2nd February 2018, Volume 131 Number 1469

Ashok Gunawardene, Brendan Desmond, Ali Shekouh, Peter Larsen, Elizabeth Dennett

The New Zealand Ministry of Health has reported colorectal cancer (CRC) as the third most common cancer after prostate and breast, comprising almost 14% of all those registered in 2012, and as the second leading cause of cancer-related death behind only lung.1 Furthermore, data collated by the International Agency for Research on Cancer (IARC) has placed New Zealand as having one of the highest incidence rates of colorectal cancer in the world.2

Historically, outcomes of CRC in New Zealand have compared unfavourably to international standards with five-year mortality rates around 40%.3,4 While the issue of timely diagnosis will be addressed by the much-anticipated national screening programme, there is a need to re-evaluate oncological outcomes in the context of other recent advances in colorectal cancer management. These include routine multidisciplinary team discussion, surgical sub-specialisation, increased utility of neo-adjuvant and adjuvant therapies and protocolled follow-up and surveillance.

While prior cohort studies have reported on mortality in detail, they have not reported on recurrence rates. This study was conducted to examine more contemporary mortality rates in a New Zealand cohort, and to provide data on recurrence rates.

Methods

Patients with newly diagnosed adenocarcinoma of the colon or rectum between January 2010 and December 2012 at a single tertiary hospital were included in the study. Consecutive patients undergoing curative intent surgery during this time-frame were included provided their follow-up was also at the host institution and patients with a prior history of colorectal malignancy were excluded. Patients with stage IV disease at presentation were included in the study if they had no residual disease following their definitive treatment. This study complies with regulations for audit at the host institution and meets the New Zealand definition of observation research.

A standardised protocol for post-operative follow-up and surveillance was used during the study period and involved three- to six-monthly surgical outpatient visits and yearly computed tomography (CT) scan of the chest, abdomen and pelvis for the first three years. A complete colonoscopy was performed within a year of surgery if one had not been completed pre-operatively with a repeat colonoscopy after three years. Yearly outpatient appointments were scheduled between years three to five and three-monthly serum carcinoembryonic antigen (CEA) was checked for all five years. Surgery for rectal cancer was typically carried out by three colorectal surgeons and colon cancers by general and colorectal surgeons at the hospital.

Time-to-recurrence was calculated from the date of surgery to either histological confirmation or clinical-radiological evidence of loco-regional or distant disease recurrence on review at a multidisciplinary meeting, where the date of meeting was taken as the end-point. Patients that were alive and without evidence of disease recurrence on 1 November 2016 were censored. The site of primary cancer was taken from the operation note and disease stage based on classification in accordance with the American Joint Committee on Cancer (AJCC) Seventh edition.5

Statistical analysis was performed using IBM SPSS Statistics Version 24.0. Categorical variables were compared between groups using Chi squared or Fischer’s exact tests. Kaplan-Meier curves were plotted to evaluate time-to-recurrence and the log-rank test performed to compare groups by stage. Univariate and multivariate survival analyses were performed using Cox proportional hazards model including variables with p<0.1 on univariate analysis into a multivariate model optimised through the backward step-wise elimination method. The null hypothesis was rejected when p values were equal to or below 0.05.

Results

Two hundred and thirty-seven patients were included in the study, including 116 males and 121 females with a median age of 71 years (range 32 to 91 years). Two hundred and nine (88.2%) were of European, 10 (4.2%) of Māori and nine (3.8%) patients were of Pacific Island ethnicity. Fifty-nine (24.9%) cancers were rectal, 21 (8.9%) recto-sigmoid and the remaining 157 (66.2%) colonic. Thirty-six (15.2%) patients required an emergency operation, which was most commonly due to obstruction (Table 1).

Table 1: Site, stage and presentation of patients in the cohort.

 

Patients

N=237 (%)

Age

71 (range 32–91)

Gender M:F

116:121

Comorbidities

0

156 (65.8)

1

63 (26.6)

2+

18 (7.6)

Site

Rectum

59 (24.9)

Recto-sigmoid

21 (8.9)

Colon

157 (66.2)

Stage

I

39 (16.5)

II

90 (38.0)

III

98 (41.4)

IV

10 (4.2)

Presentation

Emergency, all

36 (15.2)

Obstruction

26 (11.0)

Perforation

10 (4.2) 

The operations performed for primary colorectal cancer were, right hemicolectomy or extended right hemicolectomy (n=97); transverse colectomy (n=3); left hemicolectomy (n=9); anterior resection (n=81); Hartmann’s procedure (n=15); abdominal perineal resection (APR) (n=20); subtotal colectomy (n=11); proctocolectomy (n=2); transanal excision (n=1).

Of the 10 patients with stage IV disease at presentation, one had an oophorectomy for an ovarian metastasis at the time of initial surgery. The remaining nine had surgery for synchronous metastases following an interval (range 98–215 days) after the initial surgery. Six patients underwent segmental liver resection, two hemi-hepatectomy and one open excision of a lingular mass of the lung.

The median follow-up duration was 61 months (46–81 months) and overall survival rate was 68.6%. In total, 59 (24.9%) patients developed disease recurrence at a median time of 14.0 months. The annual recurrence rates were 10.5% for year one, 7.6% for year two, 3.8% in year three, 2.1% in year four and 0.8% for year five and greater. The median time to local recurrence was 12.5 months (IQR 8.5–26.0) with a similar time for distant recurrence at 14.0 months (IQR 8.0–25.0).

Of 59 patients with rectal cancer, 39 (66.1%) underwent neo-adjuvant therapy. Fifteen (25.4%) of these received short-course radiotherapy and 24 (40.7%) long-course chemo-radiation. Seventy-seven (32.5%) patients of the full cohort received adjuvant chemotherapy: 2.6% for stage I, 18.2% for stage II, 72.7% for stage III and 100% for stage IV.

The most common site of disease recurrence was the liver (n=25) followed by the lung (n=16) and local (n=15) recurrence in colorectal cancer patients combined (Figure 1).The overall recurrence rate was 26.4% for colon and 20.3% for rectal cancer (p=0.35); liver recurrences occurred in 11.8% of colon and 6.8% of rectal cancers (p=0.28); lung recurrences in 5.6% of colon and 10.2% of rectal cancers (p=0.23) and local recurrence occurred in 7.9% of colon cancer and 1.7% of rectal cancer (p=0.09). Recurrences are summarised by stage in Table 2.

Figure 1: Sites of recurrences.

c 

Table 2: Recurrences by stage and site of primary.

Stage recurrence site

Colon

Rectum

Total

Stage I

21

18

39

Local

0 (0)

0 (0)

0 (0)

Distant

0 (0)

2 (11.1)

2 (5.1)

Stage II

73

17

90

Local

8 (10.9)

1 (5.9)

9 (10.0)

Distant

7 (9.59)

4 (23.5)

11 (12.2)

Stage III

77

21

98

Local

6 (7.8)

0 (0)

6 (6.1)

Distant

21 (27.3)

4 (22.2)

25 (25.5)

Stage IV

7

3

10

Local

0 (0)

0 (0)

0 (0)

Distant

5 (71.4)

1 (33.3)

6 (60.0) 

In Figure 2, Kaplan-Meier curves reveal recurrences to increase significantly by advancing disease stage (p=0.001). The recurrence rates by stage are 5.1% in stage I, 22.2% in stage II, 31.6% in stage III and 60% in stage IV. On univariate analysis, stage, emergency presentation, histological grade, lymphovascular invasion and perineural invasion were associated with recurrence although following multivariate analysis only disease stage was found to be independently predictive (Table 3).

Figure 2: Kaplan-Meier Curves for time-to-recurrence by stage. 

c 

Table 3: Univariate and multivariate analysis: time-to-recurrence.

 

 

Univariate analysis

Multivariate analysis

HR (95% confidence interval)

p value

HR (95% confidence interval)

p value

Stage

I

 

 

 

 

II

4.88 (1.14–20.87)

0.033

4.70 (1.10–20.12)

0.04

III

7.76 (1.76–30.8)

0.006

6.90 (1.64–29.03)

0.008

IV

15.01 (3.02–74.52)

0.001

12.40 (2.45–62.76)

0.002

Emergency

1.84 (1.002–3.36)

0.049

-

 

Colonic site

1.48 (0.79–2.79)

0.23

-

 

High grade

2.11 (1.17–3.79)

0.013

-

 

Lymph vascular invasion

1.79 (1.06–3.02)

0.03

-

 

Perineural invasion

2.41 (1.32–4.12)

0.004

1.82 (0.98–3.39)

0.06 

Recurrences were detected as a result of CEA testing in 24 patients (40.7%), clinical symptoms or signs in 16 patients (27.1%) and routine imaging in 19 patients (32.2%).

Discussion

In this study we found a recurrence rate of 24.9% at five years, with a median time to recurrence of 14 months and mortality rate of 31.4%; these findings are comparable to the modern international literature.6–18 Recurrence most commonly occurred in the liver, followed by lung, with a local recurrence rate of 6.3%. Recurrence was strongly linked to disease stage.

Twenty-five out of 59 (42.4%) recurrences were observed in the first year following surgery and 43/59 (72.9%) within the first two years, with annual recurrence rates decreasing sharply thereafter. That the majority of recurrences occur within the first two years post-operatively is reflected in several guidelines, which recommend increased frequency of follow-up clinic appointments within the first two to three years, such as those of the American Cancer Society, National Comprehensive Cancer Network and European Society for Medical Oncology.19–21 Studies with longer formal follow-up programmes have demonstrated low recurrence rates beyond five years also and this includes Seo et al’s study of 4,023 patients that revealed 36 (0.9%) recurrences occurred beyond five years.17

Our results are consistent with others in demonstrating the liver as the most common site of disease recurrence for rectal and colonic primaries combined.9,11,12,14,22,23 Recurrences in the lung were relatively more common for rectal primaries compared with colonic, occurring in 10.2% and 5.6% respectively, although this difference was not statistically significant (p=0.23). This pattern is frequently attributed to systemic venous drainage of rectal cancers via the pelvic veins.24 Although much less common, we found other sites of disease recurrence to include bone, brain, distant lymph nodes, peritoneum and ovary. While we did not observe any recurrences in the spleen, this has also been reported previously, albeit infrequently.25

We found recurrence rates to range from 5.1% to 60% in stages I to IV and multivariate analysis identified disease stage as a strong independent predictor of recurrence, with a greater than four-fold increase for stage II, six-fold for stage III and 12-fold for stage IV when compared to stage I (Table 2). Pathological features including histological grade, lymphovascular invasion and perineural invasion were not identified as independent predictors of time-to-recurrence in this study, perhaps due to a relatively small sample, as these tumour characteristics have been shown elsewhere to independently predict poor prognosis and are considered high-risk features of stage II CRC when selecting patients for adjuvant chemotherapy.19–21

Routine post-operative surveillance is carried out with the aim of earlier detection of disease recurrence and with it to maximise the chance of curative treatment and improve survival. While it was beyond the remit of this study to evaluate the treatment and survival in patients subsequent to the development of recurrence,five-year survival rates of 40% have been reported following surgery for metastatic recurrence involving the liver with similar outcomes being reported for the lung.26 In our series, CEA measurement lead to more detection than CT surveillance and overall 73% of recurrences were detected by these surveillance investigations.

The all-cause mortality rate in this cohort of 31.4% compares favourably to rates of approximately 40% reported historically in New Zealand and suggests a continuing trend of improving outcome.3,18,27 Data from the SEER database reveals a five-year mortality of 35.1%, although this is not limited to patients undergoing surgery with curative intent.28

The proportion of patients receiving adjuvant therapy in this cohort, 32.5%, falls short of the 40% reported by Buchwald et al, who have demonstrated a steady increase in the use of adjuvant therapy at their tertiary centre between 1993 to 2009.29 The proportion of stage III colon cancer patients receiving adjuvant chemotherapy was 58.4% in this cohort, which is very similar to 59% reported in the PIPER project report. Likewise, 34.7% of patients with non-metastatic rectal cancer in this cohort compared with 36% in the PIPER project report received adjuvant chemotherapy.30 For rectal cancer, 66.1% of patients received some form of neo-adjuvant therapy, which falls between the national average of 52% reported in the PIPER project report and 82% achieved in the 2009 cohort reported by Buchwald et al.29,30 While the figures presented in this cohort are in concordance with the national average, there is still room for improvement in this regard.

We report a very low rate of local recurrence following rectal cancer surgery of 1.7%, which compares well against modern international studies, where rates of 2.4–10% are frequently reported.15 Rectal cancer surgery at the unit is performed exclusively by surgeons with subspecialist training and may go towards explaining the low local recurrence rates and indeed the one case of local recurrence was in a patient that needed an APR and declined surgery for 15 months. This also reflects a large improvement from rates in excess of 20% reported in the literature prior to the introduction of total mesorectal excision, adjuvant therapies, multidisciplinary team discussion and surgical subspecialisation.27,31

The authors acknowledge that the current study has limitations. An assessment of compliance with post-operative follow-up protocol, including completion of follow-up, would be useful in order to evaluate the risk of bias, which is inherent in retrospective studies. Additionally, the outcomes presented here are those of a single hospital and may not be representative of New Zealand in general and, as shown in the PIPER project report, variations in practice and outcomes do exist across the country.

The outcomes reported in this study are encouraging and may reflect advances in colorectal cancer care, including the multidisciplinary team model, increasing use of adjuvant therapy, surgical specialisation and protocolled surveillance and follow-up.

Conclusion

While New Zealand outcomes in colorectal cancer have historically compared unfavourably against international standards, the outcomes observed in this cohort reflect improvement. However, ongoing quality improvement to further reduce mortality and recurrence remains an important health priority for this group of patients.

Summary

In this article, the authors describe the patterns of disease recurrence in patients having curative surgery for colorectal cancer at a hospital in New Zealand over a three-year period. With five years of follow-up data, one in four patients experienced the cancer returning and this is in keeping with data reported in the worldwide literature. Recurrence most commonly occurred within the first two years, and the liver and lungs were the most common sites of recurrence.

Abstract

Aim

To describe the patterns of recurrence in a contemporaneous cohort of patients undergoing surgery with curative intent for colorectal adenocarcinoma at a New Zealand hospital with five-year follow-up.

Method

Patients with colorectal cancer undergoing potentially curative surgery between January 2010 and December 2012 were followed up for a median of 61 months with three-monthly CEA (carcinoembryonic antigen), a colonoscopy after one year and yearly computed tomography scans of the chest, abdomen and pelvis for the first three years.

Results

Overall, 59/237 (24.9%) of patients experienced disease recurrence, the most common sites being the liver, followed by the lung and local recurrence. Recurrence rates did not differ significantly between colon and rectal cancer and ranged from 5.1% in stage I to 60% in stage IV. Seventy-three percent of all recurrences were observed within the first 24 months post-operatively.

Conclusion

While New Zealand outcomes in colorectal cancer have historically compared unfavourably against international standards, the outcomes observed in this cohort are encouraging and may reflect advances in care, including multidisciplinary team discussion, increased use of adjuvant therapy, surgical subspecialisation and protocolled surveillance and follow-up.

Author Information

Ashok Gunawardene, Department of Surgery & Anaesthesia, University of Otago, Wellington;
Brendan Desmond, Department of General Surgery, Capital & Coast DHB, Wellington;
Ali Shekouh, Department of General Surgery, Capital & Coast DHB, Wellington;
Peter Larsen, Surgery & Anaesthesia, University of Otago, Wellington;
Elizabeth Dennett, Department of Surgery & Anaesthesia, University of Otago; Capital & Coast DHB, Wellington.

Correspondence

Dr Ashok Gunawardene, Department of Surgery & Anaesthesia, University of Otago, 23A Mein Street, Wellington 6242.

Correspondence Email

ashok.gunawardene@otago.ac.nz

Competing Interests

Dr Gunawardene reports grants from Wellington Surgical Research Trust, Phil & Teds, outside the submitted work. Dr Larsen reports grants from Wellington Medical Research Foundation, Lotteries Health Research, University of Otago Research Grant, outside the submitted work.

References

  1. Ministry of Health. 2015. Cancer: New registrations and deaths 2012. Wellington: Ministry of Health.
  2. Ferlay J, Soerjomataram I, Dikshit R, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. International journal of cancer. 2015; 136(5):E359–E86.
  3. Aye PS, Elwood JM, Stevanovic V. Comparison of cancer survival in New Zealand and Australia, 2006–2010. N Z Med J. 2014; 127(1407):14–26.
  4. Alafeishat L, Elwood M, Ioannides S. Cancer mortality and incidence trends comparing New Zealand and Australia for the period 2000–2007. The New Zealand Medical Journal. 2014; 127(1400):9.
  5. Compton C, Fenoglio-Preiser CM, Pettigrew N, Fielding LP. American Joint Committee on Cancer prognostic factors consensus conference. Cancer. 2000; 88(7):1739–57.
  6. Dent OF, Newland RC, Chan C, et al. Trends in pathology and long-term outcomes after resection of colorectal cancer: 1971–2013. ANZ Journal of Surgery. 2017; 87(1-2):34–8.
  7. Sørensen CG, Karlsson WK, Pommergaard H-C, et al. The diagnostic accuracy of carcinoembryonic antigen to detect colorectal cancer recurrence–A systematic review. International Journal of Surgery. 2016; 25:134–44.
  8. Kim HS, Lee MR. Diagnostic accuracy of elevated serum carcinoembryonic antigen for recurrence in postoperative stage II colorectal cancer patients: comparison with stage III. Annals of coloproctology. 2013; 29(4):155–9.
  9. Fora A, Patta A, Attwood K, et al. Intensive radiographic and biomarker surveillance in stage II and III colorectal cancer. Oncology. 2012; 82(1):41–7.
  10. Su B-B, Shi H, Wan J. Role of serum carcinoembryonic antigen in the detection of colorectal cancer before and after surgical resection. World J Gastroenterol. 2012; 18(17):2121–6.
  11. Banaszkiewicz Z, Jarmocik P, Frasz J, et al. Usefulness of CEA concentration measurement and classic colonoscopy in follow-up after radical treatment of colorectal cancer. Polish Journal of Surgery. 2011; 83(6):310–8.
  12. Chen C-H, Hsieh M-C, Lai C-C, et al. Lead time of carcinoembryonic antigen elevation in the postoperative follow-up of colorectal cancer did not affect the survival rate after recurrence. International journal of colorectal disease. 2010; 25(5):567–71.
  13. Yakabe T, Nakafusa Y, Sumi K, et al. Clinical significance of CEA and CA19-9 in postoperative follow-up of colorectal cancer. Annals of surgical oncology. 2010; 17(9):2349–56.
  14. Park IJ, Choi G-S, Lim KH, et al. Serum carcinoembryonic antigen monitoring after curative resection for colorectal cancer: clinical significance of the preoperative level. Annals of surgical oncology. 2009; 16(11):3087–93.
  15. Räsänen M, Carpelan-Holmström M, Mustonen H, et al. Pattern of rectal cancer recurrence after curative surgery. International journal of colorectal disease. 2015; 30(6):775–85.
  16. Seo SI, Lim SB, Yoon YS, et al. Comparison of recurrence patterns between ≤5 years and >5 years after curative operations in colorectal cancer patients. Journal of surgical oncology. 2013; 108(1):9–13.
  17. Cho YB, Chun H-K, Yun HR, et al. Clinical and pathologic evaluation of patients with recurrence of colorectal cancer five or more years after curative resection. Diseases of the colon & rectum. 2007; 50(8):1204–10.
  18. Keating J, Yong D, Cutler G, Johnston J. Multidisciplinary treatment of colorectal cancer in New Zealand: survival rates from 1997–2002. The New Zealand Medical Journal. 2006; 119(1242).
  19. Meyerhardt JA, Mangu PB, Flynn PJ, et al. Follow-up care, surveillance protocol, and secondary prevention measures for survivors of colorectal cancer: American Society of Clinical Oncology clinical practice guideline endorsement. Journal of Clinical Oncology. 2013; 31(35):4465–70.
  20. Labianca R, Nordlinger B, Beretta G, et al. Early colon cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Annals of oncology. 2013; 24(suppl_6):vi64-vi72.
  21. El-Shami K, Oeffinger KC, Erb NL, et al. American Cancer Society colorectal cancer survivorship care guidelines. CA: a cancer journal for clinicians. 2015; 65(6):427–55.
  22. Shibutani M, Maeda K, Nagahara H, et al. Significance of CEA and CA19-9 combination as a prognostic indicator and for recurrence monitoring in patients with stage II colorectal cancer. Anticancer research. 2014; 34(7):3753–8.
  23. Hara M, Kanemitsu Y, Hirai T, et al. Negative serum carcinoembryonic antigen has insufficient accuracy for excluding recurrence from patients with Dukes C colorectal cancer: analysis with likelihood ratio and posttest probability in a follow-up study. Diseases of the colon & rectum. 2008; 51(11):1675.
  24. Pugh SA, Shinkins B, Fuller A, et al. Site and stage of colorectal cancer influence the likelihood and distribution of disease recurrence and postrecurrence survival: data from the FACS randomized controlled trial. Annals of surgery. 2016; 263(6):1143–7.
  25. Carriquiry LA, Piñeyro A. Should carcinoembryonic antigen be used in the management of patients with colorectal cancer? Diseases of the colon & rectum. 1999; 42(7):921–9.
  26. Primrose JN, Perera R, Gray A, et al. Effect of 3 to 5 years of scheduled CEA and CT follow-up to detect recurrence of colorectal cancer: the FACS randomized clinical trial. Jama. 2014; 311(3):263–70.
  27. Frizelle F, Emanuel J, Keating J, Dobbs B. A multicentre retrospective audit of outcome of patients undergoing curative resection for rectal cancer. The New Zealand Medical Journal (Online). 2002; 115(1156).
  28. Siegel RL, Miller KD, Fedewa SA, et al. Colorectal cancer statistics, 2017. CA: a cancer journal for clinicians. 2017; 67(3):177–93.
  29. Buchwald P, Hall C, Davidson C, et al. Improved survival for rectal cancer compared to colon cancer: the four cohort study. ANZ journal of surgery. 2016.
  30. The PIPER Project: An Internal Examination of Colorectal Cancer Management in New Zealand : Jackson C, Firth M, Hinder V, et al. http://www.fmhs.auckland.ac.nz/assets/fmhs/sms/ctnz/docs/THE%20PIPER%20PROJECT%20Final%20deliverable%20report%207%20August%202015%20(HRC%2011_764%20FINDLAY).pdf
  31. Obrand DI, Gordon PH. Incidence and patterns of recurrence following curative resection for colorectal carcinoma. Diseases of the colon & rectum. 1997; 40(1):15–24.