Curr Treat Options Gastro DOI 10.1007/s11938-017-0121-7

Colon (J Anderson, Section Editor)

Screening and Surveillance of Colorectal Cancer Using CT Colonography Manoj Kumar, MD Brooks D. Cash, MD* Address * Division of Gastroenterology, University of South Alabama, 6000 University Commons, 75 University Blvd S, Mobile, AL, 36688, USA Email: [email protected]

* Springer Science+Business Media, LLC 2017

This article is part of the Topical Collection on Colon Keywords Colorectal cancer screening I CT colonography I Virtual colonoscopy I Colorectal cancer surveillance I Colon polyp detection

Opinion Statement Colorectal cancer (CRC) is a common cancer among throughout the world with the highest rates in developed countries such as the USA. There is ample evidence demonstrating the beneficial effects of colorectal cancer screening and, largely thanks to screening initiatives and insurance coverage, epidemiologic analyses show a steady decline in both CRC incidence and mortality rates over the last several decades. However, screening rates for CRC in the US remain low and approximately 1 in 3 adults between the ages of 50 and 75 years has not undergone any form of CRC screening, highlighting the need for additional accurate, minimally invasive, and acceptable screening options. Computed tomography colonography (CTC) has emerged as a viable alternative to existing CRC screening tests and research continues to enhance our knowledge regarding the ability of CTC to play a meaningful role in optimizing CRC screening in areas where it is available. This review highlights recent publications of salient research in the field of CTC. CTC continues to evolve, with lower radiation doses and greater evidence of its ability to identify clinical relevant colonic and extracolonic abnormalities. Recent evidence has bolstered the currently recommended CTC screening interval of 5 years and has reiterated the cost-effectiveness of CTC as a CRC screening examination. Additionally, emerging evidence suggests a role for CTC as a polyp and CRC surveillance modality as well as a preoperative adjunct in patients with established CRC. Data supporting the safety and patient acceptance of CTC also has continued to accumulate and CTC has recently been endorsed as an appropriate test for CRC screening in multiple important guidelines and recommendations. CTC is poised to become an important option in the CRC screening and surveillance arena.

Colon (J Anderson, Section Editor)

Introduction Colorectal cancer (CRC) is the second most common cancer among women and third most common cancer among men worldwide with an estimated 1.4 million cases and nearly 700,000 deaths in 2012 [1]. In the USA, Congress authorized coverage of certain CRC screening tests beginning in 1998 [2] and epidemiologic analyses in this country show a steady decline in both CRC incidence and mortality rates over the last several decades, attributed to screening, reduced prevalence of risk factors, and improved treatments [3, 4]. Level 1 evidence exists to support routine CRC screening for reducing CRC incidence and mortality, presumably through endoscopic polypectomy and early detection of CRC [5]. Nevertheless, the Centers for Disease Control (CDC) reported in 2013 that screening rates for CRC in the USA remain low and that approximately 1 in 3 adults between the ages of 50 and 75 years has not undergone any form of CRC screening [6]. This translates into more than 20 million people in the USA who remain unscreened and at risk for CRC-related morbidity and mortality. With a lifetime risk for the development of CRC without screening estimated to be between 3 and 5%, such staggering numbers have compelled researchers and health care policy makers to seek alternatives to overcome barriers to screening for this costly and largely preventable disease, including the development of new CRC screening tests. Numerous options for CRC screening currently exist, with colonoscopy remaining the most popular, since it facilitates visualization of the entire colonic mucosa as well as polypectomy, interrupting the adenomacarcinoma sequence and decreasing CRC incidence. Colonoscopy is endorsed as the preferred CRC screening test by several guideline issuing bodies such as the US Multisociety Task Force (MSTF) on Colon Cancer and the American College of Gastroenterology. However, due to its cost, inconvenience, invasiveness, and medical resource requirements, individuals often elect to use other screening methods or forego screening altogether,

resulting in suboptimal compliance rates in individuals at normal CRC risk as well as those at increased risk [7]. Thus, there is a need for additional accurate, accessible, and acceptable alternatives to colonoscopy to improve population screening and benefit. Computed tomographic colonography (CTC) is one of several alternative tests developed for CRC screening and, among the various CRC screening test options, appears to be the closest to colonoscopy for the accurate detection of advanced and non-advanced colonic neoplasia. CTC is similar to colonoscopy in that it permits visualization of the entire colonic mucosa, rather than evaluating just the distal portion (as with flexible sigmoidoscopy (FS)) or relying on the detection of shed products of colonic neoplasia in the stool (e.g., fecal immunochemical testing (FIT), fecal occult blood testing (FOBT), or fecal DNA testing (sDNA)). While CTC has been available for more than 20 years and is considered an acceptable screening alternative, it has failed to gain widespread acceptance from payers for this purpose. Even though the 2008 MSTF screening guidelines recommended CTC every 5 years as an acceptable CRC screening strategy, there has been minimal uptake of CTC nationwide, likely due to the lack of insurance coverage [8]. Most recently, the US Preventive Services Task Force (USPSTF) recommendations for CRC screening included CTC as an acceptable option among other effective and established CRC screening tests, largely based on the accumulated data supporting the accuracy of CTC for CRC screening [9••]. Surveillance after a diagnosis of colonic neoplasia represents another important role for colonoscopy [10]. The evidence for CTC as a colonic neoplasia surveillance modality is less robust than for screening, but evaluation of CTC for surveillance has important implications for its future integration into routine practice. This chapter will highlight recent publications and data surrounding the performance of CTC for CRC screening and surveillance.

CTC Background First developed in 1994, [11] CTC has become an increasingly common method for colorectal evaluation. Contrary to colonoscopy, CTC does not require sedation and therefore has the potential to be less disruptive with regards to

Screening and Surveillance of Colorectal Cancer Using CTC

Kumar and Cash

procedural burden in the form of sedation risks and lost productivity. As a noninvasive CRC screening modality, CTC is also inherently less risky than colonoscopy involving polypectomy, due to the lower potential for polypectomyassociated adverse events. Similar to other CRC screening tests that require colonoscopy confirmation if positive, patients undergoing CTC for CRC screening should be asymptomatic and at average risk for the development of CRC [8]. Additional indications for CTC include incomplete colonoscopic screening or surveillance as well as CRC screening in patients who are at increased risk for colonoscopy-related complications (e.g., patients on anticoagulation medication or those with anesthesia risks). Recently, CTC has shown potential utility in the preoperative evaluation and subsequent management of patients with colon cancer. Sato et al. found that CTC successfully identified CRC location with a sensitivity of 98.9% and that it was useful in illustrating the feeding artery of identified cancers, facilitating surgical planning and precise lymph node dissection [12]. Contraindications to CTC include a history of intestinal obstruction, suspected peritonitis, recent abdominal surgery, pregnancy, and history of allergy to iodinated contrast. In patients with allergy to iodinated contrast, using water as a contrast agent rather than diatrizoate sodium can be done, but the efficacy of this technique has not been widely studied. [13]. Conventional CTC requires patients to consume a purgative bowel preparation similar to colonoscopy. The bowel preparation with CTC also typically includes dilute barium sulfate and diatrizoate sodium for stool and colonic fluid tagging, respectively. This tagging permits post-acquisition digital removal and cleansing of CTC images to enhance the differentiation of stool from polyps during CTC interpretation. During the CTC procedure, the colon is distended with volume and pressure regulated insufflation of CO2 via a rectal catheter. Once a scout film demonstrates adequate colonic insufflation, image acquisition is performed using a low radiation dose (approximately 2–4 mSv) helical CT scan protocol. The raw CT image data is then processed by sophisticated software tools into CTC images, which may be viewed and manipulated in both 2D and 3D formats (Figs. 1, 2, and 3). Most CTC readers utilize a primary 3D reading format which closely resembles the endoscopic view during colonoscopy and use 2D images for problem-solving or additional query of images suspicious for intracolonic neoplasia. Some CTC readers, however, prefer to use 2D images as the primary reading modality and, in experienced hands, the accuracies of either reading format are comparable. CTC interpretation consists of both antegrade and retrograde image review and there are a variety of viewing options such as the “filet” view or virtual dissection method which straightens and flattens the colon, enabling it to resemble an open pathologic specimen. Computed aided detection (CAD), a standard practice in other radiologic arenas, has also emerged as a viable option for CTC interpretation, but remains investigational and lacking in terms of convincing outcomes data [14]. A reporting system known as the CT colonography reporting and data system (C-RADS) classification has been developed to standardize the description and management recommendations of lesions found on CTC (Table 1) [15]. This classification scheme describes colonic lesions with respect to their shape, size, and number. C-RADS also provides recommendations for disposition based on the intracolonic findings. C-RADS also provides a scoring system

Colon (J Anderson, Section Editor)

Fig. 1. a 3-Dimensional CTC depiction of a 20 mm pedunculated cecal polyp. b 2-Dimensional CTC depiction of the same 20 mm pedunculated cecal polyp (blue arrow). Note the retained dependent residual colonic effluent tagged with oral contrast (white layer).

Fig. 2. a 3-Dimensional CTC depiction of a 8 mm pedunculated polyp. b 2-Dimensional CTC depiction of the same 8 mm sessile polyp (blue arrow). Note the adherent oral contrast on the dependent tip of the polyp.

Screening and Surveillance of Colorectal Cancer Using CTC

Kumar and Cash

Fig. 3. a 3-Dimensional depiction of a 9 mm sigmoid polyp in the prone position. b The same sigmoid polyp shown in the supine position. c The same sigmoid polyp using CTC tissue color density. d 2-Dimensional depiction of the same sigmoid polyp (blue arrow).

for extracolonic structures (Table 2). Most extracolonic abnormalities identified with CTC have no clinically relevant implications, so it is critical that CTC interpreters and clinicians ordering the test understand the C-RADS-reporting system in order to prevent the performance of low-yield additional evaluations that can impair the cost effectiveness of CTC. The extracolonic portion of CTC is typically interpreted by radiologists with body-imaging experience due to their experience interpreting cross-sectional imaging. It is important to recognize that due to the reduced doses of radiation used in CTC and the lack of intravenous contrast material during the study, extracolonic images obtained during CT are not of the same quality as dedicated abdominopelvic CT scans. Multiple studies have demonstrated that CTC has similar sensitivity compared with colonoscopy in CRC screening populations, especially for the detection of neoplastic lesions ≥10 mm in diameter. In one landmark study, Pickhardt et al. compared the performance of CTC to colonoscopy for the detection of colorectal neoplasia in 1233 asymptomatic patients undergoing CRC screening and showed a sensitivity of 93.8% for adenomatous polyps ≥10 mm in diameter and 88.7% for polyps ≥6 mm in diameter [16]. The

Colon (J Anderson, Section Editor) Table 1. C-RADS intracolonic CTC classification. C0. Inadequate study/awaiting prior comparisons • Inadequate prep: cannot exclude lesions 910 mm owing to presence of fluids/feces • Inadequate insufflations: one or more colonic segments collapsed on both views • Awaiting prior colon studies for comparison Cl. Normal colon or benign lesion; continue routine screeninga • No visible abnormalities of the colon • No polyp ≥6 mm • Lipoma or inverted diverticulum • Nonneoplastic findings— e.g., colonic diverticula C2. Intermediate polyp or intermediate findings; surveillance or colonoscopy recommendedb • Intermediate polyp 6–9 mm, G3 in number • Indeterminate findings, cannot exclude polyp 96 mm in technically adequate exam C3. Polyp, possibly advanced adenoma ; follow- up colonoscopy recommended • Polyp 910 mm • ≥3 polyps, each 6–9 mm C4. Colonic mass, likely malignant; surgical consultation recommendedc • Lesion compromises bowel lumen, demonstrates extracolonic invasion Reproduced by permission [15] a Every 5–10 years b Evidence suggests surveil lance can be delayed at least 3 years, subject to i ndi vi dual patient circumstance c Communicate to referring physician as per accepted guidelines for communication, such as ACR practice guideline for communication: Diagnostic radiology. Subject to local practice, endoscopic biopsy may be indicated

corresponding sensitivity of colonoscopy for these lesions was 87.5 and 92.3%, respectively. Other studies evaluating CTC as a CRC screening test have produced similar results to Pickhardt et al. [17]. In an effort to evaluate the performance of CTC outside of expert centers, a multicenter trial known as the National CT Colonography Trial was undertaken by the American College of Radiology Imaging Network (ACRIN). This study enrolled 2500 asymptomatic subjects and found CTC to have a sensitivity for polyps ≥10 mm of 90% [18]. Regge et al. published the results of another multicenter study with approximately 900 patients at increased risk for CRC and found CTC to have an overall negative predictive value of 96.3% compared to colonoscopy for the detection of advanced colorectal neoplasia [19]. CTC has also been compared with other CRC screening modalities and has shown promising results. Graser demonstrated CTC sensitivity to be not only comparable to colonoscopy for the detection of advanced adenomas, but also superior to FS, FIT, and FOBT [20].

CTC: Recent Research Over the last 12–18 months, several noteworthy publications have increased our knowledge regarding the use of CTC as a primary CRC screening modality.

Screening and Surveillance of Colorectal Cancer Using CTC

Kumar and Cash

Table 2. C-RADS extracolonic CTC classification E0. Limited exam. Compromised by artifact; evaluation of extracolonic soft tissues is severely limited. E1. Normal exam or anatomic variant. No extracolonic abnormalities visible. a. Anatomic variant; e.g., retroaortic left renal vein E2. Clinically unimprotant findings. No work—up indicated. Examples: a. Liver, kidney: simple cysts b. Gallbladder: cholelithiasis without cholecystitis c. Vertebra: hemangioma E3. Likely unimportant finding, incompletely characterized. Subject to local practice and patient preference, work—up may be indicated. Examples: a. Kidney: minimally complex or homogeneously hyperattenuating cyst E4. Potentially Important Finding. Communicate to referring physician as per accepted practice guidelines. a. Kidney: solid renal mass b. Lymphadenopathy c. Vasculature: aortic aneurysm d. Lung: non-uniformly calcified parenchymal nodule ≥1 cm Reproduced by permission [15]

Some of these have addressed issues related to CTC performance characteristics and accuracy, while others have focused on historical concerns associated with CTC, such as radiation exposure and the frequency and impact of extracolonic findings.

CTC and Screening Intervals Until recently, there was no data regarding the optimal screening intervals for CTC screening after an initial negative study. The American Cancer Society and MSTF recommend a 5-year screening interval after initial negative CTC, as does the C-RADS reporting system. However, these recommendations were not backed by definitive evidence. Pickhardt et al. recently reported both quantitative and qualitative data regarding colonic neoplasia detected at repeat CTC screening after initial negative CTC screening [21••]. In this retrospective observational study of 5640 initial negative CTC screening examinations (no polyps ≥6 mm) performed before 2010, 1429 (25.3%) patients returned for repeat CTC screening (mean interval 5.7 years ±0.9; range 4.5–10.7 years). One hundred seventy-three of these patients (12.1%) undergoing repeat CTC screening had at least one (206 total) polyp ≥6 mm detected on CTC. Placing this data in context, these investigators reported a prevalence of colon polyps ≥6 mm in 14.3% similar asymptomatic patients undergoing initial CTC screening in a previous report [22]. Importantly, 29.5% (61 of 207) of the polyps ≥6 mm that were noted on repeat screening were able to be identified on initial CTC images as polyps G6 mm and 12.6% (26 of 207) were missed on initial screening (i.e., identified on retrospective examination of initial CTC examinations). Large polyps, advanced neoplasia (advanced adenomas and cancer), and invasive cancer were seen in 3.8% (55 of 1429), 2.8% (40 of 1429), and 0.14% (2 of

Colon (J Anderson, Section Editor) 1429), respectively, at follow-up screening, compared with 5.2% (p = 0.02), 3.2% (p = 0.52), and 0.45% (p = 0.17), respectively, at initial screening. These investigators concluded that the lower rates of clinically important colorectal neoplasia at 5 years compared to initial screening supports currently recommended CTC screening intervals.

CTC and Radiation Exposure One of the frequently voiced concerns by critics of CTC is the unknown effects of radiation exposure. What is often lost in this argument is the fact that the American College of Radiology (ACR) has provided clear practice guidelines specifying use of low dose, non-enhanced multiple detector CT technique among asymptomatic individuals for CRC screening in order to maintain minimal radiation exposure with CTC [23]. Techniques such as automatic exposure control (AEC), which continuously adjusts the tube current according to patient volume, has been suggested as another way to decrease the radiation dose associated with CTC [24•]. CTC with these low dose protocols has decreased the radiation dose to ≤2 mSv and has shown equivalent polyp detection capability and excellent sensitivity and specificity in asymptomatic patients at average risk for CRC as well in patients with increased risk for CRC [20, 25–27]. While the exact risk of adverse outcomes with very low-dose medical radiation delivery such as that with CTC is not known and is thought to be incalculable, additional efforts should be made to minimize radiation dose. Nevertheless, Gonzales et al. estimated the risk of radiation-related cancer is about 0.05% from a single screen at age 60 with a low dose CTC protocol of 8 mSv for males and 9 mSv for females, and modeling studies have demonstrated that the potential benefit from CTC in preventing CRC and death exceeds the hypothetical risks of very low-dose radiation exposure [28]. It is likely that in the near future radiation doses associated with CTC will be below 1 mSv. One recent study assessed the effect of a low-tube-voltage technique and iterative reconstruction (IR) on the radiation dose and image quality of CTC. [29] This study showed a significantly higher contrast-to-noise ratio (CNR) and better visual scores on 2D and 3D CTC images with combined use of the low-tube-voltage (100 kVp) technique and IR. Another study by Lambert and colleagues showed comparable image quality and interpretation accuracy using ultra-low dosing via a variety of techniques, most notably iterative reconstruction, compared to standard dose (approximately 4 mSv) CTC [30]. Importantly, the detection of clinically unimportant extracolonic findings (C-RADS category E2) was lower with the low-dose CTC approach, while the detection of potentially significant extracolonic findings (C-RADS category E4) was unchanged.

CTC and Extracolonic Findings The cost-effectiveness of CTC has been repeatedly evaluated due to the cost of the test itself, the costs of colonoscopy and polypectomy performed in response to positive CTC, and the costs associated with additional evaluations resulting from detection of extracolonic abnormalities. On one hand, extracolonic imaging might be viewed as a desirable feature of CTC, however the vast majority of extracolonic lesions require no addition evaluation and defensive additional

Screening and Surveillance of Colorectal Cancer Using CTC

Kumar and Cash

diagnostic testing for unimportant abnormalities is thought by some to significantly degrade the cost-effectiveness of CTC. An early study of costs associated with CTC showed an 8.6% rate of extracolonic findings among a cohort of 2195 asymptomatic people resulting in an additional diagnostic cost of $98.56 USD per patient screened. [31] Pickhardt and colleagues recently published an analysis of extracolonic findings derived from a long-term CTC screening program affiliated with the University of Wisconsin. This analysis included 7952 asymptomatic individuals undergoing first time CTC screening over a period of 99 months and found a 9.1% prevalence of E3 findings (725/7952) [32••]. Consideration for further imaging was suggested in 83.8% (608/725) and 65 patients were lost to follow-up. Conditions requiring treatment or surveillance were ultimately diagnosed in 8.3% (55/660), including eight malignant neoplasms. In the remaining 605 patients, 25 (4.1%) underwent invasive biopsy or surgery to prove benignity and 278 (46.0%) received additional imaging follow-up. Pickhardt et al. also explored the prevalence and outcomes associated with potentially significant extracolonic findings (C-RADS category E4) in the Wisconsin CTC population [33] There were 2.5% (202/7952) patients with E4 findings, for which additional imaging (56%; 113/202) or clinical follow-up (44%; 89/ 202) were recommended. Twenty-two patients were lost to follow-up. Of the remaining 180 patients, 68% (123/180) had clinically significant disease, including 23% (42/180) with malignant or potentially malignant neoplasms and 32% (57/180) with abdominal aortic or other visceral artery aneurysms requiring treatment or surveillance. These findings highlight the importance of adherence to the C-RADS reporting system and using a judicious approach to the evaluation and management of unexpected extracolonic findings with attention to costs and potential risks and benefits of invasive diagnostic testing. A recent study by Plumb and colleagues evaluated the maximum rate of falsepositive diagnoses that patients and health care professionals were willing to accept in exchange for detection of extracolonic malignancy using CTC for CRC screening [34]. In this study, 52 patients and 50 health care professionals undertook two discrete choice experiments where they chose between unrestricted CTC that examined intra- and extracolonic organs or CTC restricted to the colon. They found that patients and health care professionals are willing to tolerate high rates of false-positive diagnoses with CTC in exchange for diagnosis of extracolonic malignancy. Perhaps not surprisingly, patients were much more willing than health care professionals to accept false-positive extracolonic findings on CTC that resulted in follow-up testing (9 99.8 vs 40% for radiologic follow-up and 999.8 vs 5% for invasive follow-up, both P G .001).

CTC and Diminutive Colonic Polyps Another controversial aspect of using CTC as a CRC screening modality is the standard convention of not reporting diminutive (G6 mm) polyps due to the lower specificity associated with these lesions on CTC. However, it is important to recognize that best estimates of the prevalence of advanced histology in diminutive lesions is about 1.7% with 0% prevalence of carcinoma [35]. The majority of polyps removed by gastroenterologists during colonoscopy are either small (6– 9 mm) or diminutive and a substantial proportion of these polyps are non-

Colon (J Anderson, Section Editor) neoplastic. CTC may offer a unique opportunity to gain additional information regarding the natural history of small and diminutive colorectal polyps. The results of a recent study by Nolthenius emphasized the correlation between polyp size and the potential of identifying advance neoplastic features [36••]. This study utilized a volumetric growth assessment to predict which small polyps were likely to become advanced adenomas with non-interventional surveillance. Similar to other natural history trials using CTC observation, [37] they found that the majority of 6–9 mm polyps do not progress to advanced neoplasia within 3 years and the lesions with advanced status can be identified based on their rate of volume increase on serial CTC examinations. These findings can be viewed as supporting the current practice of not reporting diminutive polyps seen on CTC as well as the recommended screening interval of 5 years.

CTC and Serrated Polyps Sessile serrated adenomas/polyps (SSP) have been the subject of significant investigation over the past decade as clinicians have recognized that these lesions follow an alternative pathway to develop into colorectal cancer compared to the historical adenoma-carcinoma sequence. These lesions are thought to be a leading cause of “interval cancers” that are discovered after a negative screening examination. Sessile serrated lesions have a predilection for the proximal colon, are often minimally protruding above the colonic mucosa, and are easily obscured by colonic mucous, making them more challenging to identify compared to their adenomatous counterparts. Thus, any recommended CRC screening method should have the capacity to accurately detect these lesions and the ability of CTC to detect SSP is an important consideration that requires additional evidence. One study of screening CTC found the prevalence of non-diminutive serrated lesions (≥6 mm) to be 3.1%, similar to prevalence data from colonoscopy trials evaluating SSP detection [38]. However, direct comparison data between the two modalities is scant. Ijspeert et al. recently compared CTC with colonoscopy with respect to detection of high-risk SSP (SSP ≥10 mm or SSP with dysplasia) [39]. Among 1276 participants undergoing colonoscopy, 4.3% had ≥1 high-risk SSP versus 0.8% of 982 undergoing CTC. The colonoscopy group also had more flat, high-risk SSP overall, more high-risk SSP in the proximal colon, and more SSP with dysplasia detected than CTC. One of the explanations offered for the above findings was the use of a reduced bowel preparation with CTC (consisting of only one iodinated agent). Nevertheless, larger prospective comparisons of CTC and colonoscopy accuracy for the detection of SSP are needed and this issue remains a concern. Until CTC is demonstrated to be as accurate as colonoscopy for identifying SSP, retaining a shorter interval for screening examinations is a prudent approach.

CTC Safety CTC is considered a remarkably safe modality for CRC screening. Published perforation rates associated with screening and diagnostic CTC suggest a range of 0.005 to 0.03%, [40–42] lower than published perforation rates associated with diagnostic and therapeutic colonoscopy (0.1–0.2%) [43–49]. Most perforations associated with CTC have occurred in patients with high-risk occlusive

Screening and Surveillance of Colorectal Cancer Using CTC

Kumar and Cash

lesions undergoing diagnostic studies. Replacement of room-air manual insufflation of the colon with volume and pressure regulated automated CO2 insufflation has likely contributed to the reduced perforation risk with CTC. Among individuals who qualify for CTC due to incomplete colonoscopy, concern has been voiced regarding the safety of performing CTC on the same day as the incomplete endoscopic examination, especially if there was any tissue sampling performed during colonoscopy. Lara et al. recently published a retrospective analysis of 6260 patients undergoing colonoscopy, 198 (3.1%) of whom underwent same-day CTC due to incomplete colonoscopy [50••]. Thirty-four (17%) of these patients had undergone a total of 72 polypectomies immediately prior to same-day CTC. In this study, there were no reported complications or perforations during both short- and long-term follow-up of all patients undergoing same-day CTC. The majority of polyps removed prior to same-day CTC in this study were diminutive or small, with only 5 (7%) polyps ≥10 mm removed prior to colonoscopy, and nearly 75% of the polypectomies prior to same-day CTC were from the left colon. While this report is reassuring, more data is required regarding the safety of same-day colonoscopy after incomplete colonoscopy with polypectomy.

Economics of Screening CTC CTC has been demonstrated to be a cost-effective CRC screening test in multiple cost-effective analyses using a variety of different models, cost and outcomes estimates, and performance variables [51–56]. A recently reported analysis by Pyenson et al. compared the cost-effectiveness of CTC to colonoscopy and predicted a cost of $4.0 billion for CTC in all Medicare enrollees compared with $5.7 billion for colonoscopy, using an average CTC screening cost of $439 and an average screening colonoscopy cost of $1036. [57]. In the past 2 decades, CTC advances in the technical aspects of the performance of CTC and continued accumulation of convincing data related to its accuracy for the identification of advanced colonic neoplasia in the screening setting has generated enthusiasm for its use as an alternative to colonoscopy. Numerous insurers currently cover screening CTC, [58] however, the Centers for Medicare and Medicaid Services (CMS)-mandated coverage of CTC screening for Medicare beneficiaries remains lacking. Multiple patient advocacy groups and professional societies such as the ACR and the American Gastroenterological Association have advocated for CMS coverage of CTC screening in the Medicare population, either outright or via a trial program and published data from several sources have shown that the accuracy of CTC screening in Medicare aged patients is similar to that observed in younger patients undergoing CTC screening [59–61]. It will be interesting to see whether the most recent statement of the United States Preventative Services Task Force (USPSTF) on CRC screening, which included CTC as an acceptable screening option, will have any effect on future CMS coverage determinations. [9••].

CTC for Surveillance Because of the strong association of colorectal polyps with subsequent development of colorectal neoplasia, they are removed during

Colon (J Anderson, Section Editor) colonoscopy and patients are recommended to undergo surveillance colonoscopy at future intervals, based on the size, number, and histology of resected lesions [62–65]. In clinical practice, all polyps are routinely resected, even though there is low probability of polyps G10 mm harboring advanced histology [66, 67]. An important area of debate is whether CTC can used to observe small polyps 6–9 mm in size. Stoop and colleagues published the results of a randomized trial of CTC vs colonoscopy for screening in 8844 patients (COCOS trial) [68]. In this trial, it was shown that CTC and colonoscopy had similar yields for the detection of advanced neoplasia, based on higher participation in the CTC group and higher procedural yield in the colonoscopy group. However, in the COCOS trial, only CTC participants with lesions ≥10 mm were referred for colonoscopy and participants with a largest lesion between 6 and 9 mm were recommended for surveillance CTC. Nolthenius et al. recently reported the outcomes observed in the prospective evaluation of patients from the COCOS trial who were included in the CTC surveillance arm, specifically evaluating the yield of surveillance CTC for advanced neoplasia, as well as the total yield of CTC screening [69]. Eighty-two patients comprised the CTC surveillance group. Thirteen patients bypassed surveillance CTC and had an early colonoscopy at a mean of 54 weeks after the index CTC. Nine of these patients (69.2%) were found to have advanced neoplasia, defined as either an advanced adenoma or invasive CRC. Among the 56 patients who ultimately completed CTC surveillance, the mean time of surveillance was 3.4 years (+/− 0.43) and 33 underwent colonoscopy based on the persistence of at least one polyp ≥6 mm. Twenty patients had no polyps ≥6 mm found on surveillance CTC and evidence of polyp regression was identified in 50%. Fifteen of the 33 completing colonoscopy after surveillance CTC were found to have advanced neoplasia. Thus, when the primary outcome of the COCOS trial was considered with that of the surveillance arm of the trial, the diagnostic yield for advanced neoplasia per 100 participants was 8.6 with CTC. Importantly, 90% of the advanced neoplasia was ≥10 mm and 30% had tubulovillous histology. No CRC was identified in the surveillance group. In addition, surveillance CTC found significantly lower extracolonic findings during surveillance compared with primary CTC screening examinations (1.8% versus 4.4–11%), tempering concern regarding accumulation of additional ancillary evaluations associated with serial CTC examination. In addition to their examination of the yield of surveillance CTC, Nolthenius et al. evaluated the perceived burden of waiting for surveillance CTC using validated questionnaires in 62 patients and found that 73% rated the experience of waiting for surveillance CTC as never or only sometimes burdensome. [70] They also found that the burden of index and surveillance CTC were similar, and that the majority of perceived burden associated with surveillance CTC was derived from waiting for results, rather than for the procedure itself. These findings, while deserving validation in a larger cohort and comparison to surveillance colonoscopy, suggest that surveillance CTC would be accepted by patients if placed into practice.

Screening and Surveillance of Colorectal Cancer Using CTC

Kumar and Cash

Among patients with established CRC who undergo surgical resection, intense follow-up surveillance is critical as it has been shown to confer a survival benefit [71–73]. Colonoscopy with standard CT imaging has been the routine practice for these patients. Recently, Porte et al. published a systematic review of the performance characteristics of surveillance CTC compared to colonoscopy and standard CT imaging in patients with a previous diagnosis of CRC [74]. These investigators identified 7 studies providing data on 880 patients that demonstrated CTC had a sensitivity for the detection of anastomotic recurrence of 95% (95% CI: 62–100) with a specificity of 100% (95% CI: 75–100). The sensitivity of CTC detection of metachronous CRC was 100%. Moreover, they concluded that CTC as a single test alternative to colonoscopy and standard CT imaging of extraintestinal organs could result in cost savings exceeding €20 million for an annual cohort of UK CRC patients. While the concept of using a single CTC in lieu of colonoscopy and standard CT imaging for CRC surveillance is attractive, additional analysis using US data and factoring in the potential effects of increased radiation doses required to adequately visualize extraintestinal organs is needed before this practice can be widely advocated.

Conclusion The essential objective behind any CRC screening program is to prevent the disease by identifying it at early, curable, or even preventative stage. Any screening modality that is accurate, cost-effective, less invasive than standard alternatives, and acceptable to patients and payers has the potential to improve the suboptimal CRC screening rates currently observed in the USA. Based on the accumulated data over the last 2 decades, CTC appears to fulfill these requirements and has the potential to improve outcomes through greater screening utilization, reapportioning scarce endoscopic resources towards higher yield, diagnostic and therapeutic procedures, and less onerous surveillance protocols for patients with a history of CRC. Additional refinement of CTC protocols with ultra-low radiation doses and enhanced detection using CAD are needed, as is outcomes data reflecting best practices in terms of integrating CTC into current CRC programs and the ability of CTC to accurately identify flat and sessile lesions. Nevertheless, with the recent USPSTF tacit endorsement of CTC as an acceptable CRC screening modality, CTC is poised to become an important player in the CRC screening and surveillance arena.

Compliance with Ethical Standards Conflict of Interest Manoj Kumar and Brooks D. Cash each declare no potential conflicts of interest. Human and Animal Rights and Informed Consent This article does not contain any studies with human or animal subjects performed by any of the authors.

Colon (J Anderson, Section Editor)

References and Recommended Reading Papers of particular interest, published recently, have been highlighted as: • Of importance; •• Of major importance 1.

Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A, et al. Global cancer statistics, 2012. CA Cancer J Clin. 2015;65:87–108. 2. Balanced Budget Act of 1997, Pub. L. No. 105–33; § 4104 (1997). 3. Siegel RL, Ward EM, Jemal A. Trends in colorectal cancer incidence rates in the United States by tumor location and stage, 1992-2008. Cancer Epidemiol Biomark Prev. 2012;21:411–6. 4. Edwards BK, Ward E, Kohler BA, Eheman C, Zauber AG, Anderson RN, et al. Annual report to the nation on the status of cancer, 1975-2006, featuring colorectal cancer trends and impact of interventions (risk factors, screening, and treatment) to reduce future rates. Cancer. 2010;116:544–73. 5. Winawer S, Fletcher R, Rex D, Bond J, Burt R, Ferrucci J, et al. Colorectal cancer screening and surveillance: clinical guidelines and rationale update based on new evidence. Gastroenterology. 2003;124:544–60. 6. http://www.cdc.gov/media/releases/2013/p1105colorectal-cancer-screening.html.(2016, August 31) 7. Taylor DP, Cannon-Albright LA, Sweeney C, Williams MS, Haug PJ, Mitchell JA, et al. Comparison of compliance for colorectal cancer screening and surveillance by colonoscopy based on risk. Genet Med. 2011;13:737–43. 8. Levin B, Lieberman DA, McFarland B, Andrews KS, Brooks D, Bond J, et al. Screening and surveillance for the early detection of colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology. Gastroenterology. 2008;134:1570–95. 9.•• US Preventive Services Task Force. Screening for colorectal cancer. JAMA. 2016;315:2564–75. Most recent update of USPSTF position on CRC screening and CRC screening tests 10. Meyerhardt JA, Mangu PB, Flynn PJ, Korde L, Loprinzi CL, Minski BD, 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. J Clin Oncol. 2013;31:4465–70. 11. Vining DJ, Gelfand DW, Bechtold RE, Scharling ES, Grishaw EK, Shifrin RY. Technical feasibility of colon imaging with helical CT and virtual reality. AJR Am J Roentgenol. 1994;162(Suppl):104. 12. Sato K, Tanaka T, Sato J, Shibata E, Nagai Y, Murono K, et al. Usefulness of preoperative CT colonography for colon cancer. Asian J Surg. 2016; doi:10.1016/j.asjsur. 2016.04.002.

13.

Ridereau-Zins C, Pilleul F, Gandon Y, Laurant V, et al. CT colonography: why? when? how? Diagn Interv Imaging. 2012;93:2–9. 14. Halligan S, Altman DG, Mallett S, Taylor SA, Burling D, Roddie M, et al. Computed tomographic colonography: assessment of radiologist performance with and without computer-aided detection. Gastroenterology. 2006;131:1690–9. 15. Zalis ME, Barish MA, Choi JR, Dachman AH, Fenlon HM, Ferucci JT, et al. CT colonography reporting and data system: a consensus proposal. Radiology. 2005;236:3–9. 16. Pickhardt PJ, Choi JR, Hwang I, Butler JA, Puckett ML, Hildebrandt HA, Wong RK, et al. Computed tomographic virtual colonoscopy for colorectal neoplasia in asymptomatic adults. N Engl J Med. 2003;349:2191–200. 17. Johnson CD, Harmsen WS, Wilson LA, Maccarty RL, Welch TJ, Ilstrup DM, et al. Prospective blinded evaluation of computed tomographic colonography for screen detection of colorectal polyps. Gastroenterology. 2003;125:311–9. 18. Johnson CD, Chen MH, Toledano AY, Heiken JP, Dachman A, Kuo MD, et al. Accuracy of CT colonography for detection of large adenomas and cancers. N Engl J Med. 2008;359:1207–17. 19. Regge D, Laudi C, Galatola G, Della Monica P, Bonelli L, Angelelli Get al. Diagnostic accuracy of computed tomographic colonography for the detection of advanced neoplasia in individuals at increased risk of colorectal cancer. JAMA 2009;301(23):2453-2461. 20. Graser A, Stieber P, Nagel D, Schäfer C, Horst D, Becker CR et al. Comparison of CT colonography, colonoscopy, sigmoidoscopy and faecal occult blood tests for the detection of advanced adenoma in an average risk population. Gut 2009;58(2):241-248. 21.•• Pickhardt PJ, Pooler BD, Mbah I, Weiss JM, Kim DH. Colorectal findings at repeat CT colonography screening after initial CT colonography screening negative for polyps larger than 5 mm. Radiology. 2016;22:160582. First large report of results evaluating serial CTC screening examinations 22. Kim DH, Pickhardt PJ, Taylor AJ, Leung WK, Winter TC, Hinshaw JL, et al. CT colonography versus colonoscopy for the detection of advanced neoplasia. N Engl J Med. 2007;357(14):1403–12. 23. American College of Radiology. ACR practice guideline for the performance of computed tomography (CT) colonography in adults. Practice Guidelines and Technical Standards; Reston, VA: 2009. 24.• Nagata K, Fujiwara M, Kanazawa H, Mogi T, Iida N, Mitsushima T, et al. Evaluation of dose reduction and image quality in CT colonography: comparison of low-

Screening and Surveillance of Colorectal Cancer Using CTC dose CT with iterative reconstruction and routine-dose CT with filtered back projection. Eur Radiol. 2015;25:221–9. Important article showing the feasibility of reduced radiation doses without compromised image quality of CTC 25. Van Gelder RE, Venema HW, Serlie IW, Nio CY, Determann RM, Tipker CA, et al. CT colonography at different radiation dose levels: feasibility of dose reduction. Radiology. 2002;224:25–33. 26. Macari M, Bini EJ, Xue X, Milano A, Katz SS, Resnick D, et al. Colorectal neoplasms: prospective comparison of thin-section low-dose multi-detector row CT colonography and conventional colonoscopy for detection. Radiology. 2002;224:383–92. 27. Iannaccone R, Laghi A, Catalano C, Brink JA, Mangiapane F, Trenna S, et al. Detection of colorectal lesions: lower-dose multi-detector row helical CT colonography compared with conventional colonoscopy. Radiology. 2003;229:775–81. 28. Berrington de Gonzalez A, Kim KP, Yee J. CT colonography: perforation rates and potential radiation risks. Gastrointest Endosc Clin N Am. 2010 Apr;20 (2):279–91. 29. Yamamura S, Oda S, Imuta M, Utsunomiya D, Yoshida M, Namimoto T, et al. Reducing the radiation dose for CT colonography. Acad Radiol. 2016;23:155–62. 30. Lambert L, Ourednicek P, Briza J, Giepmans W, Jahoda J, Hruska L, et al. Sub-milliSievert ultralow-dose CT colonography with iterative model reconstruction technique. PeerJ. 4:e1883. doi:10.7717/peerj.1883. 31. Pickhardt PJ, Hanson ME, Vanness DJ, Lo JY, Kim DH, Taylor AJ, et al. Unsuspected extracolonic findings at screening CT colonography: clinical and economic impact. Radiology. 2008;249:151–9. 32.•• Pooler BD, Kim DH, Pickhardt PJ. Indeterminate but likely unimportant extracolonic findings at screening CT colonography (C-RADS Category E3): Incidence and outcomes data from a clinical screening program. Am J Roentgenol. 2016;9:1–6. Important real-life experience showing low rates and acuity of extracolonic findings over time with surveillance CTC 33. Pooler BD, Kim DH, Pickhardt PJ. Potentially important extracolonic findings at screening CT colonography: incidence and outcomes data from a clinical screening program. AJR Am J Roentgenol. 2016;206:313–8. 34. Plumb AA, Boone D, Fitzke H, Helbren E, Mallett S, Zhu S, et al. Detection of extracolonic pathologic findings with CT colonography: a discrete choice experiment of perceived benefits versus harms. Radiology. 2014;273:144–52. 35. Butterly LF, Chase MP, Pohl H, Fiarman GS. Prevalence of clinically important histology in small adenomas. Clin Gastroenterol Hepatol. 2006;4:343–8. 36.•• Tutein Nolthenius CJ, Boellaard TN, de Haan MC, Nio CY, Thomeer MG, Bipat S, et al. Evolution of screendetected small (6-9 mm) polyps after a 3-year surveillance interval: assessment of growth with CT colonography

Kumar and Cash

compared with histopathology. Am J Gastroenterol. 2015;110:1682–90. Landmark article evaluating the natural history of small polyps left in vivo 37. Pickhardt PJ, Kim DH, Pooler BD, Hinshaw JL, Barlow D, Jensen D, et al. Assessment of volumetric growth rates of small colorectal polyps with CT colonography: a longitudinal study of natural history. Lancet Oncol. 2013;14:711–20. 38. Kim DH, Matkowskyj KA, Lubner MG, Hinshaw JL, Munoz Del Rio A, Pooler BD, et al. Serrated polyps at CT colonography: prevalence and characteristics of the serrated polyp spectrum. Radiology. 2016;280:455–63. 39. IJspeert JE, Tutein Nolthenius CJ, Kuipers EJ, van Leerdam ME, Nio CY, Thomeer MG, et al. CTcolonography vs. colonoscopy for detection of highrisk sessile serrated polyps. Am J Gastroenterol. 2016;111:516–22. 40. Burling D, Halligan S, Slater A, Noakes MJ, Taylor SA. Potentially adverse events at CT colonography in symptomatic patients: national survey of the United Kingdom. Radiology. 2006;239:464–71. 41. Sosna J, Blachar A, Amitai M, Barmeir E, Peled N. Goldberg SNet, al. Colonic perforation at CT colonography: assessment of risk in a multicenter large cohort. Radiology. 2006;239:457–63. 42. Pickhardt PJ. Incidence of colonic perforation at CT colonography: review of existing data and implications for screening of asymptomatic adults. Radiology. 2006;239:313–6. 43. Waye JD, Lewis BS, Yessayan S. Colonoscopy: a prospective report of complications. J Clin Gastroenterol. 1992;15:347–51. 44. Anderson ML, Pasha TM, Leighton JA. Endoscopic perforation of the colon: lessons from a 10-year study. Am J Gastroenterol. 2000;95:3418–22. 45. Sieg A, Hachmoeller-Eisenbach U, Eisenbach T. Prospective evaluation of complications in outpatient GI endoscopy: a survey among German gastroenterologists. Gastrointest Endosc. 2001;53:620–7. 46. Tran DQ, Rosen L, Kim R, Riether RD, Stasik JJ, Khubchandani IT. Actual colonoscopy: what are the risks of perforation? Am J Surg. 2001;67:845– 7. 47. Korman LY, Overholt BF, Box T, Winker CK. Perforation during colonoscopy in endoscopic ambulatory surgical centers. Gastrointest Endosc. 2003;58:554–7. 48. Gatto NM, Frucht H, Sundararajan V, Jacobson JS, Grann VR, Neugut AI. Risk of perforation after colonoscopy and sigmoidoscopy: a population-based study. J Natl Cancer Inst. 2003;95:230–6. 49. Bowles CJ, Leicester R, Romaya C, Swarbrick E, Williams CB, Epstein O. A prospective study of colonoscopy practice in the UK today: are we adequately prepared for national colorectal cancer screening tomorrow? Gut. 2004;53:277–83. 50.•• Lara LF, Avalos D, Huynh H, Jimenez-Cantisano B, Padron M, Pimentel R, et al. The safety of same-day CT colonography following incomplete colonoscopy with

Colon (J Anderson, Section Editor) polypectomy. United European Gastroenterol J. 2015;3:358–63. Very nice study demonstrating safety of same day CTC, even after mucosal biopsy during incomplete colonoscopy 51. Gomes M, Aldridge RW, Wylie P, Bell J, Epstein O. Cost-effectiveness analysis of 3-D computerized tomography colonography versus optical colonoscopy for imaging symptomatic gastroenterology patients. Appl Health Econ Health Policy. 2013;11:107–17. 52. Hanly P, Skally M, Fenlon H, Sharp L. Costeffectiveness of computed tomography colonography in colorectal cancer screening: a systematic review. Int J Technol Assess Health Care. 2012;28:415–23. 53. Hassan C, Pickhardt PJ. Cost-effectiveness of CT colonography. Radiol Clin N Am. 2013;51:89–97. 54. Kriza C, Emmert M, Wahlster P, Niederländer C, Kolominsky-Rabas P. An international review of the main cost-effectiveness drivers of virtual colonography versus conventional colonoscopy for colorectal cancer screening: is the tide changing due to adherence? Eur J Radiol. 2013;82:629–36. 55. Lucidarme O, Cadi M, Berger G, Taieb J, Poynard T, Grenier P, et al. Cost-effectiveness modeling of colorectal cancer: computed tomography colonography vs colonoscopy or fecal occult blood tests. Eur J Radiol. 2012;8:1413–9. 56. Pickhardt PJ. CT colonography: does it satisfy the necessary criteria for a colorectal screening test? Expert Rev Gastroenterol Hepatol. 2014;8:211–3. 57. Pyenson B, Pickhardt PJ, Sawhney TG, Berrios M. Medicare cost of colorectal cancer screening: CT colonography vs. optical colonoscopy. Abdom Imaging. 2015;40:2966–76. 58. http://www.acr.org/About-Us/Media-Center/PressReleases/2015-Press-Releases/20151111-Patient-andProvider-Groups-Call-on-Congress-to-Pass-MedicareVirtual-Colonoscopy-Coverage (2016, August 31) 59. Kim DH, Pickhardt PJ, Hanson ME, Hinshaw JL. CT colonography: performance and program outcome measures in an older screening population. Radiology. 2010;254(2):493–500. 60. Cash BD, Stamps K, McFarland EG, Spiegel AR, Wade SW. Clinical use of CT colonography for colorectal cancer screening in military training facilities and potential impact on HEDIS measures. J Am Coll Radiol. 2013;10:30–6. 61. Macari M, Nevsky G, Bonavita J, Kim DC, Megibow AJ, Babb JS. CT colonography in senior versus nonsenior patients: extracolonic findings, recommendations for additional imaging, and polyp prevalence. Radiology. 2011;259:767–74. 62. Zauber AG, Winawer SJ, O’Brien MJ, LansdorpVogelaar I, van Ballegooijen M, Hankey BF, et al. Colonoscopic polypectomy and long-term prevention of colorectal-cancer deaths. N Engl J Med. 2012;366:687–96.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72. 73.

74.

Lieberman DA, Weiss DG, Harford WV, Ahnen DJ, Provenzale D, Sontag SJ, et al. Five-year colon surveillance after screening colonoscopy. Gastroenterology. 2007;133:1077–85. Schoenfeld P, Cash B, Flood A, Dobhan R, Eastone J, Coyle W, et al. Colonoscopic screening of average-risk women for colorectal neoplasia. N Engl J Med. 2005;352:2061–8. Lieberman DA, Rex DK, Winawer SJ, Giardiello FM, Johnson DA, Levin TR. Guidelines for colonoscopy surveillance after screening and polypectomy: a consensus update by the US Multi-Society Task Force on Colorectal Cancer. Gastroenterology. 2012;143:844– 57. Lieberman D, Moravec M, Holub J, Michaels L, Eisen G. Polyp size and advanced histology in patients undergoing colonoscopy screening: implications for CT colonography. Gastroenterology. 2008;135:1100–5. Kim DH, Pickhardt PJ, Taylor AJ. Characteristics of advanced adenomas detected at CT colonographic screening: implications for appropriate polyp size thresholds for polypectomy versus surveillance. AJR Am J Roentgenol. 2007;188:940–4. Stoop EM, de Haan MC, de Wijkerslooth TR, Bossuyt PM, van Ballegooijen M, Nio CY, et al. Participation and yield of colonoscopy versus non-cathartic CT colonography in population-based screening for colorectal cancer: a randomised controlled trial. Lancet Oncol. 2012;13:55–64. Tutein Nolthenius CJ, Boellaard TN, de Haan MC, Nio CY, Thomeer MG, Bipat S, et al. Computer tomography colonography participation and yield in patients under surveillance for 6-9 mm polyps in a population-based screening trial. Eur Radiol. 2016;26:2762–70. Tutein Nolthenius CJ, Boellaard TN, de Haan MC, Nio CY, Thomeer MG, Bipat S, et al. Burden of waiting for surveillance CT colonography in patients with screendetected 6-9 mm polyps. Eur Radiol. 2016; doi:10. 1007/s00330-016-4251-4. Renehan AG, Egger M, Saunders MP, O’Dwyer ST. Impact on survival of intensive follow up after curative resection for colorectal cancer: systematic review and meta-analysis of randomised trials. BMJ. 2002;324:813. Jeffery M, Hickey B, Hider P. Follow up strategies for patients treated for non-metastatic colorectal cancer. Cochrane Database Syst Rev. 2007;1:CD002200. Primrose JN, Perera R, Gray A, Rose P, Fuller A, Corkhill 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. Porté F, Uppara M, Malietzis G, Faiz O, Halligan S, Athanasiou T, et al. CT colonography for surveillance of patients with colorectal cancer: Systematic review and meta-analysis of diagnostic efficacy. Eur Radiol. 2016; doi:10.1007/s00330-016-4319-1.