- Email: [email protected]
The management of malignant polyps in colorectal cancer screening programmes: A retrospective Italian multi-centre study
G Model YDLD-2878; No. of Pages 5
ARTICLE IN PRESS Digestive and Liver Disease xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Digestive and Liver Disease journal homepage: www.elsevier.com/locate/dld
Oncology
The management of malignant polyps in colorectal cancer screening programmes: A retrospective Italian multi-centre study Renato Fasoli a,∗ , Richard Nienstedt a , Nicola De Carli b , Fabio Monica c , Ennio Guido d , Flavio Valiante e , Franco Armelao a , Giovanni de Pretis a,f a
Provincial Department of Gastroenterology and Digestive Endoscopy, Santa Chiara Hospital, Trento, Italy Histo-pathology Unit, Santa Chiara Hospital, Trento, Italy c Gastroenterology and Digestive Endoscopy Unit, Bassano del Grappa Hospital, Bassano del Grappa, VI, Italy d Gastroenterology and Digestive Endoscopy Unit, Sant’Antonio Hospital, Padova, Italy e Gastroenterology and Digestive Endoscopy Unit, Santa Maria del Prato Hospital, Feltre, BL, Italy f Provincial Department of Gastroenterology and Digestive Endoscopy, Santa Maria del Carmine Hospital, Rovereto, TN, Italy b
a r t i c l e
i n f o
Article history: Received 2 December 2014 Accepted 18 April 2015 Available online xxx Keywords: Colorectal cancer screening Colorectal malignant polyps Histological re-evaluation Radicalisation surgery
a b s t r a c t Background: Although recognition of colorectal malignant polyps is increasing, treatment plans lack the evidence of randomised trials. Aim: To retrospectively evaluate presentation, management and outcomes of screen-detected colorectal malignant polyps, with special focus on the role of histological factors in therapeutic decision-making. Methods: We retrospectively analysed data regarding malignant polyps detected during faecal immunochemical test-based screening programmes in five centres in North-Eastern Italy between April 2008 and April 2013. Results: 306 malignant polyps in 306 patients were included; 72 patients underwent surgery directly (23.6%). Of 234 patients treated endoscopically, 133 subsequently underwent radicalisation surgery (56.8%) and in 17 there was evidence of residual disease (12.8%). Involved, unsafe (<1 mm) or invaluable resection margins and sessile morphology represented the most frequent determinants of subsequent surgery. The mean number of nodes harvested during radicalisation surgery was 7.1 ± 6.4 (range 0–29). Histological diagnosis was re-evaluated according to new guidelines in 125 cases (41%); in 18 this led to modification of the risk class (14.4%). Conclusions: Although the rate of surgical treatment following endoscopic resection is similar to other studies, residual disease at surgery was lower than most international series. Adhering to the new histological reporting system and respecting guidelines on node harvesting may favourably influence prognosis. © 2015 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.
1. Introduction Colorectal malignant polyps (T1 tumour, according to TNM classification) are cancerous polyps that have invaded the sub-mucosal layer. The diffusion of bowel cancer screening programmes has determined the increased detection of these less advanced forms of the tumour. Therefore, there is growing awareness by both clinicians and pathologists regarding the diagnosis and management.
∗ Corresponding author at: Endoscopy Unit, Imperia Hospital, Via Sant’Agata 57, 18100 Imperia, Italy. Tel.: +39 340 4662570; fax: +39 0183 537496. E-mail address: [email protected] (R. Fasoli).
Traditionally, malignant polyps have been considered low risk and therefore safely treatable by means of endoscopic resection alone if: (a) the lesions were not poorly differentiated; (b) there was no vascular or lymphatic invasion; (c) the margins of excision were not involved [1,2]; (d) an en-bloc polypectomy was performed [3]. More recently, a European panel has issued new guidelines in colorectal cancer (CRC) screening and diagnosis. Regarding the histological judgement of malignant polyps, new features such as budding and micro-staging have acquired significant weight [4]. Up to now, no controlled trial has compared endoscopic treatment alone with endoscopic + surgical treatment in the management of these lesions. In this study, we report the results of a multi-centre study of malignant polyps diagnosed during CRC screening programmes.
http://dx.doi.org/10.1016/j.dld.2015.04.011 1590-8658/© 2015 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Fasoli R, et al. The management of malignant polyps in colorectal cancer screening programmes: A retrospective Italian multi-centre study. Dig Liver Dis (2015), http://dx.doi.org/10.1016/j.dld.2015.04.011
G Model
ARTICLE IN PRESS
YDLD-2878; No. of Pages 5
R. Fasoli et al. / Digestive and Liver Disease xxx (2015) xxx–xxx
2
Presentation, endoscopic and surgical management and follow-up are analysed. A special focus has been given to residual disease at surgery following polypectomy and to the role of histological factors in determining decision-making, including a re-evaluation of endoscopic samples by dedicated pathologists. 2. Materials and methods We retrospectively evaluated data on all consecutive colorectal malignant polyps detected during screening programmes between April 2008 and April 2013 in five centres of North-Eastern Italy: Trento, Rovereto (Trento), Padova (Ospedale Sant’Antonio), Bassano (Vicenza) and Feltre (Belluno). All involved screening programmes following the protocol of total colonoscopy in subjects of both sexes, aged 50–69 years, who tested positive for faecal immunochemical occult blood test (FIT). A standardised case record form was sent to all participating centres. Demographic and baseline clinical features, endoscopic characteristics of the lesions, original histological diagnosis, management and follow-up have all been collected. The following comorbidities were considered relevant: diabetes mellitus, significant cardiovascular and respiratory conditions, chronic kidney disease, chronic liver disease, present or previous malignant tumours, and degenerative neurological conditions. Residual disease was defined as the presence of any residual carcinoma at the site of polypectomy or in the nodes, or both. Recurrence of disease was defined as the presence of locally or regionally recurrent disease or distant metastases after a curative resection. A minimum follow-up of 18 months was required for all patients, to achieve a valuable judgement of recurrent disease. Longer follow-up periods, when available, allowed an evaluation of surveillance recommendations. Single centres have followed local clinical protocols and no “a priori” attempt to homogenise clinical behaviour has been made. According to the timing of diagnosis, classification into low- and high-risk lesions has changed from the traditional approach [1,2] to a more modern one based on new guidelines [4]; consequently, the histological reporting form of endoscopically resected lesions has changed. For this reason specialised pathologists re-evaluated the cases previously enrolled.
2.1. Statistical analysis Categorical variables were reported as numbers and summarised as proportions; continuous variables were expressed as means ± standard deviation (SD). Statistical comparisons were performed using Student’s t test (comparisons between two groups) or one-way ANOVA (comparisons amongst three groups) for continuous variables (age, size of the lesions, number of nodes resected). Chi-squared test was used for categorical variables (gender, site distribution and morphology of the polyps, histological items involved in therapeutic judgement, rates of direct and radicalisation surgery). A p value of less than 0.05 was considered to represent a statistically significant difference. 2.2. Ethical approval A recently published Code of Clinical Research (National Code on Clinical Researches) [5] declared that retrospective archive studies do not need ethics approval. According to the Declaration of Helsinki, every precaution was taken to protect the privacy of the subjects involved and the confidentiality of their personal information. 3. Results In the five-year period under consideration, there were a total of 14,943 FIT-positive subjects. Of them, 13,588 underwent colonoscopy (adhesion rate: 90.9%). A total number of 306 malignant polyps in 306 patients were observed. One hundred and ninety-three patients (63%) were male. The mean age was 63.1 ± 6.2 years (range 50–69). Seventy-two cases were referred for direct surgical intervention (23.6%). The indications for direct surgery were: size of the lesion in 26 (36%), depressed morphology in 23 (32%), no-lifting sign in 6 (8%), and technical problems during resection in 6 (8%). Information is missing in 12 cases (16%). Two hundred and thirty-four cases were resected endoscopically (76.4%) and 133 of these were referred for radicalisation surgery (56.8%). Table 1 reports the demographic, clinical and endoscopic features of patients undergoing direct surgery, endoscopic resection
Table 1 Baseline characteristics of malignant polyps. Endoscopic resection only n = 101
Radicalisation surgery n = 133
Direct surgery n = 72
Total n = 306 197 (63%) 109 (37%)
p value
Gender Males Females
68 (67%) 33 (33%)
83 (62%) 50 (48%)
46 (64%) 26 (36%)
Mean age (±SD# )
63.1 (±6.9)
63.0 (±5.7)
63.1 (±5.3)
Comorbidities 0 1 ≥2 Missing data
76 (75%) 12 (12%) 5 (5%) 8 (8%)
100 (75%) 19 (14%) 4 (3%) 10 (8%)
53 (74%) 9 (12%) 3 (4%) 7 (10%)
230 (75%) 40 (13%) 12 (4%) 24 (8%)
0.979a
Site Right colon Left colon Rectum
10 (10%) 58 (57%) 33 (33%)
20 (15%) 83 (62%) 30 (23%)
30 (42%) 31 (43%) 11 (15%)
60 (20%) 172 (56%) 74 (24%)
<0.001a
Mean size
15.8 mm (±8.0)
19.2 mm (±9.0)
37.5 mm (±17.9)
Morphology Sessile/Flat Pedunculated
30 (30%) 71 (70%)
97 (73%) 36 (27%)
66 (92%) 6 (37%)
a b
63.1 (±6.2)
22.4 mm (±10.5) 193 (63%) 113 (37%)
0.735a 0.987b
<0.001b
<0.001a
Chi-squared. One way ANOVA.
Please cite this article in press as: Fasoli R, et al. The management of malignant polyps in colorectal cancer screening programmes: A retrospective Italian multi-centre study. Dig Liver Dis (2015), http://dx.doi.org/10.1016/j.dld.2015.04.011
G Model
ARTICLE IN PRESS
YDLD-2878; No. of Pages 5
R. Fasoli et al. / Digestive and Liver Disease xxx (2015) xxx–xxx
3
Table 2 Differences amongst centres in direct surgery rates.
Trento Rovereto Bassano Feltre Padova a
Endoscopic resection n= 234
Direct surgery n = 72
Total n = 306
71 (75%) 30 (83%) 38 (73%) 46 (90%) 49 (70%)
24 (25%) 8 (21%) 14 (27%) 5 (10%) 21 (30%)
95 38 52 51 70
p valuea
0.108
Chi-squared test.
Table 3 Differences amongst centres in post-polypectomy radicalisation surgery rates.
Trento Rovereto Bassano Feltre Padova a
Endoscopic resection only n = 101 (43%)
Radicalisation surgery n = 133 (57%)
Total n = 234
21 (30%) 11 (37%) 12 (32%) 31 (67%) 26 (53%)
50 (70%) 19 (63%) 26 (68%) 15 (33%) 23 (47%)
71 30 38 46 49
p valuea
0.00033
Chi-squared test.
followed by surgical radicalisation and endoscopic resection only respectively. No significant differences were observed for age, gender and comorbidities. In patients treated with direct surgery, lesion size was significantly larger; right colon localisation and sessile/flat morphology occurred with significantly higher frequency. Conversely, in patients treated only endoscopically, rectal localisation and pedunculated morphology were significantly more frequent. No significant difference amongst centres was observed in the rates of direct surgery (Table 2); conversely, a wide variability amongst centres has been observed in the rates of radicalisation surgery (Table 3). Table 4 shows the different histological items taken into consideration in the therapeutic assessment. Involved, unsafe (<1 mm) or invaluable margins of resection occurred in 88% of cases referred for radicalisation surgery. Sessile morphology was also strikingly more common in the radicalised group than in the endoscopically treated one (p < 0.005). In three patients with high-risk histological features after endoscopic resection, severe comorbidities were contraindications for surgery. 3.1. Lymph node harvesting During direct surgical intervention, a mean 9.6 ± 5.4 nodes were resected for histological examination (range 3–42). In 42 cases fewer than 12 nodes were harvested (58%). Data were not available in 9 cases (8%).
Table 4 Histological items considered for therapeutic assessment. Histological item
Radicalisation surgery N = 133
Endoscopic resection only N = 101
Piece-meal resection Incomplete resection Involved margins Invaluable margins Unsafe margins (<1 mm) Lympho-vascular invasion Depth of submucosal invasion Tumour grading Budding Sessile/flat morphology
39 (29%) 4 (3%) 57 (43%) 29 (22%) 29 (22%) 23 (17%) 47 (35%) 15 (11%) 61 (46%) 97 (73%)
14 (14%) 0 (0%) 8 (8%) 3 (3%) 11 (11%) 0 (0%) 5 (5%) 0 (0%) 35 (35%) 30 (30%)
During post-polypectomy surgical intervention, a mean 7.1 ± 5.8 nodes were resected (range 0–29). In 90 cases fewer than 12 lymph nodes were retrieved (68%). Data were not available in 10 cases (13%). The mean number of lymph nodes removed in patients who underwent direct surgery was significantly higher than in those who underwent post-polypectomy surgery (p = 0.027). 3.2. Residual disease Seventeen out of 133 patients sent to radicalisation surgery had evidence of residual disease (12.8%). Their TNM classification was as follows: T2N0: 6 cases; T2N1: 9 cases; T3N0: 2 cases. In this subgroup, the most frequently observed features were: involved/unsafe margins (n = 4), high-grade budding (n = 3), depth of sub-mucosal invasion (n = 3), lymphatic or vascular invasion (n = 2), tumour grading (n = 1), piece-meal resection (n = 1). 3.3. Follow-up Data on subsequent follow-up were available for 73 out of 101 subjects considered at low risk and treated only by endoscopic resection (72.3%). Pedunculated lesions accounted for 70% of the total, whereas 30% were sessile and flat. The mean follow-up was 44 months (range 24–54). In 90% of patients with sessile/flat lesions the protocol included an early follow-up (3–6 months), followed by a colonoscopy at 1–3 years. In pedunculated lesions a follow-up colonoscopy at 3 years was suggested in 52% of patients; 38% were seen at 1 year and 10% at 2 years. Only 37% of patients received radiological surveillance (ultrasound or computerised tomography). During the 18-month follow-up, no recurrence of disease was diagnosed in the endoscopically treated population; 3 recurrences (2 cases of distant metastases and 1 case of lymph-nodal involvement) were observed in the high-risk population referred for radicalisation surgery. 3.4. Histological re-evaluation A second opinion of histological samples was acquired in 125 cases (53% of endoscopically resected lesions). The re-evaluation led to a modification of the risk class in 18 cases (14.4%).
Please cite this article in press as: Fasoli R, et al. The management of malignant polyps in colorectal cancer screening programmes: A retrospective Italian multi-centre study. Dig Liver Dis (2015), http://dx.doi.org/10.1016/j.dld.2015.04.011
G Model YDLD-2878; No. of Pages 5
ARTICLE IN PRESS R. Fasoli et al. / Digestive and Liver Disease xxx (2015) xxx–xxx
4
Nine patients were downgraded from a high- to a low-risk class, whereas 9 patients were upgraded. The histological features considered in downgrading were: depth of invasion (n = 6), absence of vascular invasion (n = 2), negative margins (n = 1); the histological features considered in upgrading were: budding (n = 7), vertical margin involvement (n = 1), extensive sub-mucosal invasion (n = 1). Regarding therapeutic decisions, 8 patients would have had a different indication (6.4%): 5 patients resulted over-treated, and 3 under-treated.
4. Discussion No controlled trial is available that could shed light on the treatment of colorectal malignant polyps, in particular on the indication for surgical radicalisation after endoscopic removal. The most numerous contribution about this issue published so far comes from the study by Cooper et al. [6], who in 2012 reported a population-based analysis of colonoscopic polypectomy compared with surgery in 2077 patients observed in 2002–2005. Outcomes were similar in both groups, although a significant selection bias in the allocation of the treatment was present and few data were available regarding prognostic features other than the degree of differentiation of the cancer. A similar study was published in 2012, reflecting the current management of these lesions in a general setting in a few centres scattered throughout Northern England and offers interesting data regarding different issues [7]. A recent position statement of the Association of Coloproctology of Great Britain and Ireland (ACPGBI) underlines how it is more advisable and ethically appropriate to create a malignant polyp registry rather than embarking on a randomised controlled trial [8]. According to these authors, all clinicians should scrutinise regional non-randomised data carefully, particularly with the increasing detection of these malignant polyps via the national screening programmes. Careful audits of specific risk factors are recommended, to better define indications for surgical radicalisation. In light of these premises, we were stimulated to carry out a regional study on this type of lesion diagnosed within screening campaigns. To our knowledge, the present study is the first to specifically take into consideration the management of malignant polyps in a FIT-positive screening population. With regard to the site of malignant polyps in our series, our result (56%) of left colon localisation is in keeping with data published by Wasif et al. [9], whereas we had a slightly higher right colon localisation than that reported in the meta-analysis by Hassan et al. [10]. In our series 23.6% of subjects received an immediate surgical intervention (with no attempt at endoscopic removal). This figure is far below that reported in the series by Gill et al. [7], where 42% of subjects received biopsies and subsequent surgery. The difference could be due to a more aggressive endoscopic approach towards this kind of lesion observed in our population. Our indications for direct surgery are in keeping with contraindications to endoscopic resection in so-called difficult colorectal polyps (size, macroscopic morphology, no-lifting sign, and technical difficulties) [11]. Considering the indications for radicalisation surgery, several factors must be taken into account, mainly the risk of lymph-node metastases, surgical morbidity and mortality, and patients’ comorbidities.
The influence of histopathological factors on the risk of nodal metastases in early colorectal cancer has been taken in consideration in two recent meta-analyses that came to similar conclusions. A submucosal invasion deeper than 1 mm, lympho-vascular invasion, poor differentiation and tumour budding were significantly associated with lymph-node metastases [12,13]. Furthermore, a previous meta-analysis [10] showed that positive resection margins, poor differentiation of carcinoma, and vascular invasion were significantly associated with more general unfavourable clinical outcomes (residual disease, recurrent disease, lymph node metastases, haematogenous metastases, mortality). British authors have even created a score to predict the risk of residual disease following resection of a malignant polyp [8]. Overall, the frequency of surgical radicalisation reported in our study is in keeping with figures reported in the international literature [6,7]. Sessile morphology and the conditions of margins of resection of the endoscopically removed tissue appear to have played major roles in our decision-making behaviour. In this regard, as the lack of margin evaluation (mainly related to specimen fragmentation) has proved to be a frequent cause of attribution to high risk class, resection/handling of the lesion should be a major concern for endoscopists. Unlike available series recently published in the literature [6,14,15], lympho-vascular invasion, tumour grade and high-grade budding have only marginally influenced treatment plans. The study by Cooper et al. [6] has shown – in a general population setting – that comorbidities and advanced age play important roles in decision-making. Our study concerns a screening population with a relatively young age and low rates of relevant comorbidities; furthermore, counselling before colonoscopy often leads to the exclusion of severely ill and frail patients. For these reasons, histological/morphological factors have been the main determinants in the majority of cases. The incidence of residual disease in the subgroup of patients undergoing surgical intervention after endoscopic resection shows some degree of variability in the literature, both at the bowel wall and at the lymph nodes. Reported rates range between 15% and 27% [7,14,16,17]. In our experience, such a rate is slightly lower than reported in the above-mentioned series collected in the general population [7,14,16,17]. Considering that we are dealing with a screening population, this difference may hypothetically account for a less severe form of the disease in this subgroup of patients. As regards the pattern of histological parameters considered crucial in the decision on surgical radicalisation, particular value has been given to the state of tumour margins. This type of behaviour is in keeping with the quoted British score (which attributes 4 points of severity to margins lower than 1 mm) [8] and with a North American retrospective study [18]. Naqvi et al. came to different conclusions and suggested that clear resection margins, even those with less than 1 mm of clearance, might be treated safely only with surveillance [19]. Unfortunately, the paucity of patients with evidence of residual disease at surgery in our series allows neither predictors of residual disease at surgery to be analysed nor wall and lymph node involvement to be differentiated. Benhaim et al. compared lymph-node harvest from 22 patients undergoing radicalisation resection after polypectomy of a malignant lesion, with 509 patients undergoing direct colectomy for invasive CRC. Significantly fewer lymph nodes were retrieved in the former group (11.63 vs 26.3 respectively) [20]. A study by Wasif et al. came to similar and even more relevant conclusions. They described – in a general population setting – a
Please cite this article in press as: Fasoli R, et al. The management of malignant polyps in colorectal cancer screening programmes: A retrospective Italian multi-centre study. Dig Liver Dis (2015), http://dx.doi.org/10.1016/j.dld.2015.04.011
G Model YDLD-2878; No. of Pages 5
ARTICLE IN PRESS R. Fasoli et al. / Digestive and Liver Disease xxx (2015) xxx–xxx
median removal of just one node. This type of behaviour has prognostic consequences, as patients with fewer resected lymphnodes have been shown to have a survival disadvantage [9]. In our series, the mean number of resected lymph nodes at surgery is below the threshold indicated by recent guide-lines, which suggest a minimum number of 12 [21]. This occurred both in direct surgery (mean 9.6) and – more strikingly – in postpolypectomy surgery (mean 7.1). A possible explanation for this result is a higher confidence within the surgical team regarding the prognosis of patients referred to surgery after endoscopic polypectomy. Our data on follow-up in patients who only received endoscopic treatment, although limited to 18 months, confirm the favourable outcome of this population, with no recurrences diagnosed in the period considered, in keeping with the British study [7]. Proposed protocols in sessile/flat lesions comply satisfactorily with international guidelines. In pedunculated lesions the quality of surveillance intervals can be improved, with the aim of prescribing with higher frequency a 3-year endoscopic follow-up. In fact, it has been shown that the short-term outcomes of these lesions are similar to those of non-malignant pedunculated polyps [22]. Considering behaviour in planning radiological follow-up after endoscopic polypectomy, British and North American authors have different opinions: the former are prone to offer imaging follow-up based on the rate of distant metastases; the latter are less inclined to do so. In our experience radiological follow-up was offered to 37% of this subgroup of patients. As regards the subgroup referred for surgical intervention, either directly or after endoscopic resection, follow-up recommendations comply in general with international guidelines following surgery for CRC of any grade. Budding and depth of submucosal invasion represent the most frequently quoted items at the re-evaluation of histological samples. While budding is a novel parameter with high sensitivity and acceptable specificity with regard to lymph-node metastases [4], the degree of sub-mucosal infiltration has been identified very early as a risk factor for advanced disease. The varying judgement concerning degree of sub-mucosal infiltration reflects the importance of a specialised pathologist [23]. Approximately 15% of re-evaluated patients changed their risk class; in almost half of them this would have led to a different treatment. These data, taken together, underline the importance of having clinical/pathological audits in this setting. The main limitations of the study are its retrospective nature, the suboptimal collection of data regarding some items and a thorough follow-up limited to 18 months. Although our rate of surgical treatment of colorectal malignant polyps following endoscopic resection complies with that of other series carried out in general populations, the rate of residual disease at surgery lies below the majority of international experiences. Therefore, bearing in mind the potential risk of over-treatment in these patients, we believe that it is mandatory to have a rigorous multi-disciplinary discussion of treatment options in each case, as well as considering the patient’s point of view and preference. Furthermore, adhering to a new histological reporting system and respecting guidelines on lymph node harvesting could favourably influence the patient’s prognosis. The rate of endoscopically treated patients offered an imaging follow-up for the early detection of distant metastases turned out to be sub-optimal: we believe that local protocols ought to be planned
5
to stimulate the regular prescription of extra-colonic examinations and – possibly – homogenise the surveillance of these patients. Conflict of interest None declared. Funding Richard Nienstedt was holder of a two-year salary grant from Pezcoller Foundation (Trento). References [1] Bond JH. Polyp guideline: diagnosis, treatment and surveillance for patients with colorectal polyps. Practice Parameters Committee of the American College of Gastroenterology. American Journal of Gastroenterology 2000;95: 3053–63. [2] Muto T, Sawada T, Sugihara K. Treatment of carcinoma in adenomas. World Journal of Surgery 1991;15:35–40. [3] Ramirez M, Schiering S, Papaconstantinou HT, et al. Management of the malignant polyp. Clinics in Colon and Rectal Surgery 2008;21:286–90. [4] Quirke P, Risio M, Lambert R, et al. European guidelines for quality assurance in colorectal cancer screening and diagnosis. First Edition – quality assurance in pathology in colorectal cancer screening and diagnosis. Endoscopy 2012;44(Suppl. 3):SE116–30. [5] National Code on Clinical Researches published in Official Gazette numbered with 28030 at April 13, 2013. [6] Cooper GS, Xu F, Barnholtz Sloan JS, et al. Management of malignant colonic polyps: a population-based analysis of colonoscopic polypectomy versus surgery. Cancer 2012;118:651–9. [7] Gill MD, Rutter MD, Holtham SJ. Management and short term outcome of malignant colorectal polyps in the north of England. Colorectal Disease 2013;15:169–76. [8] Williams JG, Pullan RD, Hill J, et al. Management of the malignant colorectal polyp: ACPGBI position statement. Colorectal Disease 2013;15(Suppl. 2):1–38. [9] Wasif N, Etzioni D, Maggard MA, et al. Trends, patterns, and outcomes in the management of malignant colonic polyps in the general population of the United States. Cancer 2011;117:931–7. [10] Hassan C, Zullo A, Risio M, et al. Histologic risk factors and clinical outcome in colorectal malignant polyp: a pooled-data analysis. Diseases of the Colon & Rectum 2005;48:1588–96. [11] Riley SA. Colonscopic polypectomy and endoscopic mucosal resection: a practical guide; 2008. http://bsg.org.uk/pdf word docs/polypectomy 08.pdf [12] Beaton C, Twine CP, Williams GL, et al. Systematic review and meta-analysis of histopathological factors influencing the risk of lymph node metastases in early colorectal cancer. Colorectal Disease 2013;15:788–97. [13] Bosch SL, Teerenstra S, de Wilt JH, et al. Predicting lymph node metastasis in pT1 colorectal cancer: a systematic review of risk factors providing rationale for therapy decisions. Endoscopy 2013;45:827–34. [14] Ueno K, Mochizuki H, Hashiguchi Y, et al. Risk factors for an adverse outcome in early invasive colorectal carcinoma. Gastroenterology 2004;127:385–94. [15] Kim KM, Eo SJ, Shim SG, et al. Risk factors for residual cancer and lymph node metastasis after noncurative endoscopic resection of early colorectal cancer. Diseases of the Colon and Rectum 2013;56:35–42. [16] Volk EE, Goldblum JR, Petras RE, et al. Management and outcome of patients with invasive carcinoma arising in colorectal polyps. Gastroenterology 1995;109:1801–7. [17] Seitz U, Bohnacker S, Seewald S, et al. Is endoscopic polypectomy an adequate therapy for malignant colorectal adenomas? Presentation of 114 patients and review of the literature. Diseases of the Colon and Rectum 2004;47:1789–96. [18] Butte JM, Tang P, Gonen M, et al. Rate of residual disease after complete endoscopic resection of malignant colonic polyp. Diseases of the Colon and Rectum 2012;55:122–7. [19] Naqvi S, Burroughs S, Chave HS, et al. Management of colorectal polyp cancers. Annals of the Royal College of Surgeons of England 2012;94:574–8. [20] Benhaim L, Benoist S, Bachet JB, et al. Salvage colectomy for endoscopically removed malignant colon polyps: is it possible to determine the optimal number of lymph nodes that need to be harvested? Colorectal Disease 2012;14:79–88. [21] Desch CE, McNiff KK, Schneider EC, et al. American Society of Clinical Oncology/National Comprehensive Cancer Network Quality Measures. Journal of Clinical Oncology 2008;26:3631–7. [22] Freeman HJ. Long-term follow-up of patients with malignant pedunculated colon polyps after colonoscopic polypectomy. Canadian Journal of Gastroenterology 2013;27:20–4. [23] Bong-Hyeon Kye, Ji-Han Jung, Hyung-Jin Kim, et al. Tumor budding as a risk factor of lymph node metastasis in submucosal invasive T1 colorectal carcinoma: a retrospective study. BMC Surgery 2012;12:16.
Please cite this article in press as: Fasoli R, et al. The management of malignant polyps in colorectal cancer screening programmes: A retrospective Italian multi-centre study. Dig Liver Dis (2015), http://dx.doi.org/10.1016/j.dld.2015.04.011
ARTICLE IN PRESS Digestive and Liver Disease xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Digestive and Liver Disease journal homepage: www.elsevier.com/locate/dld
Oncology
The management of malignant polyps in colorectal cancer screening programmes: A retrospective Italian multi-centre study Renato Fasoli a,∗ , Richard Nienstedt a , Nicola De Carli b , Fabio Monica c , Ennio Guido d , Flavio Valiante e , Franco Armelao a , Giovanni de Pretis a,f a
Provincial Department of Gastroenterology and Digestive Endoscopy, Santa Chiara Hospital, Trento, Italy Histo-pathology Unit, Santa Chiara Hospital, Trento, Italy c Gastroenterology and Digestive Endoscopy Unit, Bassano del Grappa Hospital, Bassano del Grappa, VI, Italy d Gastroenterology and Digestive Endoscopy Unit, Sant’Antonio Hospital, Padova, Italy e Gastroenterology and Digestive Endoscopy Unit, Santa Maria del Prato Hospital, Feltre, BL, Italy f Provincial Department of Gastroenterology and Digestive Endoscopy, Santa Maria del Carmine Hospital, Rovereto, TN, Italy b
a r t i c l e
i n f o
Article history: Received 2 December 2014 Accepted 18 April 2015 Available online xxx Keywords: Colorectal cancer screening Colorectal malignant polyps Histological re-evaluation Radicalisation surgery
a b s t r a c t Background: Although recognition of colorectal malignant polyps is increasing, treatment plans lack the evidence of randomised trials. Aim: To retrospectively evaluate presentation, management and outcomes of screen-detected colorectal malignant polyps, with special focus on the role of histological factors in therapeutic decision-making. Methods: We retrospectively analysed data regarding malignant polyps detected during faecal immunochemical test-based screening programmes in five centres in North-Eastern Italy between April 2008 and April 2013. Results: 306 malignant polyps in 306 patients were included; 72 patients underwent surgery directly (23.6%). Of 234 patients treated endoscopically, 133 subsequently underwent radicalisation surgery (56.8%) and in 17 there was evidence of residual disease (12.8%). Involved, unsafe (<1 mm) or invaluable resection margins and sessile morphology represented the most frequent determinants of subsequent surgery. The mean number of nodes harvested during radicalisation surgery was 7.1 ± 6.4 (range 0–29). Histological diagnosis was re-evaluated according to new guidelines in 125 cases (41%); in 18 this led to modification of the risk class (14.4%). Conclusions: Although the rate of surgical treatment following endoscopic resection is similar to other studies, residual disease at surgery was lower than most international series. Adhering to the new histological reporting system and respecting guidelines on node harvesting may favourably influence prognosis. © 2015 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.
1. Introduction Colorectal malignant polyps (T1 tumour, according to TNM classification) are cancerous polyps that have invaded the sub-mucosal layer. The diffusion of bowel cancer screening programmes has determined the increased detection of these less advanced forms of the tumour. Therefore, there is growing awareness by both clinicians and pathologists regarding the diagnosis and management.
∗ Corresponding author at: Endoscopy Unit, Imperia Hospital, Via Sant’Agata 57, 18100 Imperia, Italy. Tel.: +39 340 4662570; fax: +39 0183 537496. E-mail address: [email protected] (R. Fasoli).
Traditionally, malignant polyps have been considered low risk and therefore safely treatable by means of endoscopic resection alone if: (a) the lesions were not poorly differentiated; (b) there was no vascular or lymphatic invasion; (c) the margins of excision were not involved [1,2]; (d) an en-bloc polypectomy was performed [3]. More recently, a European panel has issued new guidelines in colorectal cancer (CRC) screening and diagnosis. Regarding the histological judgement of malignant polyps, new features such as budding and micro-staging have acquired significant weight [4]. Up to now, no controlled trial has compared endoscopic treatment alone with endoscopic + surgical treatment in the management of these lesions. In this study, we report the results of a multi-centre study of malignant polyps diagnosed during CRC screening programmes.
http://dx.doi.org/10.1016/j.dld.2015.04.011 1590-8658/© 2015 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Fasoli R, et al. The management of malignant polyps in colorectal cancer screening programmes: A retrospective Italian multi-centre study. Dig Liver Dis (2015), http://dx.doi.org/10.1016/j.dld.2015.04.011
G Model
ARTICLE IN PRESS
YDLD-2878; No. of Pages 5
R. Fasoli et al. / Digestive and Liver Disease xxx (2015) xxx–xxx
2
Presentation, endoscopic and surgical management and follow-up are analysed. A special focus has been given to residual disease at surgery following polypectomy and to the role of histological factors in determining decision-making, including a re-evaluation of endoscopic samples by dedicated pathologists. 2. Materials and methods We retrospectively evaluated data on all consecutive colorectal malignant polyps detected during screening programmes between April 2008 and April 2013 in five centres of North-Eastern Italy: Trento, Rovereto (Trento), Padova (Ospedale Sant’Antonio), Bassano (Vicenza) and Feltre (Belluno). All involved screening programmes following the protocol of total colonoscopy in subjects of both sexes, aged 50–69 years, who tested positive for faecal immunochemical occult blood test (FIT). A standardised case record form was sent to all participating centres. Demographic and baseline clinical features, endoscopic characteristics of the lesions, original histological diagnosis, management and follow-up have all been collected. The following comorbidities were considered relevant: diabetes mellitus, significant cardiovascular and respiratory conditions, chronic kidney disease, chronic liver disease, present or previous malignant tumours, and degenerative neurological conditions. Residual disease was defined as the presence of any residual carcinoma at the site of polypectomy or in the nodes, or both. Recurrence of disease was defined as the presence of locally or regionally recurrent disease or distant metastases after a curative resection. A minimum follow-up of 18 months was required for all patients, to achieve a valuable judgement of recurrent disease. Longer follow-up periods, when available, allowed an evaluation of surveillance recommendations. Single centres have followed local clinical protocols and no “a priori” attempt to homogenise clinical behaviour has been made. According to the timing of diagnosis, classification into low- and high-risk lesions has changed from the traditional approach [1,2] to a more modern one based on new guidelines [4]; consequently, the histological reporting form of endoscopically resected lesions has changed. For this reason specialised pathologists re-evaluated the cases previously enrolled.
2.1. Statistical analysis Categorical variables were reported as numbers and summarised as proportions; continuous variables were expressed as means ± standard deviation (SD). Statistical comparisons were performed using Student’s t test (comparisons between two groups) or one-way ANOVA (comparisons amongst three groups) for continuous variables (age, size of the lesions, number of nodes resected). Chi-squared test was used for categorical variables (gender, site distribution and morphology of the polyps, histological items involved in therapeutic judgement, rates of direct and radicalisation surgery). A p value of less than 0.05 was considered to represent a statistically significant difference. 2.2. Ethical approval A recently published Code of Clinical Research (National Code on Clinical Researches) [5] declared that retrospective archive studies do not need ethics approval. According to the Declaration of Helsinki, every precaution was taken to protect the privacy of the subjects involved and the confidentiality of their personal information. 3. Results In the five-year period under consideration, there were a total of 14,943 FIT-positive subjects. Of them, 13,588 underwent colonoscopy (adhesion rate: 90.9%). A total number of 306 malignant polyps in 306 patients were observed. One hundred and ninety-three patients (63%) were male. The mean age was 63.1 ± 6.2 years (range 50–69). Seventy-two cases were referred for direct surgical intervention (23.6%). The indications for direct surgery were: size of the lesion in 26 (36%), depressed morphology in 23 (32%), no-lifting sign in 6 (8%), and technical problems during resection in 6 (8%). Information is missing in 12 cases (16%). Two hundred and thirty-four cases were resected endoscopically (76.4%) and 133 of these were referred for radicalisation surgery (56.8%). Table 1 reports the demographic, clinical and endoscopic features of patients undergoing direct surgery, endoscopic resection
Table 1 Baseline characteristics of malignant polyps. Endoscopic resection only n = 101
Radicalisation surgery n = 133
Direct surgery n = 72
Total n = 306 197 (63%) 109 (37%)
p value
Gender Males Females
68 (67%) 33 (33%)
83 (62%) 50 (48%)
46 (64%) 26 (36%)
Mean age (±SD# )
63.1 (±6.9)
63.0 (±5.7)
63.1 (±5.3)
Comorbidities 0 1 ≥2 Missing data
76 (75%) 12 (12%) 5 (5%) 8 (8%)
100 (75%) 19 (14%) 4 (3%) 10 (8%)
53 (74%) 9 (12%) 3 (4%) 7 (10%)
230 (75%) 40 (13%) 12 (4%) 24 (8%)
0.979a
Site Right colon Left colon Rectum
10 (10%) 58 (57%) 33 (33%)
20 (15%) 83 (62%) 30 (23%)
30 (42%) 31 (43%) 11 (15%)
60 (20%) 172 (56%) 74 (24%)
<0.001a
Mean size
15.8 mm (±8.0)
19.2 mm (±9.0)
37.5 mm (±17.9)
Morphology Sessile/Flat Pedunculated
30 (30%) 71 (70%)
97 (73%) 36 (27%)
66 (92%) 6 (37%)
a b
63.1 (±6.2)
22.4 mm (±10.5) 193 (63%) 113 (37%)
0.735a 0.987b
<0.001b
<0.001a
Chi-squared. One way ANOVA.
Please cite this article in press as: Fasoli R, et al. The management of malignant polyps in colorectal cancer screening programmes: A retrospective Italian multi-centre study. Dig Liver Dis (2015), http://dx.doi.org/10.1016/j.dld.2015.04.011
G Model
ARTICLE IN PRESS
YDLD-2878; No. of Pages 5
R. Fasoli et al. / Digestive and Liver Disease xxx (2015) xxx–xxx
3
Table 2 Differences amongst centres in direct surgery rates.
Trento Rovereto Bassano Feltre Padova a
Endoscopic resection n= 234
Direct surgery n = 72
Total n = 306
71 (75%) 30 (83%) 38 (73%) 46 (90%) 49 (70%)
24 (25%) 8 (21%) 14 (27%) 5 (10%) 21 (30%)
95 38 52 51 70
p valuea
0.108
Chi-squared test.
Table 3 Differences amongst centres in post-polypectomy radicalisation surgery rates.
Trento Rovereto Bassano Feltre Padova a
Endoscopic resection only n = 101 (43%)
Radicalisation surgery n = 133 (57%)
Total n = 234
21 (30%) 11 (37%) 12 (32%) 31 (67%) 26 (53%)
50 (70%) 19 (63%) 26 (68%) 15 (33%) 23 (47%)
71 30 38 46 49
p valuea
0.00033
Chi-squared test.
followed by surgical radicalisation and endoscopic resection only respectively. No significant differences were observed for age, gender and comorbidities. In patients treated with direct surgery, lesion size was significantly larger; right colon localisation and sessile/flat morphology occurred with significantly higher frequency. Conversely, in patients treated only endoscopically, rectal localisation and pedunculated morphology were significantly more frequent. No significant difference amongst centres was observed in the rates of direct surgery (Table 2); conversely, a wide variability amongst centres has been observed in the rates of radicalisation surgery (Table 3). Table 4 shows the different histological items taken into consideration in the therapeutic assessment. Involved, unsafe (<1 mm) or invaluable margins of resection occurred in 88% of cases referred for radicalisation surgery. Sessile morphology was also strikingly more common in the radicalised group than in the endoscopically treated one (p < 0.005). In three patients with high-risk histological features after endoscopic resection, severe comorbidities were contraindications for surgery. 3.1. Lymph node harvesting During direct surgical intervention, a mean 9.6 ± 5.4 nodes were resected for histological examination (range 3–42). In 42 cases fewer than 12 nodes were harvested (58%). Data were not available in 9 cases (8%).
Table 4 Histological items considered for therapeutic assessment. Histological item
Radicalisation surgery N = 133
Endoscopic resection only N = 101
Piece-meal resection Incomplete resection Involved margins Invaluable margins Unsafe margins (<1 mm) Lympho-vascular invasion Depth of submucosal invasion Tumour grading Budding Sessile/flat morphology
39 (29%) 4 (3%) 57 (43%) 29 (22%) 29 (22%) 23 (17%) 47 (35%) 15 (11%) 61 (46%) 97 (73%)
14 (14%) 0 (0%) 8 (8%) 3 (3%) 11 (11%) 0 (0%) 5 (5%) 0 (0%) 35 (35%) 30 (30%)
During post-polypectomy surgical intervention, a mean 7.1 ± 5.8 nodes were resected (range 0–29). In 90 cases fewer than 12 lymph nodes were retrieved (68%). Data were not available in 10 cases (13%). The mean number of lymph nodes removed in patients who underwent direct surgery was significantly higher than in those who underwent post-polypectomy surgery (p = 0.027). 3.2. Residual disease Seventeen out of 133 patients sent to radicalisation surgery had evidence of residual disease (12.8%). Their TNM classification was as follows: T2N0: 6 cases; T2N1: 9 cases; T3N0: 2 cases. In this subgroup, the most frequently observed features were: involved/unsafe margins (n = 4), high-grade budding (n = 3), depth of sub-mucosal invasion (n = 3), lymphatic or vascular invasion (n = 2), tumour grading (n = 1), piece-meal resection (n = 1). 3.3. Follow-up Data on subsequent follow-up were available for 73 out of 101 subjects considered at low risk and treated only by endoscopic resection (72.3%). Pedunculated lesions accounted for 70% of the total, whereas 30% were sessile and flat. The mean follow-up was 44 months (range 24–54). In 90% of patients with sessile/flat lesions the protocol included an early follow-up (3–6 months), followed by a colonoscopy at 1–3 years. In pedunculated lesions a follow-up colonoscopy at 3 years was suggested in 52% of patients; 38% were seen at 1 year and 10% at 2 years. Only 37% of patients received radiological surveillance (ultrasound or computerised tomography). During the 18-month follow-up, no recurrence of disease was diagnosed in the endoscopically treated population; 3 recurrences (2 cases of distant metastases and 1 case of lymph-nodal involvement) were observed in the high-risk population referred for radicalisation surgery. 3.4. Histological re-evaluation A second opinion of histological samples was acquired in 125 cases (53% of endoscopically resected lesions). The re-evaluation led to a modification of the risk class in 18 cases (14.4%).
Please cite this article in press as: Fasoli R, et al. The management of malignant polyps in colorectal cancer screening programmes: A retrospective Italian multi-centre study. Dig Liver Dis (2015), http://dx.doi.org/10.1016/j.dld.2015.04.011
G Model YDLD-2878; No. of Pages 5
ARTICLE IN PRESS R. Fasoli et al. / Digestive and Liver Disease xxx (2015) xxx–xxx
4
Nine patients were downgraded from a high- to a low-risk class, whereas 9 patients were upgraded. The histological features considered in downgrading were: depth of invasion (n = 6), absence of vascular invasion (n = 2), negative margins (n = 1); the histological features considered in upgrading were: budding (n = 7), vertical margin involvement (n = 1), extensive sub-mucosal invasion (n = 1). Regarding therapeutic decisions, 8 patients would have had a different indication (6.4%): 5 patients resulted over-treated, and 3 under-treated.
4. Discussion No controlled trial is available that could shed light on the treatment of colorectal malignant polyps, in particular on the indication for surgical radicalisation after endoscopic removal. The most numerous contribution about this issue published so far comes from the study by Cooper et al. [6], who in 2012 reported a population-based analysis of colonoscopic polypectomy compared with surgery in 2077 patients observed in 2002–2005. Outcomes were similar in both groups, although a significant selection bias in the allocation of the treatment was present and few data were available regarding prognostic features other than the degree of differentiation of the cancer. A similar study was published in 2012, reflecting the current management of these lesions in a general setting in a few centres scattered throughout Northern England and offers interesting data regarding different issues [7]. A recent position statement of the Association of Coloproctology of Great Britain and Ireland (ACPGBI) underlines how it is more advisable and ethically appropriate to create a malignant polyp registry rather than embarking on a randomised controlled trial [8]. According to these authors, all clinicians should scrutinise regional non-randomised data carefully, particularly with the increasing detection of these malignant polyps via the national screening programmes. Careful audits of specific risk factors are recommended, to better define indications for surgical radicalisation. In light of these premises, we were stimulated to carry out a regional study on this type of lesion diagnosed within screening campaigns. To our knowledge, the present study is the first to specifically take into consideration the management of malignant polyps in a FIT-positive screening population. With regard to the site of malignant polyps in our series, our result (56%) of left colon localisation is in keeping with data published by Wasif et al. [9], whereas we had a slightly higher right colon localisation than that reported in the meta-analysis by Hassan et al. [10]. In our series 23.6% of subjects received an immediate surgical intervention (with no attempt at endoscopic removal). This figure is far below that reported in the series by Gill et al. [7], where 42% of subjects received biopsies and subsequent surgery. The difference could be due to a more aggressive endoscopic approach towards this kind of lesion observed in our population. Our indications for direct surgery are in keeping with contraindications to endoscopic resection in so-called difficult colorectal polyps (size, macroscopic morphology, no-lifting sign, and technical difficulties) [11]. Considering the indications for radicalisation surgery, several factors must be taken into account, mainly the risk of lymph-node metastases, surgical morbidity and mortality, and patients’ comorbidities.
The influence of histopathological factors on the risk of nodal metastases in early colorectal cancer has been taken in consideration in two recent meta-analyses that came to similar conclusions. A submucosal invasion deeper than 1 mm, lympho-vascular invasion, poor differentiation and tumour budding were significantly associated with lymph-node metastases [12,13]. Furthermore, a previous meta-analysis [10] showed that positive resection margins, poor differentiation of carcinoma, and vascular invasion were significantly associated with more general unfavourable clinical outcomes (residual disease, recurrent disease, lymph node metastases, haematogenous metastases, mortality). British authors have even created a score to predict the risk of residual disease following resection of a malignant polyp [8]. Overall, the frequency of surgical radicalisation reported in our study is in keeping with figures reported in the international literature [6,7]. Sessile morphology and the conditions of margins of resection of the endoscopically removed tissue appear to have played major roles in our decision-making behaviour. In this regard, as the lack of margin evaluation (mainly related to specimen fragmentation) has proved to be a frequent cause of attribution to high risk class, resection/handling of the lesion should be a major concern for endoscopists. Unlike available series recently published in the literature [6,14,15], lympho-vascular invasion, tumour grade and high-grade budding have only marginally influenced treatment plans. The study by Cooper et al. [6] has shown – in a general population setting – that comorbidities and advanced age play important roles in decision-making. Our study concerns a screening population with a relatively young age and low rates of relevant comorbidities; furthermore, counselling before colonoscopy often leads to the exclusion of severely ill and frail patients. For these reasons, histological/morphological factors have been the main determinants in the majority of cases. The incidence of residual disease in the subgroup of patients undergoing surgical intervention after endoscopic resection shows some degree of variability in the literature, both at the bowel wall and at the lymph nodes. Reported rates range between 15% and 27% [7,14,16,17]. In our experience, such a rate is slightly lower than reported in the above-mentioned series collected in the general population [7,14,16,17]. Considering that we are dealing with a screening population, this difference may hypothetically account for a less severe form of the disease in this subgroup of patients. As regards the pattern of histological parameters considered crucial in the decision on surgical radicalisation, particular value has been given to the state of tumour margins. This type of behaviour is in keeping with the quoted British score (which attributes 4 points of severity to margins lower than 1 mm) [8] and with a North American retrospective study [18]. Naqvi et al. came to different conclusions and suggested that clear resection margins, even those with less than 1 mm of clearance, might be treated safely only with surveillance [19]. Unfortunately, the paucity of patients with evidence of residual disease at surgery in our series allows neither predictors of residual disease at surgery to be analysed nor wall and lymph node involvement to be differentiated. Benhaim et al. compared lymph-node harvest from 22 patients undergoing radicalisation resection after polypectomy of a malignant lesion, with 509 patients undergoing direct colectomy for invasive CRC. Significantly fewer lymph nodes were retrieved in the former group (11.63 vs 26.3 respectively) [20]. A study by Wasif et al. came to similar and even more relevant conclusions. They described – in a general population setting – a
Please cite this article in press as: Fasoli R, et al. The management of malignant polyps in colorectal cancer screening programmes: A retrospective Italian multi-centre study. Dig Liver Dis (2015), http://dx.doi.org/10.1016/j.dld.2015.04.011
G Model YDLD-2878; No. of Pages 5
ARTICLE IN PRESS R. Fasoli et al. / Digestive and Liver Disease xxx (2015) xxx–xxx
median removal of just one node. This type of behaviour has prognostic consequences, as patients with fewer resected lymphnodes have been shown to have a survival disadvantage [9]. In our series, the mean number of resected lymph nodes at surgery is below the threshold indicated by recent guide-lines, which suggest a minimum number of 12 [21]. This occurred both in direct surgery (mean 9.6) and – more strikingly – in postpolypectomy surgery (mean 7.1). A possible explanation for this result is a higher confidence within the surgical team regarding the prognosis of patients referred to surgery after endoscopic polypectomy. Our data on follow-up in patients who only received endoscopic treatment, although limited to 18 months, confirm the favourable outcome of this population, with no recurrences diagnosed in the period considered, in keeping with the British study [7]. Proposed protocols in sessile/flat lesions comply satisfactorily with international guidelines. In pedunculated lesions the quality of surveillance intervals can be improved, with the aim of prescribing with higher frequency a 3-year endoscopic follow-up. In fact, it has been shown that the short-term outcomes of these lesions are similar to those of non-malignant pedunculated polyps [22]. Considering behaviour in planning radiological follow-up after endoscopic polypectomy, British and North American authors have different opinions: the former are prone to offer imaging follow-up based on the rate of distant metastases; the latter are less inclined to do so. In our experience radiological follow-up was offered to 37% of this subgroup of patients. As regards the subgroup referred for surgical intervention, either directly or after endoscopic resection, follow-up recommendations comply in general with international guidelines following surgery for CRC of any grade. Budding and depth of submucosal invasion represent the most frequently quoted items at the re-evaluation of histological samples. While budding is a novel parameter with high sensitivity and acceptable specificity with regard to lymph-node metastases [4], the degree of sub-mucosal infiltration has been identified very early as a risk factor for advanced disease. The varying judgement concerning degree of sub-mucosal infiltration reflects the importance of a specialised pathologist [23]. Approximately 15% of re-evaluated patients changed their risk class; in almost half of them this would have led to a different treatment. These data, taken together, underline the importance of having clinical/pathological audits in this setting. The main limitations of the study are its retrospective nature, the suboptimal collection of data regarding some items and a thorough follow-up limited to 18 months. Although our rate of surgical treatment of colorectal malignant polyps following endoscopic resection complies with that of other series carried out in general populations, the rate of residual disease at surgery lies below the majority of international experiences. Therefore, bearing in mind the potential risk of over-treatment in these patients, we believe that it is mandatory to have a rigorous multi-disciplinary discussion of treatment options in each case, as well as considering the patient’s point of view and preference. Furthermore, adhering to a new histological reporting system and respecting guidelines on lymph node harvesting could favourably influence the patient’s prognosis. The rate of endoscopically treated patients offered an imaging follow-up for the early detection of distant metastases turned out to be sub-optimal: we believe that local protocols ought to be planned
5
to stimulate the regular prescription of extra-colonic examinations and – possibly – homogenise the surveillance of these patients. Conflict of interest None declared. Funding Richard Nienstedt was holder of a two-year salary grant from Pezcoller Foundation (Trento). References [1] Bond JH. Polyp guideline: diagnosis, treatment and surveillance for patients with colorectal polyps. Practice Parameters Committee of the American College of Gastroenterology. American Journal of Gastroenterology 2000;95: 3053–63. [2] Muto T, Sawada T, Sugihara K. Treatment of carcinoma in adenomas. World Journal of Surgery 1991;15:35–40. [3] Ramirez M, Schiering S, Papaconstantinou HT, et al. Management of the malignant polyp. Clinics in Colon and Rectal Surgery 2008;21:286–90. [4] Quirke P, Risio M, Lambert R, et al. European guidelines for quality assurance in colorectal cancer screening and diagnosis. First Edition – quality assurance in pathology in colorectal cancer screening and diagnosis. Endoscopy 2012;44(Suppl. 3):SE116–30. [5] National Code on Clinical Researches published in Official Gazette numbered with 28030 at April 13, 2013. [6] Cooper GS, Xu F, Barnholtz Sloan JS, et al. Management of malignant colonic polyps: a population-based analysis of colonoscopic polypectomy versus surgery. Cancer 2012;118:651–9. [7] Gill MD, Rutter MD, Holtham SJ. Management and short term outcome of malignant colorectal polyps in the north of England. Colorectal Disease 2013;15:169–76. [8] Williams JG, Pullan RD, Hill J, et al. Management of the malignant colorectal polyp: ACPGBI position statement. Colorectal Disease 2013;15(Suppl. 2):1–38. [9] Wasif N, Etzioni D, Maggard MA, et al. Trends, patterns, and outcomes in the management of malignant colonic polyps in the general population of the United States. Cancer 2011;117:931–7. [10] Hassan C, Zullo A, Risio M, et al. Histologic risk factors and clinical outcome in colorectal malignant polyp: a pooled-data analysis. Diseases of the Colon & Rectum 2005;48:1588–96. [11] Riley SA. Colonscopic polypectomy and endoscopic mucosal resection: a practical guide; 2008. http://bsg.org.uk/pdf word docs/polypectomy 08.pdf [12] Beaton C, Twine CP, Williams GL, et al. Systematic review and meta-analysis of histopathological factors influencing the risk of lymph node metastases in early colorectal cancer. Colorectal Disease 2013;15:788–97. [13] Bosch SL, Teerenstra S, de Wilt JH, et al. Predicting lymph node metastasis in pT1 colorectal cancer: a systematic review of risk factors providing rationale for therapy decisions. Endoscopy 2013;45:827–34. [14] Ueno K, Mochizuki H, Hashiguchi Y, et al. Risk factors for an adverse outcome in early invasive colorectal carcinoma. Gastroenterology 2004;127:385–94. [15] Kim KM, Eo SJ, Shim SG, et al. Risk factors for residual cancer and lymph node metastasis after noncurative endoscopic resection of early colorectal cancer. Diseases of the Colon and Rectum 2013;56:35–42. [16] Volk EE, Goldblum JR, Petras RE, et al. Management and outcome of patients with invasive carcinoma arising in colorectal polyps. Gastroenterology 1995;109:1801–7. [17] Seitz U, Bohnacker S, Seewald S, et al. Is endoscopic polypectomy an adequate therapy for malignant colorectal adenomas? Presentation of 114 patients and review of the literature. Diseases of the Colon and Rectum 2004;47:1789–96. [18] Butte JM, Tang P, Gonen M, et al. Rate of residual disease after complete endoscopic resection of malignant colonic polyp. Diseases of the Colon and Rectum 2012;55:122–7. [19] Naqvi S, Burroughs S, Chave HS, et al. Management of colorectal polyp cancers. Annals of the Royal College of Surgeons of England 2012;94:574–8. [20] Benhaim L, Benoist S, Bachet JB, et al. Salvage colectomy for endoscopically removed malignant colon polyps: is it possible to determine the optimal number of lymph nodes that need to be harvested? Colorectal Disease 2012;14:79–88. [21] Desch CE, McNiff KK, Schneider EC, et al. American Society of Clinical Oncology/National Comprehensive Cancer Network Quality Measures. Journal of Clinical Oncology 2008;26:3631–7. [22] Freeman HJ. Long-term follow-up of patients with malignant pedunculated colon polyps after colonoscopic polypectomy. Canadian Journal of Gastroenterology 2013;27:20–4. [23] Bong-Hyeon Kye, Ji-Han Jung, Hyung-Jin Kim, et al. Tumor budding as a risk factor of lymph node metastasis in submucosal invasive T1 colorectal carcinoma: a retrospective study. BMC Surgery 2012;12:16.
Please cite this article in press as: Fasoli R, et al. The management of malignant polyps in colorectal cancer screening programmes: A retrospective Italian multi-centre study. Dig Liver Dis (2015), http://dx.doi.org/10.1016/j.dld.2015.04.011