Abstract
Background
Tumour deposits are suggested to impact prognosis in colon cancer negatively. This study assessed the impact of tumour deposits on oncological outcomes.
Methods
Data from the Swedish Colorectal Cancer Registry for patients who underwent R0 abdominal surgery for TNM stage I–III colon cancer between 2011 and 2014 with 5-year follow-up were analysed with multivariable analysis. Patients were categorized for their tumour deposit status and compared for the local recurrence and distant metastasis rates and 5-year survivals (overall and relative). Subgroup analyses were performed according to the nodal disease status.
Results
Of 8146 stage I–III colon cancer patients who underwent R0 resection, 8014 patients were analysed (808 tumour deposits positive, 7206 tumour deposits negative). Patients with tumour deposits positive tumours had increased local recurrence and distant metastasis rates (7.2 versus 3.0 per cent; P < 0.001 and 33.9 versus 12.0 per cent; P < 0.001 respectively) and reduced 5-year overall and relative survival (56.8 per cent versus 74.9 per cent; P < 0.001 and 68.5 versus 92.6 per cent; P < 0.001 respectively). In multivariable analysis, tumour deposits moderately increased the risks of local recurrence and distant metastasis (hazard ratio 1.50, 95 per cent c.i. 1.09 to 2.07; P = 0.013 and HR 1.91, 95 per cent c.i. 1.64 to 2.23; P < 0.001 respectively) and worse 5-year overall and relative survival (hazard ratio 1.60, 95 per cent c.i. 1.40 to 1.82; P < 0.001 and excess hazard ratio 2.24, 95 per cent c.i. 1.81 to 2.78; P < 0.001 respectively). Subgroup analysis of N stages found that N1c patients had worse outcomes than N0 for distant metastasis and relative survival. For patients with lymph node metastases tumour deposits increased the risks of distant metastasis and worse overall and relative survival, except for N2b patients.
Conclusion
Tumour deposits negatively impact the prognosis in colon cancer and must be considered when discussing adjuvant chemotherapy.
The prognostic impact of tumour deposits in a population-based cohort of patients from the Swedish Colorectal Cancer Registry was analysed. Tumour deposits negatively impacted the prognosis. This population-based study exclusively analysed the prognostic impact of TDs in colon cancer. Detailed data on local recurrence, distant metastasis and survival from a robust national registry are presented. Subgroup analyses of N stages and patients with lymph node metastases were performed.
Introduction
Colon cancer is the fifth most commonly diagnosed cancer and the fifth leading cause of cancer deaths worldwide1. The TNM stage, based on pathologic findings in surgical resection specimens and radiology, remains the key determinant of prognosis. However, there is stage-independent variability in oncological outcomes, underscored by the TNM staging system2,3. The TNM staging system considers three categories and does not account for other histopathological prognostic factors, such as tumour grading, lymphovascular and perineural invasion or tumour budding. Furthermore, neither molecular nor genetic prognostic factors, such as KRAS, BRAF or MSI mutations, are integrated.
Surgery is the primary treatment for most colon cancer patients. Guidelines recommend adjuvant chemotherapy for high-risk stage II and III patients4–6. However, recent studies suggest that subgroups of patients with locally advanced tumours may benefit from neoadjuvant chemotherapy7–9. Debatably, the presence of pericolic tumour deposits (TDs) in the absence of lymph node metastases (LNMs) was designated stage III (N1c) in the seventh edition of the TNM staging system2,10–12.
TDs are associated with poor prognosis in colon cancer10,12–18. The reported prevalence of TDs is approximately 20 per cent, but is highly dependent on the studied population10,12. Despite some recent population-based publications, earlier studies on TDs in colon cancer are relatively few and limited by retrospective single-centre design and uncertainty regarding pathology quality10,12. In many publications, colon and rectal cancer patients, subsequently impacted by neoadjuvant treatment, are merged, further complicating interpretation of the results13,14,16,17. Additionally, most studies report survival as the only outcome and recurrence data are not provided12.
This population-based study aimed to assess the impact of TDs on rates of local recurrence (LR) and distant metastasis (DM), as well as overall and relative survival in colon cancer. The secondary aims were to perform subgroup analyses of the prognostic value of TDs in different N stages and in patients with LNMs.
Methods
This study was approved by the Ethical Review Board of Lund University, Sweden (2020/01769), and followed the Declaration of Helsinki guidelines. The study was not preregistered.
Swedish colorectal cancer registry
Since 2007, all patients in Sweden with adenocarcinoma of the colon have been registered in the Swedish Colorectal Cancer Registry (SCRCR) nationwide quality registry. The SCRCR contains data on patients, tumours, diagnostic work-up, treatment characteristics, histopathological examinations and short-term outcomes registered 30 days after surgery19–21. Long-term outcomes, including late complications, recurrences and deaths, are reported 3 and 5 years after primary surgery19–21. In 2008, national treatment guidelines established follow-up routines for patients treated with surgery including computed tomography (CT) of the thorax and abdomen in conjunction with serum levels of carcinoembryonic antigen 12 and 36 months after surgery, as well as colonoscopy every 5 years until the age of 754. The date of death was obtained from the Cause of Death Registry. The SCRCR is a robust registry with a low proportion of missing data and high internal and external validity for key variables useful for quality assurance and research21. To date, 70 499 patients with colon cancer have been registered in the SCRCR19. In 2011, TDs were included in the SCRCR data set19.
Study population
The study was a retrospective analysis of prospectively registered data of patients with colon cancer registered in the SCRCR between 1 January 2011 and 31 December 2014. Included in the final analysis were patients with TNM stage I–III disease who underwent R0 abdominal resection surgery (hemicolectomies/colectomies) and were alive 90 days after surgery with a registered 5-year follow-up.
Definitions
During the studied interval, staging was reported according to the seventh edition of the TNM classification of malignant tumours2. In the absence of LNMs, TDs were placed in the nodal category designated N1c.
Colon cancer was defined by the SCRCR as an adenocarcinoma with the ICD-O-3 site codes C18.0–C18.9. Tumour location was dichotomized to right-sided if situated proximal to the splenic flexure and left-sided if situated in or distal to the splenic flexure.
R0 resection was defined as negative macroscopic margins after surgery, according to the surgeon, and negative microscopic margins in the surgical resection specimen, according to the pathologist.
A colorectal surgeon was defined as an accredited colorectal surgeon or a surgeon with a clinical subspecialization in colorectal surgery.
Emergency surgery was defined as a medically indicated procedure performed during an unplanned admission.
TDs were defined as macroscopic or microscopic nests or nodules, in the pericolic adipose tissue’s lymph drainage area of a primary carcinoma without histological evidence of residual lymph node in the nodule that may represent discontinuous spread, venous invasion with extravascular spread or a totally replaced lymph node2.
Histopathological tumour grading was dichotomized to low grade (well and moderately differentiated) and high grade (poorly differentiated and undifferentiated).
Lymphovascular invasion was defined as tumour infiltration of lymphatic or venous vessels, irrespective of size and intra- or extramural location.
Perineural invasion was defined as tumour infiltration of nerves in the tumour or tumour spread along the nerve pathway.
A mucinous tumour was defined as a tumour composed of more than 50 per cent extracellular mucin. This entity includes signet-ring cell carcinoma.
LR was defined as tumour recurrence at the anastomosis/resection line or in the peritoneum, mesenteric lymph nodes adjacent to the original primary tumour as well as abdominal incisions/port sites, as documented by clinical, radiological or pathological examination, or examination at surgery or autopsy more than 90 days after primary surgery.
DM was defined as tumour recurrence in the peritoneum or mesenteric lymph nodes remote from the original primary tumour as well as in any other organ (the liver, lung, ovary, bone, brain or any other parenchymatous organ) as documented by clinical, radiological or pathological examination, or examination at surgery or autopsy more than 90 days after primary surgery.
Overall survival was defined as the proportion of observed survivors in the studied cohort of colon cancer patients with a registered 5-year follow-up from the date of surgery.
Relative survival was defined as the ratio of the proportion of observed survivors in the studied cohort of colon cancer patients with a registered 5-year follow-up from the date of surgery to the proportion of expected survivors in a comparable cancer-free population.
Statistical analysis
The categorical data are presented as absolute numbers with percentages. For intergroup comparisons of categorical data, an χ2 test was used. Continuous data are presented as median (range) and Student's t test was used for intergroup comparison. Cumulative LR, DM and overall survival rates were calculated from the time of surgery to the end of follow-up. The Kaplan–Meier method was used to analyse overall survival and the log rank test to examine differences between groups. To assess the impact of TDs on the rates of LR, DM and 5-year overall survival, univariable and multivariable Cox regression analyses were performed using the following covariates: age, sex, adjuvant chemotherapy, tumour location, T stage, lymphovascular invasion, perineural invasion and tumour grading. Only cases with complete data were analysed in multivariable analyses. Hazard ratios (HRs) were calculated with 95 per cent confidence intervals. HRs larger than 1.00 indicated worse outcome for the test category versus the reference category. Relative survival was calculated using the Ederer II method to estimate expected survival differences between groups; subsequently, Poisson regression was employed in univariable and multivariable analyses22,23. Excess hazard ratio (EHR) was calculated with 95 per cent c.i. EHR larger than 1.00 indicated worse outcome for the test category versus the reference category.
For all tests, P < 0.050 was considered statistically significant. Data analyses were undertaken using SPSS® version 28.0 for Windows® (IBM, Armonk, NY, USA) and Stata®/MP version 16.1 (Stata Corp, College Station, TX, USA).
Results
During the studied interval, 16 401 patients were registered in the SCRCR and 13 267 (80.9 per cent) had abdominal resection surgery. After exclusion, 8014 patients remained (808 TD positive and 7206 TD negative) for analysis (Fig. 1). Thus, TDs were present among 10.1 per cent of the patients in the study cohort. The median TD count was 2 (range 1–40) in TD positive patients. Among stage III patients, 189 of 3033 (6.2 per cent) were staged as N1c (Table 1).
Fig. 1.
Study flow chart
TD, tumour deposits; LR, local recurrence; DM, distant metastasis.
Table 1.
Patient characteristics, treatment details and tumour data for patients who had elective R0 abdominal resection surgery for TNM stage I–III colon cancer in Sweden, 2011–2014
| Variables | All patients (n = 8014) |
TD positive (n = 808) |
TD negative (n = 7206) |
P* |
|---|---|---|---|---|
| Age, years† | 74 (19–98) | 72 (19–98) | 74 (29–97) | 0.017‡ |
| Sex | 0.371 | |||
| Male | 3927 (49.0) | 408 (50.5) | 3519 (48.8) | |
| Female | 4087 (51.0) | 400 (49.5) | 3687 (51.2) | |
| ASA score | 0.828 | |||
| I | 1042 (13.0) | 102 (12.6) | 940 (13.0) | |
| II | 4314 (53.8) | 439 (54.3) | 3875 (53.8) | |
| III | 2396 (29.9) | 234 (29.0) | 2162 (30.0) | |
| IV | 180 (2.2) | 21 (2.6) | 159 (2.2) | |
| Missing | 82 (1.0) | 12 (1.5) | 70 (1.0) | |
| BMI, kg/m2† | 25.4 (12.8–49.4) | 25.5 (14.7–46.5) | 25.3 (12.8–49.4) | 0.033‡ |
| Missing | 443 (5.5) | 50 (6.2) | 393 (5.5) | |
| Emergency surgery | <0.001 | |||
| No | 6989 (87.2) | 648 (80.2) | 6341 (88.0) | |
| Yes | 1025 (12.8) | 160 (19.8) | 865 (12.0) | |
| Surgical approach | ||||
| Open | 6561 (81.9) | 677 (83.8) | 5884 (81.7) | 0.087 |
| Laparoscopic | 1408 (17.6) | 124 (15.3) | 1284 (17.8) | |
| Missing | 45 (0.6) | 7 (0.9) | 38 (0.5) | |
| Surgical competence | 0.038 | |||
| Colorectal surgeon | 7457 (93.0) | 736 (91.1) | 6721 (93.3) | |
| General surgeon | 502 (6.3) | 64 (7.9) | 438 (6.1) | |
| Missing | 55 (0.7) | 8 (1.0) | 47 (0.7) | |
| Intraoperative perforation | 0.022 | |||
| No | 7831 (97.7) | 782 (96.8) | 7049 (97.8) | |
| Yes | 123 (1.5) | 20 (2.5) | 103 (1.4) | |
| Missing | 60 (0.7) | 6 (0.7) | 54 (0.7) | |
| Intraoperative bleeding, ml† | 100 (0–9500) | 150 (0–6000) | 100 (0–9500) | 0.020‡ |
| Missing | 270 (3.4) | 15 (1.9) | 255 (3.5) | |
| Adjuvant chemotherapy | <0.001 | |||
| No | 5723 (71.4) | 311 (38.5) | 5412 (75.1) | |
| Yes | 2240 (28.0) | 486 (60.1) | 1754 (24.3) | |
| Missing | 51 (0.6) | 11 (1.4) | 40 (0.6) | |
| Tumour location | <0.001 | |||
| Right-sided | 4650 (58.0) | 403 (49.9) | 4247 (58.9) | |
| Left-sided | 3358 (41.9) | 405 (50.1) | 2953 (41.0) | |
| Missing | 6 (0.1) | 0 (0.0) | 6 (0.1) | |
| TNM stage | <0.001 | |||
| I | 1504 (18.8) | 0 (0.0) | 1504 (20.9) | |
| II | 3477 (43.4) | 0 (0.0) | 3477 (48.3) | |
| III | 3033 (37.8) | 808 (100) | 2225 (30.9) | |
| T stage | <0.001 | |||
| 1 | 585 (7.3) | 11 (1.4) | 574 (8.0) | |
| 2 | 1194 (14.9) | 32 (4.0) | 1162 (16.1) | |
| 3 | 4857 (60.6) | 467 (57.8) | 4390 (60.9) | |
| 4 | 1372 (17.1) | 297 (36.8) | 1075 (14.9) | |
| Missing | 6 (0.1) | 1 (0.0) | 5 (0.1) | |
| No. of lymph nodes examined† | 20 (0–99) | 21 (4–99) | 20 (0–99) | <0.001‡ |
| Missing | 58 (0.7) | 3 (0.4) | 55 (0.8) | |
| N stage | <0.001 | |||
| 0 | 4959 (61.9) | 0 (0.0) | 4959 (68.8) | |
| 1 | 1989 (24.8) | 489 (60.5) | 1500 (20.8) | |
| 2 | 1042 (13.0) | 319 (39.5) | 723 (10.0) | |
| Missing | 24 (0.3) | 0 (0.0) | 24 (0.3) | |
| N1 stage subgroups | <0.001 | |||
| 1a | 863 (43.4) | 105 (21.5) | 758 (50.5) | |
| 1b | 880 (44.2) | 185 (37.8) | 695 (46.3) | |
| 1c | 189 (9.5) | 189 (38.7) | 0 (0.0) | |
| Missing | 57 (2.9) | 10 (2.0) | 47 (3.1) | |
| N2 stage subgroups | <0.001 | |||
| N2a | 554 (53.2) | 147 (46.1) | 407 (56.3) | |
| N2b | 454 (43.6) | 165 (51.7) | 289 (40.0) | |
| Missing | 34 (3.3) | 7 (2.2) | 27 (3.7) | |
| Tumour grading | <0.001 | |||
| Low grade | 6031 (75.3) | 547 (67.7) | 5484 (76.1) | |
| High grade | 1586 (19.8) | 217 (26.9) | 1369 (19.0) | |
| Missing | 397 (5.0) | 44 (5.4) | 353 (4.9) | |
| Lymphovascular invasion | <0.001 | |||
| No | 6033 (75.3) | 378 (46.8) | 5655 (78.5) | |
| Yes | 1901 (23.7) | 425 (52.6) | 1476 (20.5) | |
| Missing | 80 (1.0) | 5 (0.6) | 75 (1.0) | |
| Perineural invasion | <0.001 | |||
| No | 6719 (83.8) | 516 (63.9) | 6203 (86.1) | |
| Yes | 955 (11.9) | 273 (33.8) | 682 (9.5) | |
| Missing | 340 (4.2) | 19 (2.4) | 321 (4.4) | |
| Mucinous tumour | 0.878 | |||
| No | 6498 (81.1) | 649 (80.3) | 5849 (81.2) | |
| Yes | 1353 (16.9) | 137 (17.0) | 1216 (16.9) | |
| Missing | 163 (2.0) | 22 (2.7) | 141 (1.9) |
Values in parentheses are percentages unless indicated otherwise; *χ2 test, except ‡Student's t test. †values are median (range). TD, tumour deposit.
As shown in Table 1, patients with TD positive tumours had a higher T and N stage, tumour grading and incidence of lymphovascular and perineural invasion. The patients with TD positive tumours were younger, more frequently underwent elective surgery and were more likely to have left-sided tumours. No difference in the frequency of postoperative or surgical complications was detected between the two groups (Supplementary materials, Table S1). Patients with TD positive tumours were more extensively treated with adjuvant chemotherapy. Among N1c patients, 41.8 per cent had adjuvant chemotherapy, whereas the corresponding figures were 53.8 per cent for N1a, 60.8 for N1b, 65.5 for N2a and 67.4 for N2b patients (P < 0.001).
Recurrence
In the overall study population, LR was registered in 273 patients (3.4 per cent) within 5 years of primary surgery (Table 2). The LR rate was higher in TD positive patients (7.2 versus 3.0 per cent; P < 0.001). Metachronous DM was recorded in 1138 patients (14.2 per cent), with a higher frequency among TD positive patients (33.9 versus 12.0 per cent; P < 0.001).
Table 2.
Recurrence data for patients who had elective R0 abdominal resection surgery for TNM stage I–III colon cancer in Sweden, 2011–2014
| All patients (n = 8014) |
TD positive (n = 808) |
TD negative (n = 7206) |
P* | |
|---|---|---|---|---|
| Local recurrence | <0.001 | |||
| No | 7741 (96.6) | 750 (92.8) | 6991 (97.0) | |
| Yes | 273 (3.4) | 58 (7.2) | 215 (3.0) | |
| Distant metastasis | <0.001 | |||
| No | 6876 (85.8) | 534 (66.1) | 6342 (88.0) | |
| Yes | 1138 (14.2) | 274 (33.9) | 864 (12.0) |
Values in parentheses are percentages. *χ2 test. TD, tumour deposit.
In multivariable Cox regression analysis, adjusted for the covariables: age, sex, adjuvant chemotherapy, tumour location, T stage, lymphovascular invasion, perineural invasion and tumour grading, a TD positive tumour was an independent risk factor for LR (HR 1.50, 95 per cent c.i. 1.09 to 2.07; P = 0.013) and DM (HR 1.91, 95 per cent c.i. 1.64 to 2.23; P < 0.001) (Table 3).
Table 3.
Univariable and multivariable analysis relating the impact of tumour deposits, N stages and lymph node metastases on recurrence after elective R0 abdominal resection surgery for TNM stage I–III colon cancer in Sweden, 2011–2014
| Local recurrence | Distant metastasis | |||||
|---|---|---|---|---|---|---|
| n | Hazard ratio | P | Hazard ratio | P | ||
| TD status | Univariable | |||||
| Negative | 7206 | 1.00 | 1.00 | |||
| Positive | 808 | 2.69 (2.01–3.60) | <0.001 | 3.38 (2.95–3.88) | <0.001 | |
| Multivariable* | ||||||
| Negative | 6508 | 1.00 | 1.00 | |||
| Positive | 732 | 1.50 (1.09–2.07) | 0.013 | 1.91 (1.64–2.23) | <0.001 | |
| N stage | Univariable | |||||
| N0 | 4959 | 1.00 | 1.00 | |||
| N1a | 863 | 2.05 (1.37–3.07) | <0.001 | 2.81 (2.32–3.40) | <0.001 | |
| N1b | 880 | 3.14 (2.21–4.44) | <0.001 | 3.63 (3.04–4.32) | <0.001 | |
| N1c | 189 | 2.35 (1.14–4.83) | 0.021 | 2.27 (1.54–3.34) | <0.001 | |
| N2a | 554 | 3.82 (2.59–5.63) | <0.001 | 5.97 (4.99–7.15) | <0.001 | |
| N2b | 454 | 7.25 (5.11–10.27) | <0.001 | 9.02 (7.57–10.75) | <0.001 | |
| Multivariable* | ||||||
| N0 | 4488 | 1.00 | 1.00 | |||
| N1a | 791 | 1.46 (0.94–2.28) | 0.091 | 2.27 (1.83–2.81) | <0.001 | |
| N1b | 803 | 2.08 (1.39–3.12) | <0.001 | 2.66 (2.16–3.28) | <0.001 | |
| N1c | 172 | 1.64 (0.78–3.43) | 0.189 | 1.71 (1.14–2.56) | 0.010 | |
| N2a | 494 | 2.47 (1.57–3.88) | <0.001 | 4.13 (3.32–5.14) | <0.001 | |
| N2b | 397 | 3.88 (2.51–6.00) | <0.001 | 6.11 (4.90–7.63) | <0.001 | |
| LN status | TD status | Univariable | ||||
| N1a | Negative | 758 | 1.00 | 1.00 | ||
| Positive | 105 | 1.36 (0.52–3.54) | 0.526 | 1.96 (1.32–2.90) | 0.001 | |
| N1b | Negative | 695 | 1.00 | 1.00 | ||
| Positive | 185 | 1.17 (0.60–2.30) | 0.641 | 2.14 (1.59–2.88) | <0.001 | |
| N2a | Negative | 407 | 1.00 | 1.00 | ||
| Positive | 147 | 1.56 (0.78–3.14) | 0.211 | 1.47 (1.08–2.00) | 0.015 | |
| N2b | Negative | 289 | 1.00 | 1.00 | ||
| Positive | 165 | 1.39 (0.78–2.46) | 0.265 | 1.43 (1.08–1.90) | 0.012 | |
| Multivariable* | ||||||
| N1a | Negative | 692 | 1.00 | 1.00 | ||
| Positive | 99 | 1.46 (0.54–3.99) | 0.455 | 1.74 (1.13–2.66) | 0.011 | |
| N1b | Negative | 631 | 1.00 | 1.00 | ||
| Positive | 172 | 1.03 (0.50–2.15) | 0.932 | 1.95 (1.40–2.71) | <0.001 | |
| N2a | Negative | 360 | 1.00 | 1.00 | ||
| Positive | 134 | 1.14 (0.55–2.39) | 0.724 | 1.40 (1.00–1.96) | 0.053 | |
| N2b | Negative | 254 | 1.00 | 1.00 | ||
| Positive | 143 | 1.23 (0.66–2.29) | 0.520 | 1.21 (0.89–1.64) | 0.222 | |
Values in parentheses are 95 per cent c.i. *Adjusted for age, sex, adjuvant chemotherapy, tumour location, T stage, lymphovascular invasion, perineural invasion and tumour grading. LN, lymph node; TD, tumour deposit.
The HR for DM was higher for patients with N1c stage than N0 (HR 1.71, 95 per cent c.i. 1.14 to 2.56; P = 0.010), but not for LR (Table 3).
In a subgroup analysis of patients with LNMs, the HR for DM was higher in N1a, N1b and N2a patients with TD positive tumours than with TD negative tumours, but not for LR (Table 3). In the N2b stage, no difference was found in HR between TD positive and TD negative tumours for LR or DM.
Survival
The 5-year overall survival rate for patients with TD positive tumours was 56.8 per cent, compared with 74.9 per cent among TD negative tumours (P < 0.001) (Fig. 2a). Five-year relative survival rates were 68.5 and 92.6 per cent respectively (P < 0.001) (Fig. 2b).
Fig. 2.
Five-year survival of patients who had elective R0 abdominal resection surgery for TNM stage I–III colon cancer in Sweden, 2011–2014
a Overall and b relative survival. a P < 0.001 (log rank test), b P < 0.001 (the Ederer II method). TD, tumour deposit.
In multivariable Cox regression analysis, adjusted for the covariables: age, sex, adjuvant chemotherapy, tumour location, T stage, lymphovascular invasion, perineural invasion and tumour grading, a TD positive tumour was an independent risk factor for decreased overall (HR 1.60, 95 per cent c.i. 1.40 to 1.82; P < 0.001) and relative survival (EHR 2.24, 95 per cent c.i. 1.81 to 2.78; P < 0.001) respectively (Table 4).
Table 4.
Univariable and multivariable analysis relating the impact of tumour deposits, N stages and lymph node metastases on survival after elective R0 abdominal resection surgery for TNM stage I–III colon cancer in Sweden, 2011–2014
| Overall survival | Relative survival | |||||
|---|---|---|---|---|---|---|
| n | Hazard ratio | P | Excess hazard ratio | P | ||
| TD status | Univariable | |||||
| Negative | 7206 | 1.00 | 1.00 | |||
| Positive | 808 | 2.02 (1.80–2.27) | <0.001 | 4.61 (3.74–5.69) | <0.001 | |
| Multivariable* | ||||||
| Negative | 6508 | 1.00 | 1.00 | |||
| Positive | 732 | 1.60 (1.40–1.82) | <0.001 | 2.24 (1.81–2.78) | <0.001 | |
| N stage | Univariable | |||||
| N0 | 4959 | 1.00 | 1.00 | |||
| N1a | 863 | 1.43 (1.25–1.64) | <0.001 | 3.16 (2.09–4.76) | <0.001 | |
| N1b | 880 | 1.65 (1.45–1.88) | <0.001 | 4.12 (2.83–6.01) | <0.001 | |
| N1c | 189 | 1.31 (0.99–1.74) | 0.058 | 2.68 (1.23–5.85) | 0.014 | |
| N2a | 554 | 2.08 (1.80–2.41) | <0.001 | 7.36 (5.13–10.55) | <0.001 | |
| N2b | 454 | 3.67 (3.20–4.21) | <0.001 | 17.41 (12.61–24.02) | <0.001 | |
| Multivariable* | ||||||
| N0 | 4488 | 1.00 | 1.00 | |||
| N1a | 791 | 1.45 (1.25–1.68) | <0.001 | 2.80 (1.92–4.10) | <0.001 | |
| N1b | 803 | 1.66 (1.43–1.93) | <0.001 | 3.58 (2.50–5.13) | <0.001 | |
| N1c | 172 | 1.20 (0.89–1.61) | 0.223 | 2.11 (1.08–4.14) | 0.030 | |
| N2a | 494 | 2.19 (1.85–2.61) | <0.001 | 5.13 (3.57–7.37) | <0.001 | |
| N2b | 397 | 3.67 (3.10–4.34) | <0.001 | 9.55 (6.76–13.48) | <0.001 | |
| LN status | TD status | Univariable | ||||
| N1a | Negative | 758 | 1.00 | 1.00 | ||
| Positive | 105 | 1.39 (0.99–1.95) | 0.061 | 3.11 (1.67–5.77) | <0.001 | |
| N1b | Negative | 695 | 1.00 | 1.00 | ||
| Positive | 185 | 1.58 (1.22–2.05) | 0.001 | 2.59 (1.58–4.27) | <0.001 | |
| N2a | Negative | 407 | 1.00 | 1.00 | ||
| Positive | 147 | 1.45 (1.08–1.93) | 0.012 | 2.10 (1.34–3.31) | 0.001 | |
| N2b | Negative | 289 | 1.00 | 1.00 | ||
| Positive | 165 | 1.43 (1.11–1.83) | 0.005 | 1.49 (1.08–2.06) | 0.016 | |
| Multivariable* | ||||||
| N1a | Negative | 692 | 1.00 | 1.00 | ||
| Positive | 99 | 1.50 (1.04–2.15) | 0.029 | 2.08 (1.09–3.96) | 0.026 | |
| N1b | Negative | 631 | 1.00 | 1.00 | ||
| Positive | 172 | 1.62 (1.22–2.16) | 0.001 | 1.98 (1.22–3.20) | 0.005 | |
| N2a | Negative | 360 | 1.00 | 1.00 | ||
| Positive | 134 | 1.37 (0.99–1.88) | 0.055 | 1.70 (1.08–2.69) | 0.023 | |
| N2b | Negative | 254 | 1.00 | 1.00 | ||
| Positive | 143 | 1.20 (0.91–1.57) | 0.191 | 1.21 (0.87–1.68) | 0.252 | |
Values in parentheses are 95 per cent c.i. *Adjusted for age, sex, adjuvant chemotherapy, tumour location, T stage, lymphovascular invasion, perineural invasion and tumour grading. TD, tumour deposit; LN, lymph node.
The relative survival was decreased in patients with N1c stage compared with N0 patients (EHR 2.11, 95 per cent c.i. 1.08 to 4.14; P = 0.014) (Table 4); however, overall survival did not differ between the two groups.
Overall, HR was reduced in a subgroup analysis of patients with LNMs. EHR for decreased relative survival was higher in N1a, N1b and N2a patients with TD positive tumours compared with TD negative (Table 4). In the N2b stage, no difference in HR or EHR was found between patients with TD positive versus negative tumours for decreased overall and relative survival.
Discussion
In this study, TDs were an independent prognostic risk factor for LR and DM, reducing overall and relative survival rates in colon cancer. The negative prognostic impact remained in a subgroup analysis of N stages and patients with LNMs.
Data on the impact of TDs on recurrence in colon cancer are limited12. In a meta-analysis of colorectal cancer patients and two studies of stage III colon cancer patients, reduced disease-free survival (DFS) among patients with TD positive tumours was reported10,15,18. In the actual N stage subset analysis, N1c stage had worse prognosis regarding DM and relative survival than N0. In subgroup analysis of the patients with LNMs, the negative impact of TDs on prognosis remained for DM in N1a-b and N2a, but not in N2b patients. In prior studies, reduced DFS has been reported across all N stages for TD positive tumours without subdivision of N stages 1 and 2 into a and b15,18. The lack of impact of TDs on relative survival among N2b patients in the present analysis might be due to a type II error, as the number of patients in this group of tumours with advanced N stage was small. Alternatively, the prognostic impact of TDs might be overridden with more advanced N stage. In rectal cancer, TDs are proposed to relate to venous tumour spread as part of the vascular highway to DM, in contrast to local spread by LNMs, perineural and lymphatic invasion10,15,24. If this suggested pathway is true for colon cancer, this might explain the observed lack of impact on LR in patients with LNMs in the present analysis.
In the current cohort, patients with TD positive tumours had worse overall and relative survival rates. In the N stage subset analysis, relative survival, but not overall survival, was reduced in the N1c stage. Among the patients with LNMs, the impact of TDs on overall survival was not detected among N2a and b patients, nor on relative survival in N2b patients, possibly because of the low number of patients or more advanced N stage. In all other N stages, TDs reduced overall and relative survival. These findings are consistent with previous studies10,12–18. In American database studies, the combination of TDs and LNMs has the worst prognosis concerning overall survival13,14,16,17; however, detailed N stage subset data are not presented. In a small study, TDs negatively impacted overall survival in N1 and N2 stage patients, but the stages were not further subdivided18. Others demonstrated reduced disease-specific survival (DSS) in the presence of TDs in colorectal cancer10. Otherwise, data on the impact of TDs in colon cancer on DSS, cancer-specific survival and relative survival are seldom provided.
The prevalence of TDs in the studied cohort, including stage I–III patients, was as expected, lower than in other studies selectively including stage III patients, as TDs are associated with more advanced tumours10,12–18. Furthermore, the study strictly excluded non-radically operated patients, stage IV, recurrences, and deaths within 90 days of surgery to ensure that the analysed recurrences were true recurrences and not tumour progression.
In the study, TDs were more common among younger patients and left-sided tumours. Another author reported the same concerning tumour location15, but the higher prevalence among young patients was a novel finding15,17,18.
In agreement with earlier reports, TDs in the present study were related to other histopathologic risk factors for poor prognosis such as higher T and N stage, high tumour grading, lymphovascular, as well as perineural invasion10,12–18. An association with extramural venous invasion has also been demonstrated10,12,24, but unfortunately, this variable was not included in the SCRCR until 2017.
Although guidelines4–6 recommend adjuvant chemotherapy for stage III colon cancer patients, N1c stage patients receive less chemotherapy than others within stage III13,14. A recent audit of N1c stage patients in the National Cancer Database proved the underutilization of adjuvant chemotherapy and improved overall survival among treated patients in the N1c group25. The duration and regimen of adjuvant chemotherapy in N1c patients need further exploration13,14,25. The underuse of adjuvant chemotherapy in the current study might be explained by the close introduction of the seventh edition of the TNM staging system and the fact that its application in local MDTs was not yet established. However, in an American study14, including patients resected between 2010 and 2014, the difference within the stage III group remained after stratification by year, a disadvantage for N1c stage patients.
The independent prognostic impact of TDs has not been addressed in studies on neoadjuvant chemotherapy in colon cancer7–9. However, preoperative TNM staging of colon cancer is based on multidetector CT, which has limited accuracy for the T and N stages26,27.
The present data are prospectively collected and from a robust, national, population-based registry19–21 reflecting routine clinical care. The risk of selection bias is eliminated, and the number of events makes multivariable analysis with adjustment for an acceptable, although limited, number of confounders possible. Limitations include possible quality variations in pathology assessment and a relatively high frequency of missing data since the study interval started immediately after the inclusion of TDs in the SCRCR data set. Moreover, staging in the SCRCR today is according to the eighth edition of the TNM staging system3. In contrast, during the study interval, the staging was according to the seventh2. Some patients with N1c tumours in the actual study might today be downstaged to stage II N0, since the definition of TDs in the eighth edition has been changed from ‘no evidence of residual lymph node in the nodule’ to also include ‘or identifiable vascular or neural structures’2,3,11,12.
Because the prognostic information of TDs is lost in the current TNM staging system, modifications are suggested10–18,24. Various scenarios with different benefits and limitations are debated. However, most suggested modifications recommend that existing TDs should not be ignored in the presence of LNMs, and the number of TDs should be considered.
The present study proves an independent prognostic value of TDs across N stages and irrespective of the presence of LNMs partially ignored by the current TNM staging system3. The findings might support modifications of the TNM staging system to prevent the loss of prognostic information. Recognition of the negative prognostic impact of TDs in colon cancer when discussing adjuvant chemotherapy at postoperative MDTs is stressed. Future topics of relevance concerning TDs in colon cancer include associations to other patient- and tumour-related negative prognostic factors, preoperative radiologic staging, standardization of pathology assessment in routine clinical care, awareness of the negative prognostic impact, and the role of neoadjuvant and adjuvant chemotherapy.
Supplementary Material
Acknowledgements
The authors thank the board of the SCRCR for providing data and biostatistician A. Åkesson (Clinical Studies Sweden—Forum South, Skåne University Hospital) for assistance with the statistical analyses.
Contributor Information
Fredrik Jörgren, Department of Surgery, Helsingborg Hospital, Lund University, Helsingborg, Sweden.
Erik Agger, Department of Surgery, Skåne University Hospital, Malmö, Lund University, Malmö, Sweden.
Marie-Louise Lydrup, Department of Surgery, Skåne University Hospital, Malmö, Lund University, Malmö, Sweden.
Pamela Buchwald, Department of Surgery, Skåne University Hospital, Malmö, Lund University, Malmö, Sweden.
Funding
This work was supported by a grant from the Foundation of Stig and Ragna Gorthon (F.J.), Helsingborg, Sweden.
Disclosure
The authors declare no conflict of interest.
Supplementary material
Supplementary material is available at BJS Open online.
Data availability
Data can be provided upon reasonable request.
Author contributions
Fredrik Jörgren (Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Supervision, Writing—review & editing), Erik Agger (Conceptualization, Investigation, Methodology, Supervision, Writing—review & editing), Marie-Louise Lydrup (Conceptualization, Investigation, Methodology, Supervision, Writing—review & editing) and Pamela Buchwald (Conceptualization, Investigation, Methodology, Supervision, Writing—review & editing).
References
- 1. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2021;71:209–249 [DOI] [PubMed] [Google Scholar]
- 2. Sobin LH, Gospodarowicz MK, Wittekind C. TNM Classification of Malignant Tumours (7th edn). Chichester: Wiley-Blackwell, 2009 [Google Scholar]
- 3. Brierley J, Gospodarowicz MK, Wittekind C. TNM Classification of Malignant Tumours (8th edn). Chichester: Wiley-Blackwell, 2017 [Google Scholar]
- 4. Regional Cancer Centres of Sweden . The National Swedish Guidelines of Colorectal Cancer Care; 2022. https://www.cancercentrum.se/samverkan/cancerdiagnoser/tjocktarm-andtarm-och-anal/tjock--och-andtarm/vardprogram/ (accessed 21 August 2023)
- 5. ESMO Clinical Practice Guidelines . Gastrointestinal Cancers. https://www.esmo.org/Guidelines/Gastrointestinal-Cancers (accessed 21 August 2023)
- 6. American Society of Clinical Oncology (ASCO) . Practice & Guidelines—Gastrointestinal Cancer. https://www.asco.org/research-guidelines/quality-guidelines/guidelines/gastrointestinal-cancer (accessed 21 August 2023)
- 7. Karoui M, Gallois C, Piessen G, Legoux JL, Barbier E, De Chaisemartin C et al. Does neoadjuvant FOLFOX chemotherapy improve the prognosis of high-risk stage II and III colon cancers? Three years’ follow-up results of the PRODIGE 22 phase II randomised multicentre trial. Colorectal Dis 2021;23:1357–1369 [DOI] [PubMed] [Google Scholar]
- 8. Jung F, Lee M, Doshi S, Zhao G, Cheung KLT, Chesney T et al. Neoadjuvant therapy versus direct to surgery for T4 colon cancer: meta-analysis. Br J Surg 2021;109:30–36 [DOI] [PubMed] [Google Scholar]
- 9. Morton D, Seymour M, Magill L, Handley K, Glasbey J, Glimelius B et al. Preoperative chemotherapy for operable colon cancer: mature results of an international randomized controlled trial. J Clin Oncol 2023;41:1541–1552 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Nagtegaal ID, Knijn N, Hugen N, Marshall HC, Sugihara K, Tot T et al. Tumor deposits in colorectal cancer: improving the value of modern staging—a systematic review and meta-analysis. J Clin Oncol 2017;35:1119–1127 [DOI] [PubMed] [Google Scholar]
- 11. Lord A, Brown G, Abulafi M, Bateman A, Frankel W, Goldin R et al. Histopathological diagnosis of tumour deposits in colorectal cancer: a Delphi consensus study. Histopathology 2021;79:168–175 [DOI] [PubMed] [Google Scholar]
- 12. Delattre JF, Selcen Oguz Erdogan A, Cohen R, Shi Q, Emile JF, Taieb J et al. A comprehensive overview of tumour deposits in colorectal cancer: towards a next TNM classification. Cancer Treat Rev 2022;103:102325. [DOI] [PubMed] [Google Scholar]
- 13. Mirkin KA, Kulaylat AS, Hollenbeak CS, Messaris E. Prognostic significance of tumor deposits in stage III colon cancer. Ann Surg Oncol 2018;25:3179–3184 [DOI] [PubMed] [Google Scholar]
- 14. Wong-Chong N, Motl J, Hwang G, Nassif GJ, Albert MR, Monson JRT et al. Impact of tumor deposits on oncologic outcomes in stage III colon cancer. Dis Colon Rectum 2018;61:1043–1052 [DOI] [PubMed] [Google Scholar]
- 15. Delattre JF, Cohen R, Henriques J, Falcoz A, Emile JF, Fratte S et al. Prognostic value of tumor deposits for disease-free survival in patients with stage III colon cancer: a post hoc analysis of the IDEA France phase III trial (PRODIGE-GERCOR). J Clin Oncol 2020;38:1702–1710 [DOI] [PubMed] [Google Scholar]
- 16. Pricolo VE, Steingrimsson J, McDuffie TJ, McHale JM, McMillen B, Shparber M. Tumor deposits in stage III colon cancer. Am J Clin Oncol 2020;43:133–138 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Zheng P, Lai C, Yang W, Chen Z. Prognostic significance of tumor deposits in combination with lymph node metastasis in stage III colon cancer: a propensity score matching study. Am Surg 2020;86:164–170 [PubMed] [Google Scholar]
- 18. Cohen R, Shi Q, Meyers JP, Jin Z, Svrcek M, Fuchs CS et al. Combining tumor deposits with the number of lymph node metastases to improve the prognostic accuracy in stage III colon cancer: a post hoc analysis of the CALGB/SWOG 80702 phase III study (Alliance). Ann Oncol 2021;32:1267–1275 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Regional Cancer Centres of Sweden . Annual Report from the Swedish Colorectal Cancer Registry; 2021. https://www.cancercentrum.se/samverkan/cancerdiagnoser/tjocktarm-andtarm-och-anal/tjock--och-andtarm/kvalitetsregister/ (accessed August 21 2023)
- 20. Kodeda K, Nathanaelsson L, Jung B, Olsson H, Jestin P, Sjövall A et al. Population-based data from the Swedish colon cancer registry. Br J Surg 2013;100:1100–1107 [DOI] [PubMed] [Google Scholar]
- 21. Moberger P, Sköldberg F, Birgisson H. Evaluation of the Swedish colorectal cancer registry: an overview of completeness, timeliness, comparability and validity. Acta Oncol 2018;57:1611–1621 [DOI] [PubMed] [Google Scholar]
- 22. The Human Mortality Database . https://www.mortality.org/ (accessed 20 April 2022)
- 23. Dickman PW, Coviello E. Estimating and modeling relative survival. Stata J 2015;15:186–215 [Google Scholar]
- 24. Lord AC, Knijn N, Brown G, Nagtegaal ID. Pathways of spread in rectal cancer: a reappraisal of the true routes to distant metastatic disease. Eur J Cancer 2020;128:1–6 [DOI] [PubMed] [Google Scholar]
- 25. Simon HL, Reif de Paula T, Spigel ZA, Keller DS. N1c colon cancer and the use of adjuvant chemotherapy: a current audit of the national cancer database. Colorectal Dis 2021;23:653–663 [DOI] [PubMed] [Google Scholar]
- 26. Korsbakke K, Dahlbäck C, Karlsson N, Zackrisson S, Buchwald P. Tumor and nodal staging of colon cancer: accuracy of preoperative computed tomography at a Swedish high-volume center. Acta Radiol Open 2019;8:2058460119888713 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Friis Olsen AS, Kierkegaard Grundestrup A, Kleif J, Thanon T, Bertelsen CA. Accuracy of preoperative staging with multidetector computed tomography in colon cancer. Colorectal Dis 2021;23:680–688 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data can be provided upon reasonable request.