Colorectal cancer (CRC) has been a challenging clinical entity, particularly in the West. It is the third most common cancer in men and the second in women worldwide according to the World Health Organization (WHO) GLOBOCAN 2012 database. Having a mortality rate of over 6,00,000/year, it accounts for 9% of the total cancer deaths. However, its incidence is declining in the West, mainly due to prevalent educational and screening programs designed to detect early cancers and their precursor polyps. A cascade of recognizable precursor lesions precedes most CRC and 95% of CRCs arise from adenomatous polyps. The WHO classifies adenomatous polyps into tubular (87%), tubulovillous (8%), and villous (5%). The adenoma–carcinoma sequence of CRC is often an indolent process spanning over 10 years making it an ideal cancer preventable by screening.[5,6] Such screening programs do not exist in India, however, multiple studies have established a rising trend in epidemiology of CRCs in India pertaining to wide adoption of Western lifestyle habits. Kerala has the highest incidence of CRC in India, and it has shown incremental year-wise trends. In such a scenario, the only way to control CRC is to focus on strategies to reliably detect and completely resect adenomas at presentation before they become malignant.
Villous adenoma (VA) is the least common (<5%) histological type of adenomatous polyp, but its tendency for malignant change is the greatest (up to 40%). It is impossible to distinguish the grades of dysplasia of VA by their macroscopic appearance under ordinary white light colonoscopy. Superficial biopsy is inaccurate in this setting, missing invasive cancer about 40% of the time.[10,11] Therefore, it becomes useful for clinical management to investigate its clinicopathological characteristics associated with tumor development.
There are only a limited number of published studies related to colorectal VA reported from India. Older age, male sex, site of colonic involvement, grade of dysplasia, and multiplicity/synchronicity are the known risk factors quoted in the literature. This study is aimed toward better correlation of these risk factors from an Indian context and a step toward early detection of VA, timely intervention, and prevention of progression to CRC. The study also aims to evaluate if ulceration of colorectal VAs at index colonoscopy can predict the presence of invasive malignancy. This is the first study comparing the clinical and demographic characteristics of VA in an Indian cohort.
MATERIALS AND METHODS
Hospital-based computerized data search was performed in Amrita Institute of Medical Sciences, Kochi, to retrieve records of patients with a diagnosis of colorectal VA between January 2000 and June 2014. Subjects having a diagnosis of colorectal VA confirmed either by histopathology biopsy or on surgical excised specimen were included in the study. Patients were excluded if adenomas were arising from areas other than colorectal sites, conditions such as familial polyposis and other cancer syndromes, or presence of synchronous cancers in the rest of colorectum.
If the patients had colonoscopy done elsewhere (n = 30/98), the slides and blocks were obtained and a second opinion was taken from our dedicated G. I. Pathology team to reduce any potential for subjective errors.
This is a cross-sectional, comparative study. The statistical analysis was done using SPSS for Mac version 15 (IBM Inc, Armonk, New York, NY 10504, USA). The percentage incidence rate of malignancy in VA was computed, and to test the statistical significance of the association of VA with clinicopathological features, Chi-square test was applied.
The clinicopathological features were compiled and evaluated for age and gender distribution, site of colon involved, and associated clinical features. The period of follow-up ranged from 0 to 11 years, with a mean value of 28 months. The mean follow-up period for the polypectomy cases was 24 months. We understand that this follow-up period is less than the minimum period of surveillance (3 years) needed for high-risk adenomas.
Most of the patients of VA with associated malignancy were <60 years, with a mean age at presentation being 59.5 ± 12.21 years, whereas patients having VA with no associated adenocarcinoma were more than 60 years, with a mean age at presentation being 66 ± 11.21 years. These results were statistically significant (P = 0.012).
Out of the entire cohort of 98 cases [Figure 1], polypectomy was done in 31 cases (15 – transanal polypectomy and 16 – snare polypectomy). All the snare polypectomies were done as upfront procedures. Only one proved to be malignant. Seven cases with various comorbidities (chronic liver disease, severe airway obstruction, poor performance of the patients, etc.) were not offered local excision and kept on surveillance. Two cases were not offered any intervention due to distant metastasis. The remaining 58 cases were subjected to surgical resection.
Our study showed that 56.1% (55/98) of VA were associated with high-grade dysplasia or associated invasive malignancy [Figure 2].
Males were more commonly affected than females in all groups (1.5:1). In VA without malignant transformation, the male-to-female ratio was 2:1. Rectum was most commonly involved. The incidence of VA in the left-sided colon was 80% [Figure 3].
Due to the absence of screening program in India, all the 98 patients presented were symptomatic. Rectal bleeding was the most common presenting complaint. The incidences of various symptoms in both groups are shown in Table 1.
Only 26% of synchronous polyps were found and had no significant impact on malignancy. Ulceration on polyps at index colonoscopy was found to be a significant risk factor of invasive carcinoma (45.5%) (P = 0.016). Most of the polyps ≥2 cm were having invasive malignancy (69.4%) [Figure 4].
When compared with grades of dysplasia, 45.5% of VAs with low-grade dysplasia and 54.5% (30/98) of VAs with high-grade dysplasia had associated invasive malignancy. In our study, the grade of dysplasia did not correlate with tumor differentiation in well and moderately differentiated tumors, but the correlation was observed in poorly differentiated tumors which tend to have high grades of dysplasia. The correlation of dysplastic change in VA to tumor differentiation was marginally statistically significant (P = 0.056)
The mean size of the polyps that had colonoscopic polypectomy was 2 cm. Out of the 89 cases that received definitive treatment and in whom surgical histopathology reports were obtained, 55 cases were diagnosed to be CRC. Of these 55, in 28 cases malignancy was not detected on colonoscopic biopsy and captured only after surgical histopathology.
There were 12 cases that were lost to follow-up, of which 7 were kept on surveillance.
Considerable progress has been made recently in both the diagnosis and treatment of colorectal VA, yet the determination of malignant potential of such lesions may be difficult. Early diagnosis by endoscopy at the precancerous stage may decrease the prevalence and mortality rates. The preventive role of emerging endoscopic imaging techniques, such as chromoendoscopy, magnification endoscopy, and confocal laser endomicroscopy, has given rise to myriad innovations such as targeted biopsies and real-time in vivo diagnosis of surface, vascular, and cellular patterns, thereby increasing the diagnostic yield for such lesions. It is important that clinicians are able to recognize clinicopathological correlates associated with its progression to malignancy. These aspects have been studied in Western populations, but comparatively, little information is available for developing countries like India. It assumes special significance since we neither have a national guideline nor clinical practice of routine CRC screening. The present study is an investigation of 98 consecutive patients with colorectal VA.
The high incidence of premalignant and malignant change in VA has long been recognized. The same observation was made in our cohort of 98 VA cases where 56.1% demonstrated a focus of invasive adenocarcinoma. This value is higher than that previously reported in the literature. Bujanda et al. have mentioned in their review that the endoscopist should be alert to some features that are suggestive of possible malignancy. The features include depressed ulceration along with size, irregular contours, deformity, a short and immobile stalk, and the inability to elevate a sessile polyp when a submucosal bleb is formed. Winawer et al. ascertained the incidence of CRC invasive growth in the National Polyp Study (n = 1418) and their finding supported the view that colorectal adenomas progress to adenocarcinomas, as well as the current practice of searching for and removing adenomatous polyps to prevent CRC. In one of the first few studies, Jahadi and Baldwin reported a 13-year single-center experience with 264 patients and concluded that although 55% of the lesions had associated malignant foci, only 22% harbored invasive carcinoma. Shinya and Wolff reported an incidence of 15% cancer in situ and 10% invasive cancer with VA. Most of the other previous studies have reported rates of invasive cancer between 1% and 40%.[14,15,16]
In our study, most patients with VA showing malignant transformation were <60 years, with a mean age at presentation being 59.5 ± 12.21 years. In comparison, those patients having VA without associated invasive adenocarcinoma were more than 60 years, with a mean age at presentation being 66 ± 11.21 years (P = 0.012). This is counterintuitive for the adenoma–carcinoma sequence. It can be postulated from this result that VA destined for malignant transformation possesses an innately aggressive disease per primum and attains malignant transformation a decade earlier, indicating toward the importance of surveillance and timely intervention. This can form a basis to model our screening program targeting an earlier age group as compared to the West. VAs without invasive malignancy were more common in the sixth decade of life, and numerically, it was found to be more in males; the values are similar to those previously reported in the literature.[14,17]
Interestingly, 45.5% of VAs with low-grade dysplasia and 54.5% of VAs with high-grade dysplasia had associated invasive malignancy. Hence, it would be best to infer from our study that VA can have associated invasive malignancy irrespective of grade of dysplasia and that the pathological grade of dysplasia on the index colonoscopic biopsy may not be representative sample of the entire lesion. Our results are partly in contrast with Cho et al. where all cases of invasive cancer arose from lesions classified as high-grade dysplasia.
In India, there is no screening program for detection of CRC, hence all the 98 patients included in the study were symptomatic. While rectal bleeding was the most common presenting complaint, it was common in both groups. Hence, clinical symptoms may not be a good differentiating factor for suggesting the possibility of malignancy. This accords with presenting symptoms in similar other studies.
In our study, 80% of the VAs arise from the left-sided colon. Hence, flexible sigmoidoscopy can be a valuable screening tool, which is cost-effective and less cumbersome, compared to colonoscopy. This has also been suggested by the Asia-Pacific Working Group for CRC and has come up with a recommendation of “stratified screening approach” for CRC in Asian population.
Synchronicity and size of polyps were not predictive of malignancy in our cohort. Similar trend was reported by Gschwantler et al. Interestingly, ulceration on colorectal polyps at index colonoscopy was found to be a predictor of malignancy. This association has never been studied so far. This fact needs further validation with population-based studies in different geographies.
The fact that 43.1% of cases of VA with either low-grade or high-grade dysplasia were missed malignancy on colonoscopic biopsy and diagnosed for the first time as malignancy on surgical histopathology supports complete excision as the only means of definitive diagnosis. Other authors have also reported low levels of detection of invasive malignancy in VA. This highlights the importance of taking biopsies that are adequate and representative. Targeted biopsies by well-trained endoscopists and using advanced colonoscopic techniques and technologies such as high magnification colonoscopy, chromoendoscopy, or narrow-band imaging will improve diagnostic yield and accuracy.
However, there is no evidence to include advanced endoscopy in surveillance program till date. The authors cite reasons for missing carcinoma on biopsy, as mentioned in Table 2.
Our study has certain limitations. It is a single-center study. The study has comparatively smaller sample size to derive long-term clinical conclusions. Second, few patients included in our study underwent colonoscopy at other centers and were referred to our center afterward. Hence, inter-observer difference may have confounded polyp size, ulceration, and multiplicity. Third, the mean period of follow-up was less than the recommended minimum period of surveillance for high-risk adenomas of 3 years.
Colorectal VA in Indian population has a higher risk of malignant transformation compared to the West. Ulceration on colorectal VAs at index colonoscopy is a strong predictor of invasive malignancy. Grading of dysplasia in VA may not be accurate, especially if the biopsy is superficial and inadequate, and therefore, it is imperative to correlate with radiology in diagnosing invasive malignancy and in taking management decisions. The authors recommend that there should be a high degree of suspicion of invasive malignancy in VAs before subjecting the patients to specific line of management.
- Incidence of malignancy arising from VA is higher for Indian population than previously reported in worldwide literature
- Malignant VA tends to occur a decade earlier in India
- Ulceration on colorectal polyps at index colonoscopy may be a predictor of underlying malignancy, but the fact needs further validation
- VAs were commonly found in the rectum and left-sided colon
- Clinical symptoms are not a good guide for identifying malignancy
- VAs of size ≥2 cm tend to be malignant
- Synchronicity had no significant impact for malignant transformation of VA
- Colonoscopic biopsy may miss malignant foci in VAs.
- VAs can have associated invasive malignancy irrespective of grade of dysplasia
- Colonoscopic biopsies must be adequate and representative of the lesion
- Advanced colonoscopic techniques will improve diagnostic yields
- Colonoscopic biopsy results must be correlated with clinical and radiological findings for management decisions of VA
- Colonic screening programs should be started around the age of 40 years in view of increased incidence of CRC and malignant transformation of VA occurring a decade earlier in our study group.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
1. Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelo M, et al Cancer incidence and mortality worldwide: Sources, methods and major patterns in GLOBOCAN 2012 Int J Cancer. 2015;136:E359–86
2. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013 CA Cancer J Clin. 2013;63:11–30
3. Castells A, Marzo-Castillejo M, Mascort JJ, Amador FJ, Andreu M, Bellas B, et al Prevención del cáncercolorrectal. Actualización 2009 Gastroenterol Hepatol. 2009;32:717.e1–58
4. Shinya H, Wolff WI. Morphology, anatomic distribution and cancer potential of colonic polyps Ann Surg. 1979;190:679–83
5. Bujanda L, Cosme A, Gil I, Arenas-Mirave JI. Malignant colorectal polyps World J Gastroenterol. 2010;16:3103–11
6. Molatore S, Ranzani GN. Genetics of colorectal polyps Tech coloproctol. 2004;8(Suppl 2):s240–2
7. Pourhoseingholi MA. Epidemiology and burden of colorectal cancer
in Asia-Pacific region: What shall we do now? Transl Gastrointest Cancer. 2014;3:169–73
8. . Consolidated Report of Hospital Based Cancer Registries: 2007-2011Last accessed on 2021 Jun 30 Available from: https://ncdirindia.org/ncrp/ALL_NCRP_REPORTS/HBCR_REPORT_2007_2011/index.htm
9. Kim EC, Lance P. Colorectal polyps and their relationship to cancer Gastroenterol Clin North Am. 1997;26:1–17
10. Galandiuk S, Fazio VW, Jagelman DG, Lavery IC, Weakley FA, Petras RE, et al Villous and tubulovillous adenomas of the colon and rectum. A retrospective review, 1964-1985 Am J Surg. 1987;153:41–7
11. Taylor EW, Thompson H, Oates GD, Dorricott NJ, Alexander-Williams J, Keighley MR. Limitations of biopsy in preoperative assessment of villous papilloma Dis Colon Rectum. 1981;24:259–62
12. Winawer SJ, Zauber AG, O'Brien MJ, Ho MN, Gottlieb L, Sternberg SS, et al Randomized comparison of surveillance intervals after colonoscopic removal of newly diagnosed adenomatous polyps. The National Polyp Study Workgroup N Engl J Med. 1993;328:901–6
13. Jahadi MR, Baldwin A Jr. Villous adenomas of the colon and rectum Am J Surg. 1975;130:729–32
14. Cho SD, Herzig DO, Douthit MA, Deveney KE. Treatment strategies and outcomes for rectal villous adenoma from a single-center experience Arch Surg. 2008;143:866–70
15. Sung JJ, Ng SC, Chan FK, Chiu HM, Kim HS, Matsuda T, et al An updated Asia pacific consensus recommendations on colorectal cancer
screening Gut. 2015;64:121–32
16. Gschwantler M, Kriwanek S, Langner E, Göritzer B, Schrutka-Kölbl C, Brownstone E, et al High-grade dysplasia and invasive carcinoma in colorectal adenomas: A multivariate analysis of the impact of adenoma and patient characteristics Eur J Gastroenterol Hepatol. 2002;14:183–8
17. Solakoğlu T, Atalay R, Köseoğlu H, Özer Sarı S, Demirezer Bolat A, Akın E, et al Analysis of 2222 colorectal polyps in 896 patients: A tertiary referreal hospital study Turk J Gastroenterol. 2014;25:175–9
18. Nivatvongs S, Nicholson JD, Rothenberger DA, Balcos EG, Christenson CE, Nemer FD, et al Villous adenomas of the rectum: The accuracy of clinical assessment Surgery. 1980;87:549–51
19. Kahi CJ, Boland CR, Dominitz JA, Giardiello FM, Johnson DA, Kaltenbach T, et al Colonoscopy surveillance after colorectal cancer
resection: Recommendations of the US multi-society task force on colorectal cancer
Am J Gastroenterol. 2016;111:337–46