 | .168 | .001 | .004 | <.001 | .073 | <.001 | ||||||||
| 0–5 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
| 5–10 | 0.712 (0.562–0.903) | 0.005 | 0.664 (0.543–0.812) | .031 | 0.665 (0.460–0.963) | .031 | 0.667 (0.539–0.824) | <.001 | 0.664 (0.522–0.844) | .001 | 0.624 (0.509–0.765) | <.001 | 0.612 (0.421–0.890) | .010 | 0.632 (0.511–0.783) | <.001 |
| >10 | 0.598 (0.266–1.346) | 0.214 | 1.176 (0.721–1.917) | .869 | 1.079 (0.438–2.656) | .869 | 1.317 (0.806–2.150) | .271 | 0.525 (0.232–1.185) | .121 | 1.061 (0.648–1.736) | .814 | 1.059 (0.428–2.618) | .901 | 1.184 (0.723–1.942) | .502 |
| Unknown/missing | 0.804 (0.413–1.567) | 0.522 | 0.695 (0.370–1.306) | .259 | 0.689 (0.218–2.178) | .526 | 0.686 (0.353–1.333) | .266 | 0.690 (0.352–1.353) | .280 | 0.649 (0.344–1.225) | .182 | 0.697 (0.219–2.222) | .542 | 0.608 (0.312–1.188) | .145 |
| Tumor differentiation | 0.001 | <.001 | <.001 | .004 | .011 | <.001 | .004 | .013 | ||||||||
| Well‐differentiated | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
| Moderately differentiated | 0.664 (0.466–0.947) | 0.024 | 0.656 (0.483–0.893) | .009 | 0.501 (0.300–0.839) | .009 | 0.757 (0.540–1.062) | .107 | 0.644 (0.451–0.921) | .016 | 0.634 (0.465–0.865) | .004 | 0.504 (0.300–0.848) | .010 | 0.740 (0.526–1.039) | .082 |
| Poorly differentiated | 1.043 (0.702–1.549) | 0.837 | 1.076 (0.764–1.515) | .878 | 1.045 (0.596–1.831) | .878 | 1.161 (0.798–1.688) | .436 | 0.933 (0.623–1.398) | .737 | 0.922 (0.650–1.308) | .651 | 0.944 (0.530–1.679) | .843 | 0.992 (0.677–1.452) | .966 |
| Unknown/missing | 0.547 (0.324–0.923) | 0.024 | 0.788 (0.532–1.166) | .233 | 0.355 (0.159–0.791) | .011 | 0.907 (0.594–1.384) | .651 | 0.675 (0.398–1.145) | .145 | 0.962 (0.646–1.434) | .851 | 0.482 (0.215–1.082) | .077 | 1.104 (0.719–1.694) | .651 |
| Completeness of local resection | ||||||||||||||||
| R1 vs. R0 | 1.743 (0.434–7.007) | 0.434 | 2.018 (0.754–5.404) | .628 | 1.628 (0.227–11.647) | .628 | 2.290 (0.855–6.134) | .099 | ||||||||
| ypT stage | <0.001 | <.001 | <.001 | <.001 | ||||||||||||
| ypT0 | 1 | 1 | 1 | 1 | ||||||||||||
| ypT1 | 1.095 (0.511–2.347) | 0.816 | 1.118 (0.596–2.094) | .294 | 2.062 (0.533–7.974) | .294 | 1.171 (0.624–2.199) | .623 | ||||||||
| ypT2 | 1.260 (0.823–1.929) | 0.287 | 1.269 (0.883–1.826) | .049 | 2.417 (1.002–5.828) | .049 | 1.198 (0.823–1.743) | .347 | ||||||||
| ypT3 | 3.365 (2.371–4.775) | <0.001 | 3.663 (2.717–4.939) | <.001 | 8.147 (3.764–17.635) | <.001 | 3.435 (2.528–4.667) | <.001 | ||||||||
| ypT4 | 2.927 (1.956–4.380) | <0.001 | 2.974 (2.112–4.188) | <.001 | 7.326 (3.225–16.640) | <.001 | 2.682 (1.880–3.826) | <.001 | ||||||||
| ypN stage | <0.001 | <.001 | <.001 | <.001 | <.001 | <.001 | <.001 | <.001 | ||||||||
| ypN0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
| ypN1 | 2.557 (2.016–3.243) | <0.001 | 2.958 (2.427–3.604) | <.001 | 2.867 (1.993–4.124) | <.001 | 2.866 (2.328–3.529) | <.001 | 2.165 (1.686–2.780) | <.001 | 2.376 (1.928–2.928) | <.001 | 1.992 (1.368–2.900) | <.001 | 2.324 (1.864–2.898) | <.001 |
| ypN2 | 3.717 (2.614–5.285) | <0.001 | 3.893 (2.861–5.297) | <.001 | 4.282 (2.527–7.258) | <.001 | 3.776 (2.729–5.224) | <.001 | 2.898 (1.986–4.227) | <.001 | 2.734 (1.961–3.814) | <.001 | 2.408 (1.368–4.240) | .002 | 2.814 (1.986–3.987) | <.001 |
| ypTN stage | <0.001 | <.001 | <.001 | <.001 | ||||||||||||
| ypT0N0 | 1 | 1 | 1 | 1 | ||||||||||||
| I | 1.225 (0.783–1.916) | 0.374 | 1.079 (0.730–1.595) | .054 | 2.518 (0.985–6.435) | .054 | 1.000 (0.666–1.501) | .999 | ||||||||
| IIA | 2.638 (1.776–3.916) | <0.001 | 2.924 (2.094–4.084) | <.001 | 6.588 (2.789–15.562) | <.001 | 2.786 (1.976–3.928) | <.001 | ||||||||
| IIB | 2.349 (1.430–3.856) | 0.001 | 1.947 (1.252–3.028) | <.001 | 5.742 (2.182–15.109) | <.001 | 1.907 (1.212–3.001) | .005 | ||||||||
| IIC | 3.730 (1.566–8.887) | 0.003 | 3.674 (1.736–7.776) | .017 | 7.030 (1.419–34.831) | .017 | 3.777 (1.780–8.012 | .001 | ||||||||
| IIIA | 2.163 (1.178–3.971) | 0.013 | 2.793 (1.730–4.510) | .025 | 3.884 (1.185–12.728) | .025 | 2.829 (1.735–4.614) | <.001 | ||||||||
| IIIB | 5.257 (3.561–7.760) | <0.001 | 5.772 (4.147–8.034) | <.001 | 12.762 (5.449–29.890) | <.001 | 5.374 (3.822–7.557) | <.001 | ||||||||
| IIIC | 6.263 (3.604–10.885) | <0.001 | 7.790 (4.925–12.320) | <.001 | 20.924 (7.851–55.766) | <.001 | 6.215 (3.786–10.203) | <.001 | ||||||||
| Vascular invasion | ||||||||||||||||
| Positive vs. negative | 1.569 (0.919–2.680) | 0.099 | 1.945 (1.279–2.957) | .010 | 2.429 (1.236–4.775) | .010 | 1.777 (1.134–2.784) | .012 | 1.096 (0.621–1.934) | .753 | 1.226 (0.785–1.917) | .370 | 1.492 (0.716–3.108) | .285 | 1.032 (0.638–1.670) | .898 |
| Neural invasion | ||||||||||||||||
| Positive vs. negative | 1.446 (0.886–2.361) | 0.140 | 2.231 (1.599–3.112) | .107 | 1.700 (0.892–3.238) | .107 | 2.381 (1.697–3.341) | <.001 | 0.900 (0.538–1.507) | .689 | 1.345 (0.947–1.912) | .098 | 0.834 (0.419–1.659) | .605 | 1.498 (1.044–2.149) | .028 |
| Circumferential resection margin | ||||||||||||||||
| ≤1 vs. >1 mm | 2.798 (1.156–6.777) | 0.023 | 2.188 (1.036–4.618) | .340 | 1.974 (0.488–7.981) | .340 | 2.493 (1.180–5.267) | .017 | 2.007 (0.820–4.912) | .127 | 1.367 (0.641–2.917) | .418 | 1.334 (0.325–5.483) | .689 | 1.582 (0.740–3.381) | .236 |
Abbreviations: AJCC/CAP, American Joint Committee on Cancer/College of American Pathologists; CI, confidence interval; DFS, disease‐free survival; DMFS, distant metastasis‐free survival; HR, hazard ratio; LRFS, local recurrence‐free survival; OS, overall survival; TRG, tumor regression grading.
The AJCC/CAP TRG System Predicted Survival and Risk of Disease Progression
The median follow‐up time was 41 months. To test the prognostic value of the AJCC/CAP TRG system, the discovery cohort patients were subjected to Kaplan‐Meier survival analysis. As shown, differences of survival outcomes (OS, DFS, LRFS, and DMFS) in the four AJCC/CAP TRG categories were detected in the discovery cohort (all p < .001; supplemental online Fig. 5A–D). The same findings were further demonstrated in the validation cohort (all p < .001; Fig. 1A–D). Importantly, pairwise comparison showed that any two of the four TRG categories had a distinguished outcome (OS, DFS, LRFS, DMFS) ratio in the validation cohort (except TRG1 vs. TRG2 LRFS, p = .068; all other p < .05, Table 2; supplemental online Table 4).
Figure 1.
Kaplan‐Meier curve analysis of overall survival (A), disease‐free survival (B), LRFS (C), and DMFS (D), comparing the validation cohort patients in different American Joint Committee on Cancer/College of American Pathologists TRG categories. Abbreviations: CI, confidence interval; HR, hazard ratio; TRG, tumor regression grading.
Similarly, in the validation cohort, apparent differences were observed between any two of the four AJCC/CAP TRG categories regarding risk of death, accumulative relapse, local recurrence, and distant metastasis (Fig. 2).
Figure 2.
Kaplan‐Meier curve analysis of risk of death (A), disease relapse (B), local recurrence (C), and distant metastasis (D), comparing the validation cohort patients in different American Joint Committee on Cancer/College of American Pathologists TRG categories. Abbreviations: CI, confidence interval; HR, hazard ratio; TRG, tumor regression grading.
The AJCC/CAP TRG System Guided Personalized Adjuvant Chemotherapy Treatment
In light of the favorable outcome of the AJCC/CAP TRG0 subgroup, we further studied if AJCC/CAP TRG0 patients would benefit from adjuvant chemotherapy. After propensity score matching of age, gender, cT stage, cN stage, tumor differentiation, and location from anal verge. The Kaplan‐Meier analysis revealed that the AJCC/CAP TRG0 category patients treated with or without adjuvant chemotherapy had similar 5‐year OS, DFS, LRFS, and DMFS (p = .633, HR, 1.18; 95% CI, 0.586–2.404; p = .213, HR, 0.648; 95% CI, 0.326–1.291; p = .087, HR, 0.03; 95% CI, 0–23.196, p = .295, HR, 0.646; 95% CI, 0.324–1.286; Fig. 3). By contrast, the AJCC/CAP TRG1–3 subset patients treated without adjuvant chemotherapy had a worse 5‐year OS than those with adjuvant chemotherapy (HR, 1.75; 95% CI, 1.07–2.87; supplemental online Fig. 6A; HR, 1.6; 95% CI, 1.13–2.28; supplemental online Fig. 7A; HR, 1.68; 95% CI, 1–2.86; supplemental online Fig. 8A; with all p < .05). However, the AJCC/CAP TRG1–3 subset patients treated with or without adjuvant chemotherapy had similar DFS, LRFS, and DMFS (supplemental online Figs. 6B–D, 7B–D, and Fig. 8B–D, with all p > .05).
Figure 3.
Kaplan‐Meier curve analysis of oval survival (A), disease‐free survival (B), local recurrence‐free survival (C), and distant metastasis‐free survival (D) for American Joint Committee on Cancer/College of American Pathologists tumor regression grading 0 category patients treated with or without adjuvant chemotherapy. Abbreviations: CI, confidence interval; HR, hazard ratio.
Multivariate Analysis
All factors that had significance in univariate analysis were included in the Cox regression multivariate analysis. In the validation cohort, the AJCC/CAP TRG category, age, tumor distance from anal verge, tumor differentiation status, and ypN stage were independent prognosticators for OS (Table 2). The AJCC/CAP TRG category, tumor distance from anal verge, tumor differentiation status, and ypN stage were also significant independent prognostic factors for disease progression and distant metastasis. The AJCC/CAP TRG category, tumor differentiation status, and ypN stage were associated with local recurrence. Nonetheless, prognostic significance of tumor distance from anal verge and differentiation status was not observed in pairwise comparison in the validation cohort. By contrast, the prognostic value of the AJCC/CAP TRG system was demonstrated in any two of the four categories at validation cohort. Altogether, the AJCC/CAP TRG four‐category system was an independent prognostic factor for risk of death, disease relapse, local recurrence, and distant metastasis for LARC.
Discussion
This is the largest cohort study to date to detect the prognostic value of AJCC/CAP TRG system on outcomes in patients with LARC. The AJCC/CAP TRG system appears to be an independent prognostic factor for LARC. Importantly, pairwise survival comparison showed that any two of the four AJCC/CAP TRG categories were significantly different, not only in outcomes but also in risk of disease progression. In addition, adding adjuvant chemotherapy to AJCC/CAP TRG0 subgroup patients did not improve their long‐term outcomes. By contrast, the adjuvant chemotherapy conferred an OS benefit to patients of the AJCC/CAP TRG1–3 categories. The multivariate regression analysis finally confirmed that the AJCC/CAP TRG system was indeed an independent prognostic surrogate for LARC. Taken together, our findings highlighted the importance of the AJCC/CAP TRG system in the clinical care of patients with LARC, not only in prognosis but also in guiding the adjuvant chemotherapy selection.
The reported three‐ to five‐category TRG systems all has two extremes: pCR and nonresponse [2, 11, 12]. However, the arrangement of intermediate parts between these two extremes is different [12]. The five‐category TRG system (Mandard, Dowrak/Rödel) had three intermediate parts (TRG1–3), which was too sensitive to diverge the subset patients, because the difference between intermediate parts (TRG1 vs. TRG2, or TRG2 vs. TRG3) was too subtle. When the five‐category TRG system (Mandard, Dowrak/Rödel) was merged into a three‐category system, the three intermediate parts were collapsed to one intermediate part accordingly. Also, the single intermediate part integrated the individual not only close to pCR but also close to nonresponse, which was highly heterogeneous in pathologic response to nCRT. In contrast, the AJCC/CAP TRG system divides the intermediate part into two subgroups, near pCR (TRG1) and near nonresponse (TRG2). This arrangement increased the reproducibility during the pathologic evaluation process, because each category would be easily defined by pathologists. Indeed, interobservers variability of present study was small between different pathologists, suggesting AJCC/CAP TRG system is highly reproducible.
In 2014, Trakarnsanga et al. compared the accuracies of the well‐established TRG systems in a cohort of 563 patients with LARC and found that the four‐category AJCC/CAP TRG system was more accurately prognostic for recurrence and had the highest concordance indices (0.694) than the three‐category Mandard/Dowrak/Memorial Sloan‐Kettering Cancer Center TRG systems [12]. Here, we proved the prognostic value of AJCC/CAP TRG system in 3,096 patients with LARC (Fig. 1; Table 2). More importantly, the Kaplan‐Meier curves had no overlap between any two of the four AJCC/CAP TRG categories in the validation cohort (Fig. 1; Table 2; supplemental online Table 4), demonstrating that the AJCC/CAP TRG system is powerful for prognosis.
An ideal tumor biomarker would advance precision medicine by prognosticating for populations in general and determining individual risk [16, 17]. Our study raised another important implementation of the AJCC/CAP TRG system in modifying individual risk of disease progression. Previously, the ypT and/or ypN stage were reported to be powerful prognosticators [8, 18, 19, 20, 21, 22]. In this study, we found that the Kaplan‐Meier curves for outcomes (OS, DFS, LRFS, and DMFS) and risk of disease progression (death, disease relapse, local recurrence, and distant metastasis) had overlap between some consecutive two categories (ypT0 vs. ypT1, ypT0 vs. ypT2, ypT1 vs. ypT2, ypT3 vs. ypT4, ypN1 vs. ypN2; supplemental online Figs. 1–4; supplemental online Table 1–3). Therefore, the AJCC/CAP TRG system is more accurate in defining risk of disease progression than the other routinely used clinical factors for LARC.
Currently, the NCCN guidelines recommend adjuvant chemotherapy for LARC, although the subgroup of patients who would benefit from adjuvant chemotherapy remains controversial [23, 24]. Take AJCC/CAP TRG0 (pCR) patients for example, the low disease progression rate questioned the need to administer adjuvant chemotherapy after TME [25, 26, 27]. Our findings suggested that, for the AJCC/CAP TRG0 category subgroup, adjuvant chemotherapy did not confer any advantage to outcomes (OS, DFS, LRFS, and DMFS) or reduce the risk of disease progression (Fig. 3). In accordance with our FOWARC trial and other studies [28, 29, 30], our findings also suggested that patients with poor TRG response (AJCC/CAP TRG1–3 categories) would benefit from postoperative chemotherapy (supplemental online Figs. 6A, 7A, 8A) and should be considered for intensified adjuvant chemotherapy. In addition, because of the favorable outcome of tumor complete regression (ypT0 cases), the organ‐preserving strategy, including local excision without lymphadenectomy and Watch and Wait [31, 32, 33], was raised, aiming to achieve a high quality of life, such as long‐term urinary, sexual, and defecation functions. However, caution is required before using such an approach, because 5%–10% of ypT0 patients (4.2% in our study) have been observed to be ypN+ in the residual mesorectum area [11].
Limitations
This study has limitations. First, it was a retrospectively designed, multicenter study and was prone to bias from loss of follow‐up and missed events of death and recurrences. Second, the median follow‐up was 41 months, and only 38.3% of cases had more than 50 months of follow‐up. Moreover, lack of standardization in neoadjuvant and adjuvant chemotherapy potentially affected the long‐term outcome. We overcame these limitations by incorporating a large and prospectively maintained cohort, which enabled us to confer enough power to evaluate the survival differences. Importantly, we demonstrated our discovery set finding at an independent validation set to guarantee the clinical significance of AJCC/CAP TRG system was indeed uniform.
Conclusion
Based on this large size cohort study, the four‐category AJCC/CAP TRG system has proved to be powerful in predicting individual prognosis and guiding the adjuvant chemotherapy strategy selection for patients with LARC.
Author Contributions
Conception/design: Hai‐Yang Chen, Li‐Li Feng, Ming Li, Huai‐Qiang Ju, Yi Ding, Mei Lan, Jaffer A Ajani, Xiang‐Bo Wan
Provision of study material or patients: Ming Li, Huai‐Qiang Ju, Yi Ding, Mei Lan, Wei‐Dong Han, Jin‐Yi Lang, Xiang‐Bo Wan
Collection and/or assembly of data: Li Yu, Ming‐Biao Wei, Xiao‐Lin Pang, Fang He, Shuai Liu, Wei‐Dong Han
Data analysis and interpretation: Hai‐Yang Chen, Jian Zheng, Yan Ma, Chu‐Yang Lin, Xiang‐Bo Wan
Manuscript writing: Hai‐Yang Chen, Ping Lan, Mei‐Jin Huang, Yi‐Feng Zou, Zu‐Li Yang, Ting Wang, Jin‐Yi Lang, Guy R. Orangio, Vitaliy Poylin, Xiang‐Bo Wan
Final approval of manuscript: Hai‐Yang Chen, Li‐Li Feng, Ming Li, Huai‐Qiang Ju, Yi Ding, Mei Lan, Shu‐Mei Song, Wei‐Dong Han, Li Yu, Ming‐Biao Wei, Xiao‐Lin Pang, Fang He, Shuai Liu, Jian Zheng, Yan Ma, Chu‐Yang Lin, Ping Lan, Mei‐Jin Huang, Yi‐Feng Zou, Zu‐Li Yang, Ting Wang, Jin‐Yi Lang, Guy R. Orangio, Vitaliy Poylin, Jaffer A Ajani, Wei‐Hu Wang, Xiang‐Bo Wan
Disclosures
The authors indicated no financial relationships.
Supporting information
See http://www.TheOncologist.com for supplemental material available online.
Supplementary Figure 1 Kaplan‐Meier curve analysis of OS (A), DFS (B), LRFS (C), DMFS (D), comparing discovery cohort patients of different ypT stages. Kaplan‐Meier curve analysis of OS (E), DFS (F), LRFS (G), and DMFS (H), comparing discovery cohort patients of different ypN stages.
Supplementary Figure 2. Kaplan‐Meier curve analysis of OS (A), DFS (B), LRFS (C), and DMFS (D), comparing validation cohort patients of different ypT stages. Kaplan‐Meier curve analysis of OS (E), DFS (F), LRFS (G), and DMFS (H), comparing validation cohort patients of different ypN stages.
Supplementary Figure 3. Kaplan‐Meier curve analysis of risk of death (A), disease relapse (B), local recurrence (C) and distant metastasis (D), comparing discovery cohort patients of different ypT stages. Kaplan‐Meier curve analysis of risk of death (E), disease relapse (F), local recurrence (G) and distant metastasis (H), comparing validation cohort patients of different ypT stages.
Supplementary Figure 4. Kaplan‐Meier curve analysis of risk of death (A), disease relapse (B), local recurrence (C) and distant metastasis (D), comparing the discovery cohort patients of different ypN stages. Kaplan‐Meier curve analysis of risk of death (E), disease relapse (F), local recurrence (G) and distant metastasis (H), comparing validation cohort patients of different ypN stages.
Supplementary Figure 5. Kaplan‐Meier curve analysis of OS (A), DFS (B), LRFS (C), and DMFS (D), comparing the discovery cohort patients in different AJCC/CAP TRG categories.
Supplementary Figure 6. Kaplan‐Meier curve analysis of OS (A), DFS (B), LRFS (C), and DMFS (D) for AJCC/CAP TRG1 category patients treated with or without adjuvant chemotherapy.
Supplementary Figure 7. Kaplan‐Meier curve analysis of OS (A), DFS (B), LRFS (C), and DMFS (D) for AJCC/CAP TRG2 category patients treated with or without adjuvant chemotherapy.
Supplementary Figure 8 . Kaplan‐Meier curve analysis of OS (A), DFS (B), LRFS (C), and DMFS (D) for AJCC/CAP TRG3 category patients treated with or without adjuvant chemotherapy.
Supplementary Table 1 Univariate analysis of risk of death, disease relapse, local recurrence and distant metastasis in discovery cohort
Supplementary Table 2. Pairwise Kaplan‐Meier curve comparisons of ypT stages in discovery and validation cohort
Supplementary Table 3. Pairwise Kaplan‐Meier curve comparisons of ypN stages in discovery and validation cohort
Supplementary Table 4. Pairwise Kaplan‐Meier curve comparisons of TRG categories in validation cohort
Supplementary Table 5. Description of AJCC/CAP TRG
Acknowledgments
With deepest grief and highest reverence, the authors gratefully thank the support of Dr. Lei Wang, ex‐dean of the Sixth Affiliated Hospital of Sun Yat‐sen University, who devoted his whole life to colorectal cancer and radiation enteritis field. His noteworthy dedication to the cause as well as his extraordinary benevolence, dauntlessness, and selflessness will be forever remembered and honored. Dear Dr. Wang, we miss you.
This work is supported by National Natural Science Foundation of China (81872188); National Natural Science Foundation for Young Scholars of China (81703080); International Center for Genetic Engineering and Biotechnology grant (CRP/CHN16–04_EC).
No part of this article may be reproduced, stored, or transmitted in any form or for any means without the prior permission in writing from the copyright holder. For information on purchasing reprints contact commercialreprints@wiley.com. For permission information contact permissions@wiley.com.
Disclosures of potential conflicts of interest may be found at the end of this article.
Contributor Information
Wei‐Hu Wang, Email: wangweihu88@163.com.
Xiang‐Bo Wan, Email: wanxbo@mail.sysu.edu.cn.
References
- 1. Sauer R, Becker H, Hohenberger W et al. Preoperative versus postoperative chemoradiotherapy for rectal cancer. N Engl J Med 2004;351:1731–1740. [DOI] [PubMed] [Google Scholar]
- 2. Kitz J, Fokas E, Beissbarth T et al. Association of plane of total mesorectal excision with prognosis of rectal cancer secondary analysis of the CAO/ARO/AIO‐04 phase 3 randomized clinical trial. JAMA Surg 2018;153:e181607. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Maas M, Nelemans PJ, Valentini V et al. Long‐term outcome in patients with a pathological complete response after chemoradiation for rectal cancer: A pooled analysis of individual patient data. Lancet Oncol 2010;11:835–844. [DOI] [PubMed] [Google Scholar]
- 4. Garcia‐Aguilar J, Chow OS, Smith DD et al. Effect of adding mFOLFOX6 after neoadjuvant chemoradiation in locally advanced rectal cancer: A multicentre, phase 2 trial. Lancet Oncol 2015;16:957–966. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Smith JJ, Strombom P, Chow OS et al. Assessment of a watch‐and‐wait strategy for rectal cancer in patients with a complete response after neoadjuvant therapy. JAMA Oncol 2019;5:e185896. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Fokas E, Ströbel P, Fietkau R et al. Tumor regression grading after preoperative chemoradiotherapy as a prognostic factor and individual‐level surrogate for disease‐free survival in rectal cancer. J Natl Cancer Inst 2017;109. [DOI] [PubMed] [Google Scholar]
- 7. Mandard AM, Dalibard F, Mandard JC et al. Pathologic assessment of tumor regression after preoperative chemoradiotherapy of esophageal carcinoma. Clinicopathologic correlations. Cancer 1994;73:2680–2686. [DOI] [PubMed] [Google Scholar]
- 8. Abdul‐Jalil KI, Sheehan KM, Kehoe J et al. The prognostic value of tumour regression grade following neoadjuvant chemoradiation therapy for rectal cancer. Color Dis 2014;16;O16–25. [DOI] [PubMed] [Google Scholar]
- 9. Saltz J, Gupta R, Hou L et al. Spatial organization and molecular correlation of tumor‐infiltrating lymphocytes using deep learning on pathology images. Cell Rep 2018;23:181–193.e7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Fokas E, Liersch T, Fietkau R et al. Tumor regression grading after preoperative chemoradiotherapy for locally advanced rectal carcinoma revisited: Updated results of the CAO/ARO/AIO‐94 trial. J Clin Oncol 2014;32:1554–1562. [DOI] [PubMed] [Google Scholar]
- 11. Lee YC, Hsieh CC, Chuang JP. Prognostic significance of partial tumor regression after preoperative chemoradiotherapy for rectal cancer: A meta‐analysis. Dis Colon Rectum 2013;56:1093–1101. [DOI] [PubMed] [Google Scholar]
- 12. Trakarnsanga A, Gönen M, Shia J et al. Comparison of tumor regression grade systems for locally advanced rectal cancer after multimodality treatment. J Natl Cancer Inst 2014;106:dju248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Amin MB, Edge S, Green F et al. Definitions. In: AJCC Cancer Staging Manual 8th Edition. Chicago, IL: American College of Surgeons; 2020. [Google Scholar]
- 14. Mace AG, Pai RK, Stocchi L et al. American joint committee on cancer and college of American pathologists regression grade: A new prognostic factor in rectal cancer. Dis Colon Rectum 2015;58:32–44. [DOI] [PubMed] [Google Scholar]
- 15. Xu J, Wan XB, Huang XF et al. Serologic antienzyme rate of Epstein‐Barr virus DNase‐specific neutralizing antibody segregates TNM classification in nasopharyngeal carcinoma. J Clin Oncol 2010;28:5202–5209. [DOI] [PubMed] [Google Scholar]
- 16. Fakhry C, Lacchetti C, Rooper LM et al. Human papillomavirus testing in head and neck carcinomas: ASCO clinical practice guideline endorsement of the college of American pathologists guideline. J Clin Oncol 2018;36:3152–3161. [DOI] [PubMed] [Google Scholar]
- 17. Duchesne GM, Woo HH, Bassett JK et al. Timing of androgen‐deprivation therapy in patients with prostate cancer with a rising PSA (TROG 03.06 and VCOG PR 01‐03 [TOAD]): A randomised, multicentre, non‐blinded, phase 3 trial. Lancet Oncol 2016;17:727–737. [DOI] [PubMed] [Google Scholar]
- 18. Sclafani F, Chau I, Cunningham D et al. Prognostic role of the LCS6 KRAS variant in locally advanced rectal cancer: Results of the EXPERT‐C trial. Ann Oncol 2015;26:1936–1941. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. von den Grün JM, Hartmann A, Fietkau R et al. Can clinicopathological parameters predict for lymph node metastases in ypT0‐2 rectal carcinoma? Results of the CAO/ARO/AIO‐94 and CAO/ARO/AIO‐04 phase 3 trials. Radiother Oncol 2018;128:557–563. [DOI] [PubMed] [Google Scholar]
- 20. Fokas E, Fietkau R, Hartmann A et al. Neoadjuvant rectal score as individual‐level surrogate for disease‐free survival in rectal cancer in the CAO/ARO/AIO‐04 randomized phase III trial. Ann Oncol 2018;29:1521–1527. [DOI] [PubMed] [Google Scholar]
- 21. Rubbia‐Brandt L, Giostra E, Brezault C et al. Importance of histological tumor response assessment in predicting the outcome in patients with colorectal liver metastases treated with neo‐adjuvant chemotherapy followed by liver surgery. Ann Oncol 2007;18:299–304. [DOI] [PubMed] [Google Scholar]
- 22. Kim NK, Baik SH, Seong JS et al. Oncologic outcomes after neoadjuvant chemoradiation followed by curative resection with tumor‐specific mesorectal excision for fixed locally advanced rectal cancer: Impact of postirradiated pathologic downstaging on local recurrence and survival. Ann Surg 2006;244:1024–1030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Carvalho C, Glynne‐Jones R. Challenges behind proving efficacy of adjuvant chemotherapy after preoperative chemoradiation for rectal cancer. Lancet Oncol 2017;18:e354–e363. [DOI] [PubMed] [Google Scholar]
- 24. Van den Eynde M, Mlecnik B, Bindea G et al. The link between the multiverse of immune microenvironments in metastases and the survival of colorectal cancer patients. Cancer Cell 2018;34:1012–1026.e3. [DOI] [PubMed] [Google Scholar]
- 25. Zhou J, Qiu H, Lin G et al. Is adjuvant chemotherapy necessary for patients with pathological complete response after neoadjuvant chemoradiotherapy and radical surgery in locally advanced rectal cancer? Long‐term analysis of 40 ypCR patients at a single center. Int J Colorectal Dis 2016;31:1163–1168. [DOI] [PubMed] [Google Scholar]
- 26. Maas M, Nelemans PJ, Valentini V et al. Adjuvant chemotherapy in rectal cancer: Defining subgroups who may benefit after neoadjuvant chemoradiation and resection: A pooled analysis of 3,313 patients. Int J Cancer 2015;137:212–220. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Dossa F, Acuna SA, Rickles AS et al. Association between adjuvant chemotherapy and overall survival in patients with rectal cancer and pathological complete response after neoadjuvant chemotherapy and resection. JAMA Oncol 2018;4:930–937. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Deng Y, Chi P, Lan P et al. Modified FOLFOX6 with or without radiation versus fluorouracil and leucovorin with radiation in neoadjuvant treatment of locally advanced rectal cancer: Initial results of the Chinese FOWARC multicenter, open‐label, randomized three‐arm phase III trial. J Clin Oncol 2016;34:3300–3307. [DOI] [PubMed] [Google Scholar]
- 29. Khrizman P, Niland JC, ter Veer A et al. Postoperative adjuvant chemotherapy use in patients with stage II/III rectal cancer treated with neoadjuvant therapy: A National Comprehensive Cancer Network analysis. J Clin Oncol 2013;31:30–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Hong TS, Ryan DP. Adjuvant chemotherapy for locally advanced rectal cancer: Is it a given? J Clin Oncol. 2015;33:1878–1880. [DOI] [PubMed] [Google Scholar]
- 31. Renehan AG, Malcomson L, Emsley R et al. Watch‐and‐wait approach versus surgical resection after chemoradiotherapy for patients with rectal cancer (the OnCoRe project): A propensity‐score matched cohort analysis. Lancet Oncol 2016;17:174–183. [DOI] [PubMed] [Google Scholar]
- 32. van der Valk MJM, Hilling DE, Bastiaannet E et al. Long‐term outcomes of clinical complete responders after neoadjuvant treatment for rectal cancer in the International Watch & Wait Database (IWWD): An international multicentre registry study. Lancet 2018;391:2537–2545. [DOI] [PubMed] [Google Scholar]
- 33. Rullier E, Perez RO. Surgery or a watch‐and‐wait approach for rectal cancer? Lancet Oncol 2019;20:189–190. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
See http://www.TheOncologist.com for supplemental material available online.
Supplementary Figure 1 Kaplan‐Meier curve analysis of OS (A), DFS (B), LRFS (C), DMFS (D), comparing discovery cohort patients of different ypT stages. Kaplan‐Meier curve analysis of OS (E), DFS (F), LRFS (G), and DMFS (H), comparing discovery cohort patients of different ypN stages.
Supplementary Figure 2. Kaplan‐Meier curve analysis of OS (A), DFS (B), LRFS (C), and DMFS (D), comparing validation cohort patients of different ypT stages. Kaplan‐Meier curve analysis of OS (E), DFS (F), LRFS (G), and DMFS (H), comparing validation cohort patients of different ypN stages.
Supplementary Figure 3. Kaplan‐Meier curve analysis of risk of death (A), disease relapse (B), local recurrence (C) and distant metastasis (D), comparing discovery cohort patients of different ypT stages. Kaplan‐Meier curve analysis of risk of death (E), disease relapse (F), local recurrence (G) and distant metastasis (H), comparing validation cohort patients of different ypT stages.
Supplementary Figure 4. Kaplan‐Meier curve analysis of risk of death (A), disease relapse (B), local recurrence (C) and distant metastasis (D), comparing the discovery cohort patients of different ypN stages. Kaplan‐Meier curve analysis of risk of death (E), disease relapse (F), local recurrence (G) and distant metastasis (H), comparing validation cohort patients of different ypN stages.
Supplementary Figure 5. Kaplan‐Meier curve analysis of OS (A), DFS (B), LRFS (C), and DMFS (D), comparing the discovery cohort patients in different AJCC/CAP TRG categories.
Supplementary Figure 6. Kaplan‐Meier curve analysis of OS (A), DFS (B), LRFS (C), and DMFS (D) for AJCC/CAP TRG1 category patients treated with or without adjuvant chemotherapy.
Supplementary Figure 7. Kaplan‐Meier curve analysis of OS (A), DFS (B), LRFS (C), and DMFS (D) for AJCC/CAP TRG2 category patients treated with or without adjuvant chemotherapy.
Supplementary Figure 8 . Kaplan‐Meier curve analysis of OS (A), DFS (B), LRFS (C), and DMFS (D) for AJCC/CAP TRG3 category patients treated with or without adjuvant chemotherapy.
Supplementary Table 1 Univariate analysis of risk of death, disease relapse, local recurrence and distant metastasis in discovery cohort
Supplementary Table 2. Pairwise Kaplan‐Meier curve comparisons of ypT stages in discovery and validation cohort
Supplementary Table 3. Pairwise Kaplan‐Meier curve comparisons of ypN stages in discovery and validation cohort
Supplementary Table 4. Pairwise Kaplan‐Meier curve comparisons of TRG categories in validation cohort
Supplementary Table 5. Description of AJCC/CAP TRG