- Email: [email protected]
Disease Characteristics, Clinical Management, and Outcomes of Young Patients With Colon Cancer: A Population-based Study
Accepted Manuscript Disease Characteristics, Clinical Management, and Outcomes of Young Patients with Colon Cancer: A Population-Based Study Laura Rodriguez, Kelly Brennan, Safiya Karim, Sulaiman Nanji, Sunil V. Patel, Christopher M. Booth PII:
S1533-0028(18)30168-3
DOI:
10.1016/j.clcc.2018.06.007
Reference:
CLCC 480
To appear in:
Clinical Colorectal Cancer
Received Date: 13 April 2018 Revised Date:
18 June 2018
Accepted Date: 27 June 2018
Please cite this article as: Rodriguez L, Brennan K, Karim S, Nanji S, Patel SV, Booth CM, Disease Characteristics, Clinical Management, and Outcomes of Young Patients with Colon Cancer: A Population-Based Study, Clinical Colorectal Cancer (2018), doi: 10.1016/j.clcc.2018.06.007. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
RI PT
ACCEPTED MANUSCRIPT
Disease Characteristics, Clinical Management, and Outcomes of Young Patients with Colon Cancer: A Population-Based Study
SC
Laura Rodriguez1, Kelly Brennan2, Safiya Karim2,3,
1
M AN U
Sulaiman Nanji3,4, Sunil V. Patel3,4, Christopher M. Booth2,3,5
Canadian Cancer Trials Group
2
Division of Cancer Care and Epidemiology, Queen’s University Cancer Research Institute, Departments of Oncology3, Surgery4, Public Health Sciences5
TE D
Queen’s University, Kingston, Canada
AC C
EP
Corresponding author: Dr. Christopher Booth Division of Cancer Care and Epidemiology Queen’s University Cancer Research Institute 10 Stuart Street Kingston, ON, K7L 3N6, Canada Telephone: (613) 533-6895 Fax: (613) 533-6794 Email: [email protected]
Running title: Young patients with colon cancer Word Count: 2615 June 14, 2018
1
ACCEPTED MANUSCRIPT
ABSTRACT Introduction: The incidence of colorectal cancer in young patients is increasing. Here we evaluate whether disease characteristics, management, and outcomes of patients with colon
RI PT
cancer differs among patients <40 years compared with older patients.
Methods: Using the Ontario Cancer Registry, all cases of colon cancer (stage I, II, III) treated with surgery in Ontario during 2002–2008 were identified. Electronic records of treatment
SC
identified use of surgery and adjuvant chemotherapy (ACT). Pathology reports were obtained for a random 25% sample of all cases. A Cox model was used to identify factors associated with
M AN U
overall (OS) and cancer-specific survival (CSS).
Results: The study population included 6775 patients; age distribution was 2%, 5%, 14%, 79% for ≤40, 41-50, 51-60, >60 years of age. Compared to patients >60 years, younger patients (<40) were more likely to have lymphovascular invasion (35% vs 27%, p=0.005), T3/4 tumours
TE D
(88% vs 79%, p=0.005) and node positive disease (58% vs 41%, p<0.001). Stage distribution varied by age: stage I, 8% (<40 years) vs 19% (>60 years); stage II, 34% vs 40%; stage III, 58% vs 41% (p<0.001). ACT was delivered more commonly for patients <40 years compared to >60
EP
years for stage II (50% vs 13%, p<0.001) and stage III (≥92% vs 57%, p<0.001) disease. Adjusted OS (HR 0.32, 95%CI 0.21-0.49) and CSS (HR 0.41, 95%CI 0.26-0.64) were superior
AC C
for patients <40 compared to those >60 years of age. Conclusions: Young patients with colon cancer have more aggressive and advanced disease but improved outcomes compared to older patients.
2
ACCEPTED MANUSCRIPT
INTRODUCTION Colorectal cancer (CRC) is the second leading cause of death from cancer in men and the third leading cause of death from cancer in women in Canada (1). Recent declines in the
RI PT
incidence and mortality rates of CRC have been attributed to adoption of screening which is recommended for adults >50 years of age (2, 3). However, emerging data suggests that the
incidence of CRC in young adults (i.e. <40 years) is increasing (4, 5). There are conflicting
SC
reports in the literature as to whether disease biology varies by age. While some studies have found that younger patients present with more aggressive disease and advanced stage compared
M AN U
with older patients (6-8); other studies contradict these findings (9, 10). However, much of this literature comes from single-centre and/or pooled institutional case series. There is also uncertainty in the literature about whether the outcome of CRC differs by age. Some investigators have reported that younger patients have inferior outcomes (11-13), while others
TE D
have reported superior survival among younger patients (6, 14, 15).
The objective of this study was to determine whether disease characteristics, management and outcomes were different in young patients (≤40) compared with older age patients (> 60)
AC C
EP
within a Canadian population.
3
ACCEPTED MANUSCRIPT
METHODS Study Design and Population This is a population-based, retrospective cohort study of patients treated for colon cancer
RI PT
in the Canadian province of Ontario. Ontario has a population of approximately 13.5 million people and a single-payer universal health insurance program. The study population included a 25% random sample of all patients who underwent resection of primary colon cancer in Ontario
SC
between 2002 and 2008. To identify the study cohort, we used the Ontario Cancer Registry
(OCR) to identify all incident cases of colon cancer in Ontario diagnosed during 2000-2008. The
M AN U
OCR does not capture stage of disease; therefore, we obtained surgical pathology reports for a random sample of 25% of cases. The substantial workload burden of manual record review precluded us from including 100% of cases in Ontario. Cases were randomly selected using a computer generated random sample algorithm. Reports were not available for patients with
TE D
surgery in 2005; as such the study cohort is restricted to patients who had surgery in 2002-2004 and 2006-2008. Because we relied on review of surgical pathology reports to assign stage of disease, the study excluded patients with rectal cancer as neoadjuvant therapy would
EP
undoubtedly influence the pathologic findings at surgery. The study was approved by the Research Ethics Board of Queen’s University, Kingston, Canada and the institutional review
AC C
board at Sunnybrook Health Sciences Centre, Toronto, Canada. This study was designed, analyzed, and reported in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) statement (16, 17). Data Sources and Linkage
The OCR is a passive, population-based cancer registry that captures diagnostic and demographic information on at least 98% of all incident cases of cancer in the province of
4
ACCEPTED MANUSCRIPT
Ontario (18). The OCR also provides information about vital status and cause of death. Records of hospitalization from the Canadian Institute for Health Information (CIHI) provided information about surgical procedures; these records are known to have a very high level of
RI PT
completeness for colorectal cancer surgery (19). Provincial physician billing records from the Ontario Health Insurance Plan, treatment records from regional cancer centers, and provincial records of chemotherapy delivery were used to identify chemotherapy utilization. A team of
SC
trained data abstractors reviewed the pathology reports and entered information about extent of disease into an electronic database. Patient with evidence of metastatic disease based on the
M AN U
pathology report from the primary tumour resection (i.e. liver/omentum/peritoneum, ovary) were classified as stage IV and excluded from the current study. These datasets were linked using unique encoded identifiers and analyzed at the Institute for Clinical Evaluative Sciences (ICES). Exposures and Outcomes
TE D
Patient age was classified at the time of surgery as < 40, 41-50, 51-60, >60 years. The age groups were created a priori; we selected 40 as the upper limit of the younger group as we were most interested in those patients in their 20s and 30s. Indicators of the socioeconomic status
EP
(SES) of the community in which patients resided at diagnosis were linked as described previously (20). Quintiles (Q) of the median household income were based on the household
AC C
income distribution for the full province of Ontario. Q1 represents the communities where the poorest 20% of the Ontario population resided. Co-morbidity was classified using the Charlson Index modified for administrative data based on all non-cancer diagnoses recorded during any hospital admission within 5 years prior to surgery (21). Adjuvant chemotherapy (ACT) was defined as chemotherapy initiated within 16 weeks after surgery. High-risk stage II colon cancer was defined as having at least one of the following:
5
ACCEPTED MANUSCRIPT
T4 tumour, poorly differentiated histology, < 12 lymph nodes, lymphovascular invasion. Study data-sets did not allow us to identify bowel perforation or obstruction. Disease laterality was defined from the pathology report based on right-sided (cecum, ascending colon, hepatic flexure,
RI PT
transverse colon) versus left-sided (splenic flexure, descending colon, sigmoid colon, rectosigmoid colon) resections.
Overall (OS) and cancer-specific survival (CSS) were determined from date of primary
SC
tumour resection. To account for possible cause of death miscoding, CSS included death from any cancer. Complete information about vital status in the OCR was available up to December
M AN U
31, 2014; cause of death was available up to December 31, 2012. Statistical Analysis
The primary objective was to compare disease characteristics, management, and outcome of patients <40 years of age with those >60 years. Comparisons of proportions between study
TE D
groups were made using the chi-square test. Survival was determined from date of surgery using the Kaplan-Meier technique. The association between patient-, disease-, and treatment-related factors with overall/cancer-specific survival was evaluated using the Cox proportional hazards
EP
regression model. A priori we planned sub-group analyses to explore whether management and outcome of stage III colon cancer differed by age groups. To explore effect modification of the
AC C
association of age groups and survival by adjuvant chemotherapy we tested for interaction between ACT and age groups in the Cox proportional hazards regression model for patients with stage 3 disease. Results were considered statistically significant at p-value < 0.05. As per institutional policy, data that relate to <6 patients are not reported due to small size. All analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC).
6
ACCEPTED MANUSCRIPT
RESULTS Study population The linked administrative data-sets identified 25 613 potentially eligible patients who
RI PT
underwent resection of colon cancer in 2002-2008 (Supplemental eFigure 1). Surgical pathology reports were reviewed for 7519 randomly selected cases. The age, sex, comorbidity, and survival of the randomly selected cases did not differ significantly from those of the 18 094 cases not
SC
selected (Supplemental eTable and eFigure 2). Among the 7519 randomly selected cases, 270 (4%) were excluded as shown in Supplemental eFigure 1; cases with stage 0 (n=31) and resected
M AN U
stage IV (n=443) disease were excluded from the study population. Accordingly, the study population included 6775 patients with stage I, II, III colon cancer. Characteristics of the study population are shown in Table 1. Overall the median age was 72 years and 51% were male. Stage distribution among all cases was: 19% stage I, 38% stage II, and 43% stage III. Young
(p<0.001). Disease Characteristics
TE D
patients were more likely to be of higher SES, less co-morbidity, and more advanced disease
EP
Compared to patients >60 years of age, younger patients were more likely to have leftsided tumours (50% vs 44%, p<0.001), lymphovascular invasion (35% vs 27%, p=0.005), T3/4
AC C
tumours (88% vs 79%, p=0.005) and node positive disease (58% vs 41%, p<0.001). Stage distribution accordingly varied by age: stage I, 8% (<40 years) vs 19% (>60 years); stage II, 34% vs 40%; stage III, 58% vs 41% (p<0.001). Treatment delivery
Surgical care and delivery of chemotherapy is described in Table 2. Length of stay was shorter in young patients compared to those >60 years (median 7 vs 8 days, p<0.001). Lymph
7
ACCEPTED MANUSCRIPT
node harvest was substantially greater in patients ≤40 compared to >60 years (median 20 vs 14, p<0.001) Compared to patients >60 years of age, adjuvant chemotherapy was delivered more
RI PT
commonly for younger patients for stage II (50% vs 13%, p<0.001) and stage III (>92% vs 57%, p<0.001) disease. ACT treatment rates for high-risk stage II disease were 71% for younger
patients compared to 17% for patients >60 years (p<0.001). ACT treatment rates for stage II
SC
disease without high risk features were 36% for younger patients and 8% for those >60 years (p<0.001). Chemotherapy was initiated more promptly after surgery for younger patient than
M AN U
those >60 years (48% within 8 weeks of surgery vs 18%, p=0.002). Outcomes
Unadjusted five-year OS/CSS for the full study population was 80%/82% (<40 years), 71%/71% (41-50), 72%/75% (51-60), and 59%/68% (>60) (Figure 1).
TE D
Factors associated with OS and CSS in the full study population are shown in Table 3. Adjusted OS (HR 0.32, 95%CI 0.21-0.49) and CSS (HR 0.41, 95%CI 0.26-0.64) was superior among patients <40 years compared to those >60 years of age. Similar results were observed
EP
when stratified Cox model were generated for those patients with stage III disease (Table 4) [OS HR 0.43 (95%CI 0.27-0.68); CSS HR 0.49 (95%CI 0.31-0.78)] and stage II disease [OS HR 0.09
AC C
(95%CI 0.02-0.35); CSS HR 0.16 (95%CI 0.04-0.64)] (data not shown). Adjuvant chemotherapy was associated with improved survival in the overall cohort and the stage III sub-group. Among the stage III sub-group, a test for interaction between ACT and age suggested there was no differential effect of ACT on survival by age (OS p=0.831, CSS p=0.851).
8
ACCEPTED MANUSCRIPT
DISCUSSION In this study we have explored disease characteristics, management and outcome of early
RI PT
stage colon cancer among patients 40 years and younger. Our data demonstrate that younger patients have more aggressive histology and more advanced stage at diagnosis. They are more likely to receive adjuvant chemotherapy, have a shorter interval between surgery and initiating chemotherapy, and a greater lymph node harvest. Despite more aggressive and extensive disease,
SC
young patients have improved long-term overall and cancer-specific survival compared to older
M AN U
patients. Our data do not suggest that adjuvant chemotherapy has a differential effect across age groups.
Our results are consistent with several single center retrospective studies. Murata et al, reported a retrospective study of 2338 patients with diagnosis of CRC between 2004 and 2012, patients <40 years (n=81) old had a lower incidence of right-sided colon cancer and higher
TE D
number of retrieved lymph nodes, and were more likely to have undergone postoperative adjuvant treatment compared with the older group; however, survival was comparable to older patients (22). Another consecutive series of CRC patients from 1999-2007 (n=2220), reported
EP
young patients having more aggressive and advanced cancer stages (14). However, older patients
AC C
had higher postoperative mortality and were less likely to receive chemotherapy. OS was 69, 68, and 40% (p<0.001) among young (16-53 years), median (53-84 years) and old (85-97 years) group respectively; CSS was 70, 70, and 52% (p<0.003) respectively. One notable limitation of this study was the lack of adjusted survival analyses. The proportion of cancer-related deaths was lower among older patients. A study from the Memorial Sloan-Kettering Cancer Center, evaluated a prospective database of 1327 surgical stage I–III colon cancer patients (1990-2001) (15). Compared to older patients, those <40 years of age were more likely to have left-sided
9
ACCEPTED MANUSCRIPT
tumors and higher nodal yields, and were more likely to receive adjuvant chemotherapy. The 5year CSS was similar in both study groups: 86% vs 87%, but 5- year OS was significantly higher in the younger patient cohort (84% vs 73%, p=0.001).
RI PT
Our results are also consistent with some tumor characteristics, clinical presentation and outcomes, reported by previous population-based cohort studies from the Surveillance,
Epidemiology, and End Results (SEER) registry. O’Connell et al, reported that CRC patients
SC
<40 years old had more aggressive histology and more advanced stage than older patients.
Younger patients had superior survival for stage II and IV lesions but comparable survival for
M AN U
stage I and III tumors (6). Wang et al found that patients <50 years had later stage presentation, more aggressive pathological features, and higher lymph node yields (3). The 5-year CSS were 65%, 67%, and 63%, for patients <40, 41 to 50 and > 50 years old patients respectively. Despite having higher risk disease, in univariate and multivariate analyses younger patients had
TE D
improved cancer survival.
A consistent finding in our study and existing reports is the observation that younger patients present with more advanced stage disease. This may be explained by the fact that current
EP
screening guidelines do not recommend routine screening for this seemingly low-risk population
AC C
(23). This could result in younger patients being diagnosed once symptomatic and thus at a later stage. In addition, younger patients may be diagnosed at a later stage due to delays in seeking medical attention and subsequently referral to specialist care (24). Younger patients symptoms may be more likely to be attributed to benign causes (i.e. hemorrhoids) than older patients. The paradoxical observation that younger patients have superior survival may relate to less comorbidity, fewer complications related to surgery, improved tolerance of adjuvant chemotherapy, or intrinsic differences in tumour biology. The greater nodal harvest among
10
ACCEPTED MANUSCRIPT
young patients may also be a mitigating factor; however controversy exists as to whether greater lymph node yield is intrinsically therapeutic, whether it leads to appropriate upstaging, whether it serves as a marker of quality care, or whether it is as an epiphenomenon of underlying host-
RI PT
tumour interaction (25, 26). Our data also show higher use of adjuvant chemotherapy in younger patients for both stage III and stage II disease. Among the latter, ACT rates may be
inappropriately high for young patients with stage II disease who do not have any high-risk
SC
pathologic features.
There is emerging interest in the role of human papilloma virus (HPV) in colorectal
M AN U
cancer (27). While our data do not offer any specific insights into this issue, further work is needed to understand if HPV may explain the rising incidence of CRC in younger populations. This study has several limitations. The major limitation was its retrospective design, in which we cannot ascertain causation, as overall survival and treatment choice is subject to
TE D
considerable selection bias that cannot be entirely mitigated with risk adjustment. It is worth noting that one would expect fundamental differences in younger patients with CRC as they are more likely to present with symptomatic disease whereas older patients are more likely to be
EP
diagnosed through screening programs. Although we manually reviewed surgical pathology reports for all cases, we did not have access to other elements of the patient chart. This prevented
AC C
us from being able to provide any additional information regarding risk factors for colon cancer (i.e. family history, smoking status, diabetes). The existing data sources also lack information regarding performance status and patient preferences regarding treatment selection. We also were unable to explore the proportion of patients with Lynch Syndrome or microsatellite instability which one would expect to disproportionately affect younger patients. Finally, a significant limitation of this study is the small number of young patients included in the study
11
ACCEPTED MANUSCRIPT
(n= 107). Despite these limitations, the use of a population-based data provides a large sample size and reduces many of the selection and referral biases that plague single centre institutional case series.
RI PT
In summary, our data demonstrate that patients <40 years of age have more advanced disease and more aggressive biology than older patients. It is unclear if this is driven by
fundamental differences in disease biology or whether it simply reflects the fact that older
SC
patients are more likely to have CRC screening. Despite adverse prognostic features, younger patients have improved outcomes compared to older patients. Further work is needed to
AC C
EP
TE D
M AN U
understand if this apparent disconnect is explained by intrinsic differences in tumour biology.
12
ACCEPTED MANUSCRIPT
ACKNOWLEDGEMENTS Parts of this material are based on data and information provided by Cancer Care Ontario.
authors and not necessarily those of Cancer Care Ontario.
RI PT
However, the analysis, conclusions, opinions and statements expressed herein are those of the
Parts of this material are based on data and information compiled and provided by CIHI.
author, and not necessarily those of CIHI.
SC
However, the analyses, conclusions, opinions and statements expressed herein are those of the
M AN U
This study was supported by the Institute for Clinical Evaluative Sciences (ICES), which is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care (MOHLTC). The opinions, results and conclusions reported in this paper are those of the authors and are independent from the funding sources. No endorsement by ICES or the Ontario
TE D
MOHLTC is intended or should be inferred.
Dr. Booth had full access to all the data in the study and takes responsibility for the integrity of
FUNDING
EP
the data and the accuracy of the data analysis.
AC C
Dr. Booth is supported as a Canada Research Chair in Population Cancer Care. This work was also supported by the Canada Foundation for Innovation and the Canadian Institutes of Health Research.
DISCLOSURE
The authors report no conflicts of interest.
13
ACCEPTED MANUSCRIPT
REFERENCES
AC C
EP
TE D
M AN U
SC
RI PT
1. Canadian Cancer Society. Canadian Cancer Statistics 2017. Toronto: Canadian Cancer Society; 2017. Avilable at: http://www.cancer.ca/en/cancer-information/cancer-101/canadiancancer-statistics-publication/?region=on. Last accessed on November 10, 2017. 2. National Cancer Institute. SEER Cancer Stat Facts: Colon and Rectum Cancer. National Cancer Institute. Bethesda, MD. Avilabhle at: http://seer.cancer.gov/statfacts/html/colorect.html. Last accessed on November 10, 2017. 3. Wang R, Wang MJ, Ping J. Clinicopathological Features and Survival Outcomes of Colorectal Cancer in Young Versus Elderly: A Population-Based Cohort Study of SEER 9 Registries Data (1988-2011). Medicine (Baltimore). 2015;94(35):e1402. 4. Ahnen DJ, Wade SW, Jones WF, Sifri R, Mendoza Silveiras J, Greenamyer J, et al. The increasing incidence of young-onset colorectal cancer: a call to action. Mayo Clin Proc. 2014;89(2):216-24. 5. Bailey CE, Hu CY, You YN, Bednarski BK, Rodriguez-Bigas MA, Skibber JM, et al. Increasing disparities in the age-related incidences of colon and rectal cancers in the United States, 1975-2010. JAMA Surg. 2015;150(1):17-22. 6. O'Connell JB, Maggard MA, Liu JH, Etzioni DA, Livingston EH, Ko CY. Do young colon cancer patients have worse outcomes? World J Surg. 2004;28(6):558-62. 7. Lee PY, Fletcher WS, Sullivan ES, Vetto JT. Colorectal cancer in young patients: characteristics and outcome. Am Surg. 1994;60(8):607-12. 8. Leff DR, Chen A, Roberts D, Grant K, Western C, Windsor AC, et al. Colorectal cancer in the young patient. Am Surg. 2007;73(1):42-7. 9. Paraf F, Jothy S. Colorectal cancer before the age of 40: a case-control study. Dis Colon Rectum. 2000;43(9):1222-6. 10. Chung YF, Eu KW, Machin D, Ho JM, Nyam DC, Leong AF, et al. Young age is not a poor prognostic marker in colorectal cancer. Br J Surg. 1998;85(9):1255-9. 11. Palmer ML, Herrera L, Petrelli NJ. Colorectal adenocarcinoma in patients less than 40 years of age. Dis Colon Rectum. 1991;34(4):343-6. 12. Marble K, Banerjee S, Greenwald L. Colorectal carcinoma in young patients. J Surg Oncol. 1992;51(3):179-82. 13. Fu JF, Huang YQ, Yang J, Yi CH, Chen HL, Zheng S. Clinical characteristics and prognosis of young patients with colorectal cancer in Eastern China. World J Gastroenterol. 2013;19(44):8078-84. 14. Derwinger K, Kodeda K, Gerjy R. Age aspects of demography, pathology and survival assessment in colorectal cancer. Anticancer Res. 2010;30(12):5227-31. 15. Quah HM, Joseph R, Schrag D, Shia J, Guillem JG, Paty PB, et al. Young age influences treatment but not outcome of colon cancer. Ann Surg Oncol. 2007;14(10):2759-65. 16. Connell L, MacDonald R, McBride T, Peiperl L, Ross A, Simpson P, et al. Observational Studies: Getting Clear about Transparency. Plos Med. 2014;11(8). 17. Vandenbroucke JP, von Elm E, Altman DG, Gotzsche PC, Mulrow CD, Pocock SJ, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): Explanation and Elaboration. Ann Intern Med. 2007;147(8):W163-W94. 18. Clarke EA, Marrett LD, Kreiger N. Cancer registration in Ontario: a computer approach. IARC Sci Publ. 1991(95):246-57. 19. Li X, King C, deGara C, White J, Winget M. Validation of colorectal cancer surgery data from administrative data sources. BMC Med Res Methodol. 2012;12:97. 14
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
20. Mackillop WJ, Zhang-Salomons J, Groome PA, Paszat L, Holowaty E. Socioeconomic status and cancer survival in Ontario. J Clin Oncol. 1997;15(4):1680-9. 21. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol. 1992;45(6):613-9. 22. Murata A, Akiyoshi T, Ueno M, Fukunaga Y, Nagayama S, Fujimoto Y, et al. Clinicopathological characteristics of young patients with sporadic colorectal cancer. Surg Today. 2016;46(10):1166-75. 23. Canadian Task Force on Preventive Health Care. Recommendations on screening for colorectal cancer in primary care. Canadian Medical Assoc J. 2016; 188(5): 340-348. 24. Scott RB, Rangel LE, Osler TM, Hyman NH. Rectal cancer in patients under the age of 50 years: the delayed diagnosis. American journal of surgery. 2016;211(6):1014-8. 25. Del Paggio JC, Nanji S, Wei X, MacDonald PH, Booth CM. Lymph node evaluation for colon cancer in routine clinical practice: a population-based study. Curr Oncol. 2017;24(1):e35e43. 26. Del Paggio JC, Peng Y, Wei X, Nanji S, MacDonald PH, Krishnan Nair C, et al. Population-based study to re-evaluate optimal lymph node yield in colonic cancer. The British journal of surgery. 2017;104(8):1087-96. 27. Baandrup L, Thomsen LT, Olesen TB, Andersen KK, Norrild B, Kjaer SK. The prevalence of human papillomavirus in colorectal adenomas and adenocarcinomas: a systematic review and meta-analysis. Eur J Cancer. 2014;50(8):1446-61.
15
ACCEPTED MANUSCRIPT
Supplemental eFigure 1. Identification of patients with resected colon cancer in Ontario during 2002-2008. All colon cancer diagnosed 2000-2008 N=45383
SC
Colon cancer with primary surgery within 6 months of diagnosis N=37847
RI PT
No primary surgery within 6 months of diagnosis N=7536
M AN U
Surgery not in 2002-2004 or 2006-2008 N=12234
Colon cancer with primary surgery in 2002-2004 or 2006-2008 N=25613
TE D
Random Selection
EP
Pathology Report Reviewed N=7519
AC C
Colon cancer surgical cases N=7249
Pathology Report Not Reviewed N=18094
Excluded cases N=270 Histology N=56 Rectal N=134 Inconsistent dates N=36 Pre-op chemo N=44
Stage 0 N=31 Stage IV N= 443 Stage I/II/III cases N=6775
16
ACCEPTED MANUSCRIPT
Table 1. Demographic and disease-related characteristics of patients with resected colon cancer in Ontario 2002-2008 (n=6775).
N=107
46% 54%
96% ≤5% ≤2%
EP
5,296 (78%) 1,185-1,190 (18%) 288-293 (4%)
11% 23% 18% 21% 25% ≤2%
TE D
1,436 (21%) 1,617 (24%) 1,460 (22%) 1,217 (18%) 1,025-1,030 (15%) 15-20 (0%)
3,110 (46%) 3,320-3,325 (49%) 340-345 (5%)
54% 46%
50% 46% ≤5%
N=963
M AN U
3,307 (49%) 3,468 (51%)
N=341
RI PT
N=6,775
18% 19% 18% 24% 20% 0%
45% 55%
SC
≤40
AC C
Characteristic Patient-related Sex Female Male Socioeconomic Status1 Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 Unknown Charlson Comorbidity 0 1-2 3+ Disease-related Laterality Left Right Both/Unstated
Overall
Age Group 41-50 51-60
17% 22% 21% 20% 19% 0%
>60
N=5,364
p value 0.013
49% 51% <0.001 22% 24% 22% 17% 14% 0% <0.001
96% 4% ≤1%
89% 9% 2%
75% 20% 5% <0.001
57% 38% 5%
53% 42% 5%
44% 51% 5%
17
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
0.084 Histologic grade Low-moderate grade 5,589 (82%) 75% 81% 85% 82% High grade 1,050-1,055 (16%) 22% 18% 13% 16% Unstated 130-135 (2%) ≤5% 1% 2% 2% 0.005 Lymphovascular invasion Yes 1,890 (28%) 35% 31% 29% 27% No 4,340-4,345 (64%) 61-65% 57% 61% 65% NA 540-545 (8%) ≤5% 11% 10% 8% Tstage 0.005 pT1 405-410 (6%) ≤5% 6% 7% 6% pT2 1,040-1,045 (15%) 3-7% 11% 14% 16% pT3 4,121 (61%) 64% 60% 62% 61% pT4 1,204 (18%) 24% 23% 17% 18% <0.001 Nstage Negative 3,812 (56%) 40% 47% 50% 58% Positive 2,915-2,920 (43%) 58% 52% 49% 41% NX 40-45 (1%) ≤5% 1% 1% 1% <0.001 Stage I 1,256 (19%) 8% 14% 17% 19% II 2,599 (38%) 34% 34% 34% 40% III 2,920 (43%) 58% 52% 49% 41% 1 Quintile 1 represents the communities in which the poorest 20% of the Ontario population resided. As per Institute of Clinical Evaluative Sciences policy, cells with less than 6 cases were suppressed; results in these fields are reported as approximate values to ensure that the precise small cell value cannot be determined.
18
ACCEPTED MANUSCRIPT
Any adjuvant chemotherapy No Yes
N=107
13% 87%
TE D
12.16 8 2,043 (30%) 2,505 (37%) 2,227 (33%)
EP
16.13 14 2,071 (31%) 4,645-4,650 (69%) 55-60 (1%) All N=5519 3,192 (58%) 2,327 (42%)
N=341
N=963
N=5,364
P value 0.450
18% 82%
19% 81%
19% 81%
10.16 7 37% 39% 23%
9.97 7 38% 41% 21%
9.24 7 41% 39% 21%
12.86 8 28% 36% 36%
25.14 20 13% 84-89% ≤5% <40 N=98
18.54 16 28% 71% 1% 41-50 N=292
17.53 15 27% 72% 1% 51-60 N=797
15.55 14 32% 67% 1% >60 N=4332
M AN U
1,262 (19%) 5,513 (81%)
SC
N=6,775
AC C
Characteristic Surgical Characteristics Surgical procedure Laparoscopic Open Length of hospital stay (days) Mean Median <7 7-10 >10 Lymph node yield Mean Median <12 ≥12 NA Stages II and III Only
RI PT
Table 2. Surgical characteristics and delivery of adjuvant chemotherapy among 6775 patients with colon cancer resected in Ontario during 2002-2008. AgeGroup All ≤40 41-50 51-60 >60
<0.001
<0.001
<0.001 21% 79%
29% 71%
37% 63%
64% 36%
19
ACCEPTED MANUSCRIPT
7
7
7
1,251 (23%) 1,076 (19%) All N=2599
48% 31% <40 N=36
41% 30% 41-50 N=115
38% 25% 51-60 N=326
18% 17% >60 N=2122
2,135 (82%) 464 (18%) All N=2920
50% 50% <40 N=62
50% 50% 41-50 N=177
64% 36% 51-60 N=471
87% 13% >60 N=2210
RI PT
7
SC
Adjuvant chemotherapy for Stage II No Yes Stage III Only
0.002 7
M AN U
Time from surgery to chemotherapy Median (weeks) Proportion within 8 weeks <8 ≥8 Stage II Only
<0.001
AC C
EP
TE D
Adjuvant chemotherapy for Stage III <0.001 No 1,055-1,060 (36%) ≤5% 15% 18% 43% Yes 1,860-1,865 (64%) ≥92% 85% 82% 57% As per Institute of Clinical Evaluative Sciences policy, cells with less than 6 cases were suppressed; results in these fields are reported as approximate values to ensure that the precise small cell value cannot be determined.
20
ACCEPTED MANUSCRIPT
Figure 1. Unadjusted and adjusted overall (OS) and cancer-specific (CSS) survival of 6775 patients with stage I,II,III colon cancer treated with surgical resection in Ontario, Canada 20022008.
AC C
EP
B: Unadjusted CSS
TE D
M AN U
SC
RI PT
A: Unadjusted OS
21
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
C: Adjusted OS
AC C
EP
TE D
D: Adjusted CSS
Covariates adjusted for: sex, ses, comorbidity, stage, lvi, histologic grade, act, laterality
22
ACCEPTED MANUSCRIPT
Table 3. Variables associated with survival among 6775 patients with stages resected colon cancer in Ontario 2002-2008. Overall Survival P <0.001
60% 63%
Ref 0.83 (0.77-0.89)
65% 53% 42%
Ref 1.47 (1.34-1.62) 1.88 (1.70-2.08)
HR (95% CI)
82% 71% 75% 68%
<0.001
68% 71%
Ref 0.82 (0.76-0.90)
70% 65% 65%
Ref 1.17 (1.03-1.33) 1.18 (1.02-1.37)
TE D
<0.001
79% 62% 39%
0.009
0.001
0.387 66% 68% 71% 72% 71%
EP
1.18 (1.05-1.33) 1.20 (1.07-1.34) 1.05 (0.93-1.18) 1.03 (0.92-1.17) Ref
Ref 1.50 (1.36-1.66) 2.47 (2.19-2.78)
<0.001
0.41 (0.26-0.64) 0.77 (0.63-0.94) 0.76 (0.67-0.86) Ref
<0.001
58% 58% 63% 65% 65%
P
SC
0.32 (0.21-0.49) 0.56 (0.47-0.68) 0.62 (0.55-0.69) Ref
M AN U
80% 71% 72% 59%
5 yr CSS
RI PT
HR (95% CI)
AC C
Characteristic Patient-related Age (years) ≤40 (n=107) 41-50 (n=341) 51-60 (n=963) >60 (n=5364) Sex Male (n=3468) Female (n=3307) Charlson comorbidity score 0 (n=5296) 1 (n=847) 2+ (n=632) SES, by quintile1 1 (n=1436) 2 (n=1617) 3 (n=1460) 4 (n=1217) 5 (n=1025) Disease-related Tstage ≤T3 (n=1450) pT3 (n=4121) pT4 (n=1204)
5 yr OS
Cancer Specific Survival
1.09 (0.95-1.25) 1.11 (0.97-1.28) 1.02 (0.88-1.17) 1.01 (0.87-1.18) Ref
<0.001
<0.001 89% 69% 44%
Ref 2.11 (1.80-2.47) 3.95 (3.32-4.69)
23
ACCEPTED MANUSCRIPT
Nstage was unreported for 43 patients.
AC C
2
EP
TE D
M AN U
SC
RI PT
Nstage2 <0.001 <0.001 N0 (n=3812) 73% Ref 83% Ref N1 (n=1727) 54% 2.00 (1.82-2.19) 62% 2.43 (2.16-2.74) N2 (n=1193) 35% 3.17 (2.85-3.52) 38% 4.49 (3.95-5.10) Histologic grade <0.001 <0.001 Low-mod grade (n=5589) 64% Ref 72% Ref High grade (n=1053) 47% 1.24 (1.13-1.35) 54% 1.32 (1.19-1.47) Unstated (n=133) 71% 0.83 (0.64-1.08) 77% 0.81 (0.57-1.14) Laterality 0.759 0.455 Left (n=3110) 63% Ref 70% Ref Right (n=3323) 60% 1.00 (0.94-1.08) 69% 0.95 (0.87-1.04) Both/Unstated (n=342) 59% 1.06 (0.91-1.23) 66% 1.02 (0.84-1.23) LVI <0.001 <0.001 No (n=4342) 69% Ref 78% Ref Yes (n=1890) 44% 1.29 (1.19-1.40) 50% 1.35 (1.22-1.50) NA (n=543) 59% 1.23 (1.09-1.39) 66% 1.36 (1.17-1.59) Treatment-related Adjuvant chemotherapy <0.001 <0.001 No (n=4409) 62% Ref 73% Ref Yes (n=2366) 61% 0.47 (0.43-0.52) 64% 0.52 (0.47-0.58) 1 SES = Socioeconomic status. Quintile 1 represents the communities in which the poorest 20% of the Ontario population resided. 20 patients were excluded due to missing SES data.
HR = hazard ratio; CI = confidence interval; LVI = Lymphovascular invasion
24
ACCEPTED MANUSCRIPT
Table 4. Variables associated with survival in 2920 patients with stage III resected colon cancer in Ontario 2002-2008.
HR (95% CI)
71% 56% 58% 43%
0.43 (0.27-0.68) 0.73 (0.58-0.91) 0.75 (0.65-0.86) Ref
P
5 yr CSS
<0.001
HR (95% CI)
M AN U
Ref 0.86 (0.78-0.94)
51% 53%
0.018 Ref 0.88 (0.79-0.98)
<0.001
Ref 1.20 (1.05-1.38) 1.50 (1.29-1.74)
TE D
50% 37% 28%
54% 45% 45%
0.542 Ref 1.06 (0.90-1.25) 1.09 (0.90-1.32)
0.092
45% 41% 48% 49% 52%
75% 50% 31%
1.10 (0.94-1.28) 1.22 (1.05-1.42) 1.06 (0.90-1.24) 1.07 (0.90-1.26) Ref
Ref 1.65 (1.33-2.04) 2.59 (2.07-3.24)
<0.001
0.49 (0.31-0.78) 0.86 (0.69-1.08) 0.79 (0.68-0.92) Ref
0.002
46% 47%
P
SC
73% 56% 60% 49%
EP
Patient-related Age (years) ≤40 (n=62) 41-50 (n=177) 51-60 (n=471) >60 (n=2210) Sex Male (n=1520) Female (n=1400) Charlson comorbidity score 0 (n=2343) 1 (n=330) 2+ (n=247) SES, by quintile1 1 (n=619) 2 (n=686) 3 (n=617) 4 (n=524) 5 (n=468) Disease-related Tstage ≤T3 (n=225) pT3 (n=1928) pT4 (n=767)
5 yr OS
AC C
Characteristic
Cancer Specific Survival
RI PT
Overall Survival
0.339 51% 48% 53% 54% 56%
1.02 (0.86-1.22) 1.16 (0.98-1.37) 1.01 (0.85-1.20) 1.06 (0.89-1.28) Ref
83% 55% 34%
Ref 2.24 (1.67-3.00) 3.79 (2.81-5.11)
<0.001
<0.001
25
ACCEPTED MANUSCRIPT
EP
TE D
M AN U
SC
RI PT
Nstage <0.001 <0.001 N1 (n=1727) 54% Ref 62% Ref N2 (n=1193) 35% 1.57 (1.43-1.73) 38% 1.89 (1.70-2.11) Histologic grade <0.001 <0.001 Low-mod grade (n=2220) 49% Ref 55% Ref High grade (n=648) 36% 1.26 (1.12-1.40) 41% 1.31 (1.16-1.48) Unstated (n=52) 54% 0.90 (0.62-1.30) 60% 0.83 (0.53-1.31) Laterality 0.265 0.528 Left (n=1382) 50% Ref 56% Ref Right (n=1389) 43% 1.02 (0.93-1.13) 49% 1.03 (0.92-1.15) Both/Unstated (n=149) 43% 1.19 (0.97-1.46) 49% 1.14 (0.90-1.45) LVI <0.001 <0.001 No (n=1245) 55% Ref 62% Ref Yes (n=1445) 40% 1.24 (1.12-1.38) 44% 1.29 (1.15-1.45) NA (n=230) 44% 1.25 (1.05-1.49) 47% 1.36 (1.11-1.66) Treatment-related <0.001 Adjuvant chemotherapy <0.001 No (n=1057) 27% 37% Ref Ref Yes (n=1863) 58% 0.41 (0.37-0.45) 60% 0.45 (0.40-0.50) 1 SES = Socioeconomic status. Quintile 1 represents the communities in which the poorest 20% of the Ontario population resided. 6 patients were excluded due to missing SES data.
AC C
HR = hazard ratio; CI = confidence interval; LVI = Lymphovascular invasion
26
ACCEPTED MANUSCRIPT
Supplemental eTable. Characteristics of patients treated with surgical resection for colon cancer in Ontario 2002-2008 with and without randomly selected pathology reports. Path Yes no. (%) N=7,519
Path No no. (%) N=18,094
AC C
EP
TE D
M AN U
SC
RI PT
Age, years 20-49 470 (6%) 1,195 (7%) 50-59 1,025 (14%) 2,724 (15%) 60-69 1,804 (24%) 4,485 (25%) 70-79 2,477 (33%) 5,938 (33%) 80+ 1,743 (23%) 3,752 (21%) Sex Male 3,842 (51%) 9,403 (52%) Female 3,677 (49%) 8,691 (48%) # SES by quintile 1 1,584 (21%) 4,047 (22%) 2 1,777 (24%) 4,123 (23%) 3 1,627 (22%) 3,702 (20%) 4 1,357 (18%) 3,311 (18%) 5 1,152 (15%) 2,835 (16%) Charlson co-morbidity score 0 5890 (78%) 14330 (79%) 1 941 (13%) 2144 (12%) 2+ 688 (9%) 1620 (9%) # Socioeconomic status, Quintile 1 represents the communities where the poorest 20% of the Ontario population resided. SES data were not available for 98/25613 (0.4%) patients.
27
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Supplemental eFigure 2. Cancer-specific survival curves of patients treated with surgical resection for colon cancer in Ontario 2002-2008 with and without randomly selected pathology reports.
28
S1533-0028(18)30168-3
DOI:
10.1016/j.clcc.2018.06.007
Reference:
CLCC 480
To appear in:
Clinical Colorectal Cancer
Received Date: 13 April 2018 Revised Date:
18 June 2018
Accepted Date: 27 June 2018
Please cite this article as: Rodriguez L, Brennan K, Karim S, Nanji S, Patel SV, Booth CM, Disease Characteristics, Clinical Management, and Outcomes of Young Patients with Colon Cancer: A Population-Based Study, Clinical Colorectal Cancer (2018), doi: 10.1016/j.clcc.2018.06.007. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
RI PT
ACCEPTED MANUSCRIPT
Disease Characteristics, Clinical Management, and Outcomes of Young Patients with Colon Cancer: A Population-Based Study
SC
Laura Rodriguez1, Kelly Brennan2, Safiya Karim2,3,
1
M AN U
Sulaiman Nanji3,4, Sunil V. Patel3,4, Christopher M. Booth2,3,5
Canadian Cancer Trials Group
2
Division of Cancer Care and Epidemiology, Queen’s University Cancer Research Institute, Departments of Oncology3, Surgery4, Public Health Sciences5
TE D
Queen’s University, Kingston, Canada
AC C
EP
Corresponding author: Dr. Christopher Booth Division of Cancer Care and Epidemiology Queen’s University Cancer Research Institute 10 Stuart Street Kingston, ON, K7L 3N6, Canada Telephone: (613) 533-6895 Fax: (613) 533-6794 Email: [email protected]
Running title: Young patients with colon cancer Word Count: 2615 June 14, 2018
1
ACCEPTED MANUSCRIPT
ABSTRACT Introduction: The incidence of colorectal cancer in young patients is increasing. Here we evaluate whether disease characteristics, management, and outcomes of patients with colon
RI PT
cancer differs among patients <40 years compared with older patients.
Methods: Using the Ontario Cancer Registry, all cases of colon cancer (stage I, II, III) treated with surgery in Ontario during 2002–2008 were identified. Electronic records of treatment
SC
identified use of surgery and adjuvant chemotherapy (ACT). Pathology reports were obtained for a random 25% sample of all cases. A Cox model was used to identify factors associated with
M AN U
overall (OS) and cancer-specific survival (CSS).
Results: The study population included 6775 patients; age distribution was 2%, 5%, 14%, 79% for ≤40, 41-50, 51-60, >60 years of age. Compared to patients >60 years, younger patients (<40) were more likely to have lymphovascular invasion (35% vs 27%, p=0.005), T3/4 tumours
TE D
(88% vs 79%, p=0.005) and node positive disease (58% vs 41%, p<0.001). Stage distribution varied by age: stage I, 8% (<40 years) vs 19% (>60 years); stage II, 34% vs 40%; stage III, 58% vs 41% (p<0.001). ACT was delivered more commonly for patients <40 years compared to >60
EP
years for stage II (50% vs 13%, p<0.001) and stage III (≥92% vs 57%, p<0.001) disease. Adjusted OS (HR 0.32, 95%CI 0.21-0.49) and CSS (HR 0.41, 95%CI 0.26-0.64) were superior
AC C
for patients <40 compared to those >60 years of age. Conclusions: Young patients with colon cancer have more aggressive and advanced disease but improved outcomes compared to older patients.
2
ACCEPTED MANUSCRIPT
INTRODUCTION Colorectal cancer (CRC) is the second leading cause of death from cancer in men and the third leading cause of death from cancer in women in Canada (1). Recent declines in the
RI PT
incidence and mortality rates of CRC have been attributed to adoption of screening which is recommended for adults >50 years of age (2, 3). However, emerging data suggests that the
incidence of CRC in young adults (i.e. <40 years) is increasing (4, 5). There are conflicting
SC
reports in the literature as to whether disease biology varies by age. While some studies have found that younger patients present with more aggressive disease and advanced stage compared
M AN U
with older patients (6-8); other studies contradict these findings (9, 10). However, much of this literature comes from single-centre and/or pooled institutional case series. There is also uncertainty in the literature about whether the outcome of CRC differs by age. Some investigators have reported that younger patients have inferior outcomes (11-13), while others
TE D
have reported superior survival among younger patients (6, 14, 15).
The objective of this study was to determine whether disease characteristics, management and outcomes were different in young patients (≤40) compared with older age patients (> 60)
AC C
EP
within a Canadian population.
3
ACCEPTED MANUSCRIPT
METHODS Study Design and Population This is a population-based, retrospective cohort study of patients treated for colon cancer
RI PT
in the Canadian province of Ontario. Ontario has a population of approximately 13.5 million people and a single-payer universal health insurance program. The study population included a 25% random sample of all patients who underwent resection of primary colon cancer in Ontario
SC
between 2002 and 2008. To identify the study cohort, we used the Ontario Cancer Registry
(OCR) to identify all incident cases of colon cancer in Ontario diagnosed during 2000-2008. The
M AN U
OCR does not capture stage of disease; therefore, we obtained surgical pathology reports for a random sample of 25% of cases. The substantial workload burden of manual record review precluded us from including 100% of cases in Ontario. Cases were randomly selected using a computer generated random sample algorithm. Reports were not available for patients with
TE D
surgery in 2005; as such the study cohort is restricted to patients who had surgery in 2002-2004 and 2006-2008. Because we relied on review of surgical pathology reports to assign stage of disease, the study excluded patients with rectal cancer as neoadjuvant therapy would
EP
undoubtedly influence the pathologic findings at surgery. The study was approved by the Research Ethics Board of Queen’s University, Kingston, Canada and the institutional review
AC C
board at Sunnybrook Health Sciences Centre, Toronto, Canada. This study was designed, analyzed, and reported in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) statement (16, 17). Data Sources and Linkage
The OCR is a passive, population-based cancer registry that captures diagnostic and demographic information on at least 98% of all incident cases of cancer in the province of
4
ACCEPTED MANUSCRIPT
Ontario (18). The OCR also provides information about vital status and cause of death. Records of hospitalization from the Canadian Institute for Health Information (CIHI) provided information about surgical procedures; these records are known to have a very high level of
RI PT
completeness for colorectal cancer surgery (19). Provincial physician billing records from the Ontario Health Insurance Plan, treatment records from regional cancer centers, and provincial records of chemotherapy delivery were used to identify chemotherapy utilization. A team of
SC
trained data abstractors reviewed the pathology reports and entered information about extent of disease into an electronic database. Patient with evidence of metastatic disease based on the
M AN U
pathology report from the primary tumour resection (i.e. liver/omentum/peritoneum, ovary) were classified as stage IV and excluded from the current study. These datasets were linked using unique encoded identifiers and analyzed at the Institute for Clinical Evaluative Sciences (ICES). Exposures and Outcomes
TE D
Patient age was classified at the time of surgery as < 40, 41-50, 51-60, >60 years. The age groups were created a priori; we selected 40 as the upper limit of the younger group as we were most interested in those patients in their 20s and 30s. Indicators of the socioeconomic status
EP
(SES) of the community in which patients resided at diagnosis were linked as described previously (20). Quintiles (Q) of the median household income were based on the household
AC C
income distribution for the full province of Ontario. Q1 represents the communities where the poorest 20% of the Ontario population resided. Co-morbidity was classified using the Charlson Index modified for administrative data based on all non-cancer diagnoses recorded during any hospital admission within 5 years prior to surgery (21). Adjuvant chemotherapy (ACT) was defined as chemotherapy initiated within 16 weeks after surgery. High-risk stage II colon cancer was defined as having at least one of the following:
5
ACCEPTED MANUSCRIPT
T4 tumour, poorly differentiated histology, < 12 lymph nodes, lymphovascular invasion. Study data-sets did not allow us to identify bowel perforation or obstruction. Disease laterality was defined from the pathology report based on right-sided (cecum, ascending colon, hepatic flexure,
RI PT
transverse colon) versus left-sided (splenic flexure, descending colon, sigmoid colon, rectosigmoid colon) resections.
Overall (OS) and cancer-specific survival (CSS) were determined from date of primary
SC
tumour resection. To account for possible cause of death miscoding, CSS included death from any cancer. Complete information about vital status in the OCR was available up to December
M AN U
31, 2014; cause of death was available up to December 31, 2012. Statistical Analysis
The primary objective was to compare disease characteristics, management, and outcome of patients <40 years of age with those >60 years. Comparisons of proportions between study
TE D
groups were made using the chi-square test. Survival was determined from date of surgery using the Kaplan-Meier technique. The association between patient-, disease-, and treatment-related factors with overall/cancer-specific survival was evaluated using the Cox proportional hazards
EP
regression model. A priori we planned sub-group analyses to explore whether management and outcome of stage III colon cancer differed by age groups. To explore effect modification of the
AC C
association of age groups and survival by adjuvant chemotherapy we tested for interaction between ACT and age groups in the Cox proportional hazards regression model for patients with stage 3 disease. Results were considered statistically significant at p-value < 0.05. As per institutional policy, data that relate to <6 patients are not reported due to small size. All analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC).
6
ACCEPTED MANUSCRIPT
RESULTS Study population The linked administrative data-sets identified 25 613 potentially eligible patients who
RI PT
underwent resection of colon cancer in 2002-2008 (Supplemental eFigure 1). Surgical pathology reports were reviewed for 7519 randomly selected cases. The age, sex, comorbidity, and survival of the randomly selected cases did not differ significantly from those of the 18 094 cases not
SC
selected (Supplemental eTable and eFigure 2). Among the 7519 randomly selected cases, 270 (4%) were excluded as shown in Supplemental eFigure 1; cases with stage 0 (n=31) and resected
M AN U
stage IV (n=443) disease were excluded from the study population. Accordingly, the study population included 6775 patients with stage I, II, III colon cancer. Characteristics of the study population are shown in Table 1. Overall the median age was 72 years and 51% were male. Stage distribution among all cases was: 19% stage I, 38% stage II, and 43% stage III. Young
(p<0.001). Disease Characteristics
TE D
patients were more likely to be of higher SES, less co-morbidity, and more advanced disease
EP
Compared to patients >60 years of age, younger patients were more likely to have leftsided tumours (50% vs 44%, p<0.001), lymphovascular invasion (35% vs 27%, p=0.005), T3/4
AC C
tumours (88% vs 79%, p=0.005) and node positive disease (58% vs 41%, p<0.001). Stage distribution accordingly varied by age: stage I, 8% (<40 years) vs 19% (>60 years); stage II, 34% vs 40%; stage III, 58% vs 41% (p<0.001). Treatment delivery
Surgical care and delivery of chemotherapy is described in Table 2. Length of stay was shorter in young patients compared to those >60 years (median 7 vs 8 days, p<0.001). Lymph
7
ACCEPTED MANUSCRIPT
node harvest was substantially greater in patients ≤40 compared to >60 years (median 20 vs 14, p<0.001) Compared to patients >60 years of age, adjuvant chemotherapy was delivered more
RI PT
commonly for younger patients for stage II (50% vs 13%, p<0.001) and stage III (>92% vs 57%, p<0.001) disease. ACT treatment rates for high-risk stage II disease were 71% for younger
patients compared to 17% for patients >60 years (p<0.001). ACT treatment rates for stage II
SC
disease without high risk features were 36% for younger patients and 8% for those >60 years (p<0.001). Chemotherapy was initiated more promptly after surgery for younger patient than
M AN U
those >60 years (48% within 8 weeks of surgery vs 18%, p=0.002). Outcomes
Unadjusted five-year OS/CSS for the full study population was 80%/82% (<40 years), 71%/71% (41-50), 72%/75% (51-60), and 59%/68% (>60) (Figure 1).
TE D
Factors associated with OS and CSS in the full study population are shown in Table 3. Adjusted OS (HR 0.32, 95%CI 0.21-0.49) and CSS (HR 0.41, 95%CI 0.26-0.64) was superior among patients <40 years compared to those >60 years of age. Similar results were observed
EP
when stratified Cox model were generated for those patients with stage III disease (Table 4) [OS HR 0.43 (95%CI 0.27-0.68); CSS HR 0.49 (95%CI 0.31-0.78)] and stage II disease [OS HR 0.09
AC C
(95%CI 0.02-0.35); CSS HR 0.16 (95%CI 0.04-0.64)] (data not shown). Adjuvant chemotherapy was associated with improved survival in the overall cohort and the stage III sub-group. Among the stage III sub-group, a test for interaction between ACT and age suggested there was no differential effect of ACT on survival by age (OS p=0.831, CSS p=0.851).
8
ACCEPTED MANUSCRIPT
DISCUSSION In this study we have explored disease characteristics, management and outcome of early
RI PT
stage colon cancer among patients 40 years and younger. Our data demonstrate that younger patients have more aggressive histology and more advanced stage at diagnosis. They are more likely to receive adjuvant chemotherapy, have a shorter interval between surgery and initiating chemotherapy, and a greater lymph node harvest. Despite more aggressive and extensive disease,
SC
young patients have improved long-term overall and cancer-specific survival compared to older
M AN U
patients. Our data do not suggest that adjuvant chemotherapy has a differential effect across age groups.
Our results are consistent with several single center retrospective studies. Murata et al, reported a retrospective study of 2338 patients with diagnosis of CRC between 2004 and 2012, patients <40 years (n=81) old had a lower incidence of right-sided colon cancer and higher
TE D
number of retrieved lymph nodes, and were more likely to have undergone postoperative adjuvant treatment compared with the older group; however, survival was comparable to older patients (22). Another consecutive series of CRC patients from 1999-2007 (n=2220), reported
EP
young patients having more aggressive and advanced cancer stages (14). However, older patients
AC C
had higher postoperative mortality and were less likely to receive chemotherapy. OS was 69, 68, and 40% (p<0.001) among young (16-53 years), median (53-84 years) and old (85-97 years) group respectively; CSS was 70, 70, and 52% (p<0.003) respectively. One notable limitation of this study was the lack of adjusted survival analyses. The proportion of cancer-related deaths was lower among older patients. A study from the Memorial Sloan-Kettering Cancer Center, evaluated a prospective database of 1327 surgical stage I–III colon cancer patients (1990-2001) (15). Compared to older patients, those <40 years of age were more likely to have left-sided
9
ACCEPTED MANUSCRIPT
tumors and higher nodal yields, and were more likely to receive adjuvant chemotherapy. The 5year CSS was similar in both study groups: 86% vs 87%, but 5- year OS was significantly higher in the younger patient cohort (84% vs 73%, p=0.001).
RI PT
Our results are also consistent with some tumor characteristics, clinical presentation and outcomes, reported by previous population-based cohort studies from the Surveillance,
Epidemiology, and End Results (SEER) registry. O’Connell et al, reported that CRC patients
SC
<40 years old had more aggressive histology and more advanced stage than older patients.
Younger patients had superior survival for stage II and IV lesions but comparable survival for
M AN U
stage I and III tumors (6). Wang et al found that patients <50 years had later stage presentation, more aggressive pathological features, and higher lymph node yields (3). The 5-year CSS were 65%, 67%, and 63%, for patients <40, 41 to 50 and > 50 years old patients respectively. Despite having higher risk disease, in univariate and multivariate analyses younger patients had
TE D
improved cancer survival.
A consistent finding in our study and existing reports is the observation that younger patients present with more advanced stage disease. This may be explained by the fact that current
EP
screening guidelines do not recommend routine screening for this seemingly low-risk population
AC C
(23). This could result in younger patients being diagnosed once symptomatic and thus at a later stage. In addition, younger patients may be diagnosed at a later stage due to delays in seeking medical attention and subsequently referral to specialist care (24). Younger patients symptoms may be more likely to be attributed to benign causes (i.e. hemorrhoids) than older patients. The paradoxical observation that younger patients have superior survival may relate to less comorbidity, fewer complications related to surgery, improved tolerance of adjuvant chemotherapy, or intrinsic differences in tumour biology. The greater nodal harvest among
10
ACCEPTED MANUSCRIPT
young patients may also be a mitigating factor; however controversy exists as to whether greater lymph node yield is intrinsically therapeutic, whether it leads to appropriate upstaging, whether it serves as a marker of quality care, or whether it is as an epiphenomenon of underlying host-
RI PT
tumour interaction (25, 26). Our data also show higher use of adjuvant chemotherapy in younger patients for both stage III and stage II disease. Among the latter, ACT rates may be
inappropriately high for young patients with stage II disease who do not have any high-risk
SC
pathologic features.
There is emerging interest in the role of human papilloma virus (HPV) in colorectal
M AN U
cancer (27). While our data do not offer any specific insights into this issue, further work is needed to understand if HPV may explain the rising incidence of CRC in younger populations. This study has several limitations. The major limitation was its retrospective design, in which we cannot ascertain causation, as overall survival and treatment choice is subject to
TE D
considerable selection bias that cannot be entirely mitigated with risk adjustment. It is worth noting that one would expect fundamental differences in younger patients with CRC as they are more likely to present with symptomatic disease whereas older patients are more likely to be
EP
diagnosed through screening programs. Although we manually reviewed surgical pathology reports for all cases, we did not have access to other elements of the patient chart. This prevented
AC C
us from being able to provide any additional information regarding risk factors for colon cancer (i.e. family history, smoking status, diabetes). The existing data sources also lack information regarding performance status and patient preferences regarding treatment selection. We also were unable to explore the proportion of patients with Lynch Syndrome or microsatellite instability which one would expect to disproportionately affect younger patients. Finally, a significant limitation of this study is the small number of young patients included in the study
11
ACCEPTED MANUSCRIPT
(n= 107). Despite these limitations, the use of a population-based data provides a large sample size and reduces many of the selection and referral biases that plague single centre institutional case series.
RI PT
In summary, our data demonstrate that patients <40 years of age have more advanced disease and more aggressive biology than older patients. It is unclear if this is driven by
fundamental differences in disease biology or whether it simply reflects the fact that older
SC
patients are more likely to have CRC screening. Despite adverse prognostic features, younger patients have improved outcomes compared to older patients. Further work is needed to
AC C
EP
TE D
M AN U
understand if this apparent disconnect is explained by intrinsic differences in tumour biology.
12
ACCEPTED MANUSCRIPT
ACKNOWLEDGEMENTS Parts of this material are based on data and information provided by Cancer Care Ontario.
authors and not necessarily those of Cancer Care Ontario.
RI PT
However, the analysis, conclusions, opinions and statements expressed herein are those of the
Parts of this material are based on data and information compiled and provided by CIHI.
author, and not necessarily those of CIHI.
SC
However, the analyses, conclusions, opinions and statements expressed herein are those of the
M AN U
This study was supported by the Institute for Clinical Evaluative Sciences (ICES), which is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care (MOHLTC). The opinions, results and conclusions reported in this paper are those of the authors and are independent from the funding sources. No endorsement by ICES or the Ontario
TE D
MOHLTC is intended or should be inferred.
Dr. Booth had full access to all the data in the study and takes responsibility for the integrity of
FUNDING
EP
the data and the accuracy of the data analysis.
AC C
Dr. Booth is supported as a Canada Research Chair in Population Cancer Care. This work was also supported by the Canada Foundation for Innovation and the Canadian Institutes of Health Research.
DISCLOSURE
The authors report no conflicts of interest.
13
ACCEPTED MANUSCRIPT
REFERENCES
AC C
EP
TE D
M AN U
SC
RI PT
1. Canadian Cancer Society. Canadian Cancer Statistics 2017. Toronto: Canadian Cancer Society; 2017. Avilable at: http://www.cancer.ca/en/cancer-information/cancer-101/canadiancancer-statistics-publication/?region=on. Last accessed on November 10, 2017. 2. National Cancer Institute. SEER Cancer Stat Facts: Colon and Rectum Cancer. National Cancer Institute. Bethesda, MD. Avilabhle at: http://seer.cancer.gov/statfacts/html/colorect.html. Last accessed on November 10, 2017. 3. Wang R, Wang MJ, Ping J. Clinicopathological Features and Survival Outcomes of Colorectal Cancer in Young Versus Elderly: A Population-Based Cohort Study of SEER 9 Registries Data (1988-2011). Medicine (Baltimore). 2015;94(35):e1402. 4. Ahnen DJ, Wade SW, Jones WF, Sifri R, Mendoza Silveiras J, Greenamyer J, et al. The increasing incidence of young-onset colorectal cancer: a call to action. Mayo Clin Proc. 2014;89(2):216-24. 5. Bailey CE, Hu CY, You YN, Bednarski BK, Rodriguez-Bigas MA, Skibber JM, et al. Increasing disparities in the age-related incidences of colon and rectal cancers in the United States, 1975-2010. JAMA Surg. 2015;150(1):17-22. 6. O'Connell JB, Maggard MA, Liu JH, Etzioni DA, Livingston EH, Ko CY. Do young colon cancer patients have worse outcomes? World J Surg. 2004;28(6):558-62. 7. Lee PY, Fletcher WS, Sullivan ES, Vetto JT. Colorectal cancer in young patients: characteristics and outcome. Am Surg. 1994;60(8):607-12. 8. Leff DR, Chen A, Roberts D, Grant K, Western C, Windsor AC, et al. Colorectal cancer in the young patient. Am Surg. 2007;73(1):42-7. 9. Paraf F, Jothy S. Colorectal cancer before the age of 40: a case-control study. Dis Colon Rectum. 2000;43(9):1222-6. 10. Chung YF, Eu KW, Machin D, Ho JM, Nyam DC, Leong AF, et al. Young age is not a poor prognostic marker in colorectal cancer. Br J Surg. 1998;85(9):1255-9. 11. Palmer ML, Herrera L, Petrelli NJ. Colorectal adenocarcinoma in patients less than 40 years of age. Dis Colon Rectum. 1991;34(4):343-6. 12. Marble K, Banerjee S, Greenwald L. Colorectal carcinoma in young patients. J Surg Oncol. 1992;51(3):179-82. 13. Fu JF, Huang YQ, Yang J, Yi CH, Chen HL, Zheng S. Clinical characteristics and prognosis of young patients with colorectal cancer in Eastern China. World J Gastroenterol. 2013;19(44):8078-84. 14. Derwinger K, Kodeda K, Gerjy R. Age aspects of demography, pathology and survival assessment in colorectal cancer. Anticancer Res. 2010;30(12):5227-31. 15. Quah HM, Joseph R, Schrag D, Shia J, Guillem JG, Paty PB, et al. Young age influences treatment but not outcome of colon cancer. Ann Surg Oncol. 2007;14(10):2759-65. 16. Connell L, MacDonald R, McBride T, Peiperl L, Ross A, Simpson P, et al. Observational Studies: Getting Clear about Transparency. Plos Med. 2014;11(8). 17. Vandenbroucke JP, von Elm E, Altman DG, Gotzsche PC, Mulrow CD, Pocock SJ, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): Explanation and Elaboration. Ann Intern Med. 2007;147(8):W163-W94. 18. Clarke EA, Marrett LD, Kreiger N. Cancer registration in Ontario: a computer approach. IARC Sci Publ. 1991(95):246-57. 19. Li X, King C, deGara C, White J, Winget M. Validation of colorectal cancer surgery data from administrative data sources. BMC Med Res Methodol. 2012;12:97. 14
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
20. Mackillop WJ, Zhang-Salomons J, Groome PA, Paszat L, Holowaty E. Socioeconomic status and cancer survival in Ontario. J Clin Oncol. 1997;15(4):1680-9. 21. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol. 1992;45(6):613-9. 22. Murata A, Akiyoshi T, Ueno M, Fukunaga Y, Nagayama S, Fujimoto Y, et al. Clinicopathological characteristics of young patients with sporadic colorectal cancer. Surg Today. 2016;46(10):1166-75. 23. Canadian Task Force on Preventive Health Care. Recommendations on screening for colorectal cancer in primary care. Canadian Medical Assoc J. 2016; 188(5): 340-348. 24. Scott RB, Rangel LE, Osler TM, Hyman NH. Rectal cancer in patients under the age of 50 years: the delayed diagnosis. American journal of surgery. 2016;211(6):1014-8. 25. Del Paggio JC, Nanji S, Wei X, MacDonald PH, Booth CM. Lymph node evaluation for colon cancer in routine clinical practice: a population-based study. Curr Oncol. 2017;24(1):e35e43. 26. Del Paggio JC, Peng Y, Wei X, Nanji S, MacDonald PH, Krishnan Nair C, et al. Population-based study to re-evaluate optimal lymph node yield in colonic cancer. The British journal of surgery. 2017;104(8):1087-96. 27. Baandrup L, Thomsen LT, Olesen TB, Andersen KK, Norrild B, Kjaer SK. The prevalence of human papillomavirus in colorectal adenomas and adenocarcinomas: a systematic review and meta-analysis. Eur J Cancer. 2014;50(8):1446-61.
15
ACCEPTED MANUSCRIPT
Supplemental eFigure 1. Identification of patients with resected colon cancer in Ontario during 2002-2008. All colon cancer diagnosed 2000-2008 N=45383
SC
Colon cancer with primary surgery within 6 months of diagnosis N=37847
RI PT
No primary surgery within 6 months of diagnosis N=7536
M AN U
Surgery not in 2002-2004 or 2006-2008 N=12234
Colon cancer with primary surgery in 2002-2004 or 2006-2008 N=25613
TE D
Random Selection
EP
Pathology Report Reviewed N=7519
AC C
Colon cancer surgical cases N=7249
Pathology Report Not Reviewed N=18094
Excluded cases N=270 Histology N=56 Rectal N=134 Inconsistent dates N=36 Pre-op chemo N=44
Stage 0 N=31 Stage IV N= 443 Stage I/II/III cases N=6775
16
ACCEPTED MANUSCRIPT
Table 1. Demographic and disease-related characteristics of patients with resected colon cancer in Ontario 2002-2008 (n=6775).
N=107
46% 54%
96% ≤5% ≤2%
EP
5,296 (78%) 1,185-1,190 (18%) 288-293 (4%)
11% 23% 18% 21% 25% ≤2%
TE D
1,436 (21%) 1,617 (24%) 1,460 (22%) 1,217 (18%) 1,025-1,030 (15%) 15-20 (0%)
3,110 (46%) 3,320-3,325 (49%) 340-345 (5%)
54% 46%
50% 46% ≤5%
N=963
M AN U
3,307 (49%) 3,468 (51%)
N=341
RI PT
N=6,775
18% 19% 18% 24% 20% 0%
45% 55%
SC
≤40
AC C
Characteristic Patient-related Sex Female Male Socioeconomic Status1 Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 Unknown Charlson Comorbidity 0 1-2 3+ Disease-related Laterality Left Right Both/Unstated
Overall
Age Group 41-50 51-60
17% 22% 21% 20% 19% 0%
>60
N=5,364
p value 0.013
49% 51% <0.001 22% 24% 22% 17% 14% 0% <0.001
96% 4% ≤1%
89% 9% 2%
75% 20% 5% <0.001
57% 38% 5%
53% 42% 5%
44% 51% 5%
17
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
0.084 Histologic grade Low-moderate grade 5,589 (82%) 75% 81% 85% 82% High grade 1,050-1,055 (16%) 22% 18% 13% 16% Unstated 130-135 (2%) ≤5% 1% 2% 2% 0.005 Lymphovascular invasion Yes 1,890 (28%) 35% 31% 29% 27% No 4,340-4,345 (64%) 61-65% 57% 61% 65% NA 540-545 (8%) ≤5% 11% 10% 8% Tstage 0.005 pT1 405-410 (6%) ≤5% 6% 7% 6% pT2 1,040-1,045 (15%) 3-7% 11% 14% 16% pT3 4,121 (61%) 64% 60% 62% 61% pT4 1,204 (18%) 24% 23% 17% 18% <0.001 Nstage Negative 3,812 (56%) 40% 47% 50% 58% Positive 2,915-2,920 (43%) 58% 52% 49% 41% NX 40-45 (1%) ≤5% 1% 1% 1% <0.001 Stage I 1,256 (19%) 8% 14% 17% 19% II 2,599 (38%) 34% 34% 34% 40% III 2,920 (43%) 58% 52% 49% 41% 1 Quintile 1 represents the communities in which the poorest 20% of the Ontario population resided. As per Institute of Clinical Evaluative Sciences policy, cells with less than 6 cases were suppressed; results in these fields are reported as approximate values to ensure that the precise small cell value cannot be determined.
18
ACCEPTED MANUSCRIPT
Any adjuvant chemotherapy No Yes
N=107
13% 87%
TE D
12.16 8 2,043 (30%) 2,505 (37%) 2,227 (33%)
EP
16.13 14 2,071 (31%) 4,645-4,650 (69%) 55-60 (1%) All N=5519 3,192 (58%) 2,327 (42%)
N=341
N=963
N=5,364
P value 0.450
18% 82%
19% 81%
19% 81%
10.16 7 37% 39% 23%
9.97 7 38% 41% 21%
9.24 7 41% 39% 21%
12.86 8 28% 36% 36%
25.14 20 13% 84-89% ≤5% <40 N=98
18.54 16 28% 71% 1% 41-50 N=292
17.53 15 27% 72% 1% 51-60 N=797
15.55 14 32% 67% 1% >60 N=4332
M AN U
1,262 (19%) 5,513 (81%)
SC
N=6,775
AC C
Characteristic Surgical Characteristics Surgical procedure Laparoscopic Open Length of hospital stay (days) Mean Median <7 7-10 >10 Lymph node yield Mean Median <12 ≥12 NA Stages II and III Only
RI PT
Table 2. Surgical characteristics and delivery of adjuvant chemotherapy among 6775 patients with colon cancer resected in Ontario during 2002-2008. AgeGroup All ≤40 41-50 51-60 >60
<0.001
<0.001
<0.001 21% 79%
29% 71%
37% 63%
64% 36%
19
ACCEPTED MANUSCRIPT
7
7
7
1,251 (23%) 1,076 (19%) All N=2599
48% 31% <40 N=36
41% 30% 41-50 N=115
38% 25% 51-60 N=326
18% 17% >60 N=2122
2,135 (82%) 464 (18%) All N=2920
50% 50% <40 N=62
50% 50% 41-50 N=177
64% 36% 51-60 N=471
87% 13% >60 N=2210
RI PT
7
SC
Adjuvant chemotherapy for Stage II No Yes Stage III Only
0.002 7
M AN U
Time from surgery to chemotherapy Median (weeks) Proportion within 8 weeks <8 ≥8 Stage II Only
<0.001
AC C
EP
TE D
Adjuvant chemotherapy for Stage III <0.001 No 1,055-1,060 (36%) ≤5% 15% 18% 43% Yes 1,860-1,865 (64%) ≥92% 85% 82% 57% As per Institute of Clinical Evaluative Sciences policy, cells with less than 6 cases were suppressed; results in these fields are reported as approximate values to ensure that the precise small cell value cannot be determined.
20
ACCEPTED MANUSCRIPT
Figure 1. Unadjusted and adjusted overall (OS) and cancer-specific (CSS) survival of 6775 patients with stage I,II,III colon cancer treated with surgical resection in Ontario, Canada 20022008.
AC C
EP
B: Unadjusted CSS
TE D
M AN U
SC
RI PT
A: Unadjusted OS
21
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
C: Adjusted OS
AC C
EP
TE D
D: Adjusted CSS
Covariates adjusted for: sex, ses, comorbidity, stage, lvi, histologic grade, act, laterality
22
ACCEPTED MANUSCRIPT
Table 3. Variables associated with survival among 6775 patients with stages resected colon cancer in Ontario 2002-2008. Overall Survival P <0.001
60% 63%
Ref 0.83 (0.77-0.89)
65% 53% 42%
Ref 1.47 (1.34-1.62) 1.88 (1.70-2.08)
HR (95% CI)
82% 71% 75% 68%
<0.001
68% 71%
Ref 0.82 (0.76-0.90)
70% 65% 65%
Ref 1.17 (1.03-1.33) 1.18 (1.02-1.37)
TE D
<0.001
79% 62% 39%
0.009
0.001
0.387 66% 68% 71% 72% 71%
EP
1.18 (1.05-1.33) 1.20 (1.07-1.34) 1.05 (0.93-1.18) 1.03 (0.92-1.17) Ref
Ref 1.50 (1.36-1.66) 2.47 (2.19-2.78)
<0.001
0.41 (0.26-0.64) 0.77 (0.63-0.94) 0.76 (0.67-0.86) Ref
<0.001
58% 58% 63% 65% 65%
P
SC
0.32 (0.21-0.49) 0.56 (0.47-0.68) 0.62 (0.55-0.69) Ref
M AN U
80% 71% 72% 59%
5 yr CSS
RI PT
HR (95% CI)
AC C
Characteristic Patient-related Age (years) ≤40 (n=107) 41-50 (n=341) 51-60 (n=963) >60 (n=5364) Sex Male (n=3468) Female (n=3307) Charlson comorbidity score 0 (n=5296) 1 (n=847) 2+ (n=632) SES, by quintile1 1 (n=1436) 2 (n=1617) 3 (n=1460) 4 (n=1217) 5 (n=1025) Disease-related Tstage ≤T3 (n=1450) pT3 (n=4121) pT4 (n=1204)
5 yr OS
Cancer Specific Survival
1.09 (0.95-1.25) 1.11 (0.97-1.28) 1.02 (0.88-1.17) 1.01 (0.87-1.18) Ref
<0.001
<0.001 89% 69% 44%
Ref 2.11 (1.80-2.47) 3.95 (3.32-4.69)
23
ACCEPTED MANUSCRIPT
Nstage was unreported for 43 patients.
AC C
2
EP
TE D
M AN U
SC
RI PT
Nstage2 <0.001 <0.001 N0 (n=3812) 73% Ref 83% Ref N1 (n=1727) 54% 2.00 (1.82-2.19) 62% 2.43 (2.16-2.74) N2 (n=1193) 35% 3.17 (2.85-3.52) 38% 4.49 (3.95-5.10) Histologic grade <0.001 <0.001 Low-mod grade (n=5589) 64% Ref 72% Ref High grade (n=1053) 47% 1.24 (1.13-1.35) 54% 1.32 (1.19-1.47) Unstated (n=133) 71% 0.83 (0.64-1.08) 77% 0.81 (0.57-1.14) Laterality 0.759 0.455 Left (n=3110) 63% Ref 70% Ref Right (n=3323) 60% 1.00 (0.94-1.08) 69% 0.95 (0.87-1.04) Both/Unstated (n=342) 59% 1.06 (0.91-1.23) 66% 1.02 (0.84-1.23) LVI <0.001 <0.001 No (n=4342) 69% Ref 78% Ref Yes (n=1890) 44% 1.29 (1.19-1.40) 50% 1.35 (1.22-1.50) NA (n=543) 59% 1.23 (1.09-1.39) 66% 1.36 (1.17-1.59) Treatment-related Adjuvant chemotherapy <0.001 <0.001 No (n=4409) 62% Ref 73% Ref Yes (n=2366) 61% 0.47 (0.43-0.52) 64% 0.52 (0.47-0.58) 1 SES = Socioeconomic status. Quintile 1 represents the communities in which the poorest 20% of the Ontario population resided. 20 patients were excluded due to missing SES data.
HR = hazard ratio; CI = confidence interval; LVI = Lymphovascular invasion
24
ACCEPTED MANUSCRIPT
Table 4. Variables associated with survival in 2920 patients with stage III resected colon cancer in Ontario 2002-2008.
HR (95% CI)
71% 56% 58% 43%
0.43 (0.27-0.68) 0.73 (0.58-0.91) 0.75 (0.65-0.86) Ref
P
5 yr CSS
<0.001
HR (95% CI)
M AN U
Ref 0.86 (0.78-0.94)
51% 53%
0.018 Ref 0.88 (0.79-0.98)
<0.001
Ref 1.20 (1.05-1.38) 1.50 (1.29-1.74)
TE D
50% 37% 28%
54% 45% 45%
0.542 Ref 1.06 (0.90-1.25) 1.09 (0.90-1.32)
0.092
45% 41% 48% 49% 52%
75% 50% 31%
1.10 (0.94-1.28) 1.22 (1.05-1.42) 1.06 (0.90-1.24) 1.07 (0.90-1.26) Ref
Ref 1.65 (1.33-2.04) 2.59 (2.07-3.24)
<0.001
0.49 (0.31-0.78) 0.86 (0.69-1.08) 0.79 (0.68-0.92) Ref
0.002
46% 47%
P
SC
73% 56% 60% 49%
EP
Patient-related Age (years) ≤40 (n=62) 41-50 (n=177) 51-60 (n=471) >60 (n=2210) Sex Male (n=1520) Female (n=1400) Charlson comorbidity score 0 (n=2343) 1 (n=330) 2+ (n=247) SES, by quintile1 1 (n=619) 2 (n=686) 3 (n=617) 4 (n=524) 5 (n=468) Disease-related Tstage ≤T3 (n=225) pT3 (n=1928) pT4 (n=767)
5 yr OS
AC C
Characteristic
Cancer Specific Survival
RI PT
Overall Survival
0.339 51% 48% 53% 54% 56%
1.02 (0.86-1.22) 1.16 (0.98-1.37) 1.01 (0.85-1.20) 1.06 (0.89-1.28) Ref
83% 55% 34%
Ref 2.24 (1.67-3.00) 3.79 (2.81-5.11)
<0.001
<0.001
25
ACCEPTED MANUSCRIPT
EP
TE D
M AN U
SC
RI PT
Nstage <0.001 <0.001 N1 (n=1727) 54% Ref 62% Ref N2 (n=1193) 35% 1.57 (1.43-1.73) 38% 1.89 (1.70-2.11) Histologic grade <0.001 <0.001 Low-mod grade (n=2220) 49% Ref 55% Ref High grade (n=648) 36% 1.26 (1.12-1.40) 41% 1.31 (1.16-1.48) Unstated (n=52) 54% 0.90 (0.62-1.30) 60% 0.83 (0.53-1.31) Laterality 0.265 0.528 Left (n=1382) 50% Ref 56% Ref Right (n=1389) 43% 1.02 (0.93-1.13) 49% 1.03 (0.92-1.15) Both/Unstated (n=149) 43% 1.19 (0.97-1.46) 49% 1.14 (0.90-1.45) LVI <0.001 <0.001 No (n=1245) 55% Ref 62% Ref Yes (n=1445) 40% 1.24 (1.12-1.38) 44% 1.29 (1.15-1.45) NA (n=230) 44% 1.25 (1.05-1.49) 47% 1.36 (1.11-1.66) Treatment-related <0.001 Adjuvant chemotherapy <0.001 No (n=1057) 27% 37% Ref Ref Yes (n=1863) 58% 0.41 (0.37-0.45) 60% 0.45 (0.40-0.50) 1 SES = Socioeconomic status. Quintile 1 represents the communities in which the poorest 20% of the Ontario population resided. 6 patients were excluded due to missing SES data.
AC C
HR = hazard ratio; CI = confidence interval; LVI = Lymphovascular invasion
26
ACCEPTED MANUSCRIPT
Supplemental eTable. Characteristics of patients treated with surgical resection for colon cancer in Ontario 2002-2008 with and without randomly selected pathology reports. Path Yes no. (%) N=7,519
Path No no. (%) N=18,094
AC C
EP
TE D
M AN U
SC
RI PT
Age, years 20-49 470 (6%) 1,195 (7%) 50-59 1,025 (14%) 2,724 (15%) 60-69 1,804 (24%) 4,485 (25%) 70-79 2,477 (33%) 5,938 (33%) 80+ 1,743 (23%) 3,752 (21%) Sex Male 3,842 (51%) 9,403 (52%) Female 3,677 (49%) 8,691 (48%) # SES by quintile 1 1,584 (21%) 4,047 (22%) 2 1,777 (24%) 4,123 (23%) 3 1,627 (22%) 3,702 (20%) 4 1,357 (18%) 3,311 (18%) 5 1,152 (15%) 2,835 (16%) Charlson co-morbidity score 0 5890 (78%) 14330 (79%) 1 941 (13%) 2144 (12%) 2+ 688 (9%) 1620 (9%) # Socioeconomic status, Quintile 1 represents the communities where the poorest 20% of the Ontario population resided. SES data were not available for 98/25613 (0.4%) patients.
27
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Supplemental eFigure 2. Cancer-specific survival curves of patients treated with surgical resection for colon cancer in Ontario 2002-2008 with and without randomly selected pathology reports.
28