Combined artificial intelligence and radiologist model for predicting rectal cancer treatment response from magnetic resonance imaging: an external validation study.

PURPOSE: To evaluate an MRI-based radiomic texture classifier alone and combined with radiologist qualitative assessment in predicting pathological complete response (pCR) using restaging MRI with internal training and external validation. METHODS: Consecutive patients with locally advanced rectal cancer (LARC) who underwent neoadjuvant therapy followed by total mesorectal excision from March 2012 to February 2016 (Memorial Sloan Kettering Cancer Center/internal dataset, n = 114, 41% female, median age = 55) and July 2014 to October 2015 (Instituto do Câncer do Estado de São Paulo/external dataset, n = 50, 52% female, median age = 64.5) were retrospectively included. Two radiologists (R1, senior; R2, junior) independently evaluated restaging MRI, classifying patients (radiological complete response vs radiological partial response). Model A (n = 33 texture features), model B (n = 91 features including texture, shape, and edge features), and two combination models (model A + B + R1, model A + B + R2) were constructed. Pathology served as the reference standard for neoadjuvant treatment response. Comparison of the classifiers' AUCs on the external set was done using DeLong's test. RESULTS: Models A and B had similar discriminative ability (P = 0.3; Model B AUC = 83%, 95% CI 70%-97%). Combined models increased inter-reader agreement compared with radiologist-only interpretation (κ = 0.82, 95% CI 0.70-0.89 vs k = 0.25, 95% CI 0.11-0.61). The combined model slightly increased junior radiologist specificity, positive predictive value, and negative predictive values (93% vs 90%, 57% vs 50%, and 91% vs 90%, respectively). CONCLUSION: We developed and externally validated a combined model using radiomics and radiologist qualitative assessment, which improved inter-reader agreement and slightly increased the diagnostic performance of the junior radiologist in predicting pCR after neoadjuvant treatment in patients with LARC.

Pretreatment colostomy in patients with anal squamous cell carcinoma: Risk factors for a permanent stoma.

BACKGROUND: The current standard of care for anal squamous cell carcinoma (SCC) is concurrent chemoradiation (CRT), which enables tumor eradication while preserving the anal sphincter. Patients with locally advanced tumors, however, may experience complications that preclude treatment before stoma creation. OBJECTIVE: To evaluate the reversal rate of pretreatment stomas and the risk factors associated with nonreversal. METHODS: This single-institution retrospective cohort study using a prospective database included patients diagnosed with anal SCC from January 2008 to December 2020 who required a stoma before curative CRT. RESULTS: In total, 651 patients were identified; 65 required a stoma before chemoradiation due to obstruction (43.1%), rectovaginal fistula (20%), and perianal sepsis (36.9%). The stoma was reversed in nine patients after a mean follow-up of 35.8 months. Risk factors associated with a permanent stoma were perianal sepsis (p = 0.010), interruptions during radiotherapy for more than 7 days (p = 0.010), male sex (p = 0.013), poor performance status (Eastern Cooperative Oncology Group [ECOG] ≥ 2) (p = 0.023), large tumors (p = 0.045), and cisplatin-based chemotherapy (p = 0.047). CONCLUSIONS: Pretreatment stomas are unlikely to be reversed, and risk factors for a permanent stoma are perianal sepsis, interruptions during radiotherapy for more than 7 days, male sex, poor performance status (ECOG ≥ 2), large tumors, and cisplatin-based chemotherapy.

Outcomes of Patients With Local Regrowth After Nonoperative Management of Rectal Cancer After Neoadjuvant Chemoradiotherapy.

BACKGROUND: Clinical complete responders after chemoradiation for rectal cancer are increasingly being managed by a watch-and-wait strategy. Nonetheless, a significant proportion will experience a local regrowth, and the long-term oncological outcomes of these patients is not totally known. OBJECTIVE: The purpose of this study was to analyze the outcomes of patients who submitted to a watch-and-wait strategy and developed a local regrowth, and to compare these results with sustained complete clinical responders. DESIGN: This was a retrospective study. SETTING: Single institution, tertiary cancer center involved in alternatives to organ preservation. PATIENTS: Patients with a biopsy-proven rectal adenocarcinoma (stage II/III or low lying cT2N0M0 at risk for an abdominoperineal resection) treated with chemoradiation who were found at restage to have a clinical complete response. INTERVENTIONS: Rectal cancer patients treated with chemoradiation who underwent a watch-and-wait strategy (without a full thickness local excision) and developed a local regrowth were compared to the remaining patients of the watch-and-wait strategy. MAIN OUTCOME MEASURES: Overall survival between groups, incidence of regrowth' and results of salvage surgery. RESULTS: There were 67 patients. Local regrowth occurred in 20 (29.9%) patients treated with a watch-and-wait strategy. Mean follow-up was 62.7 months. Regrowth occurred at mean 14.2 months after chemoradiation, half of them within the first 12 months. Patients presented with comparable initial staging, lateral pelvic lymph-node metastasis, and extramural venous invasion. The regrowth group had a statistically nonsignificant higher incidence of mesorectal fascia involvement (35.0% vs 13.3%, p = 0.089). All regrowths underwent salvage surgery, mostly (75%) a sphincter-sparing procedure. 5-year overall survival was 71.1% in patients with regrowth and 91.1% in patients with a sustained complete clinical response (p = 0.027). LIMITATIONS: This study was limited by its retrospective evaluation of patient selection for a watch-and-wait strategy and outcomes, as well as its small sample size. CONCLUSIONS: Local regrowth is a frequent event when following a watch-and-wait policy (29.9%); however, patients could undergo salvage surgical treatment with adequate pelvic control. In this series, overall survival showed a statistically significant difference from patients managed with a watch-and-wait strategy who experienced a local regrowth compared to those who did not. See Video Abstract at http://links.lww.com/DCR/B773.RESULTADOS DE LOS PACIENTES CON REBROTE LOCAL, DESPUÉS DEL MANEJO NO QUIRÚRGICO DEL CÁNCER DE RECTO, DESPUÉS DE LA QUIMIORRADIOTERAPIA NEOADYUVANTEANTECEDENTES:Los respondedores clínicos completos, después de la quimiorradiación para el cáncer de recto, se tratan cada vez más mediante una estrategia de observación y espera. No obstante, una proporción significativa experimentará un rebrote local y los resultados oncológicos a largo plazo de estos pacientes, no se conocen por completo.OBJETIVO:El propósito de este estudio, fue analizar los resultados de los pacientes sometidos a una estrategia de observación y espera, que desarrollaron un rebrote local, y comparar estos resultados con respondedores clínicos completos sostenidos.DISEÑO:Este fue un estudio retrospectivo.ENTORNO CLINICO.Institución única, centro oncológico terciario involucrado en alternativas a la preservación de órganos.PACIENTES:Pacientes con un adenocarcinoma de recto comprobado por biopsia (estadio II / III o posición baja cT2N0M0, en riesgo de resección abdominoperineal), tratados con quimiorradiación, y que durante un reestadiaje, presentaron una respuesta clínica completa.INTERVENCIONES:Los pacientes con cáncer de recto tratados con quimiorradiación, sometidos a una estrategia de observación y espera (sin una escisión local de espesor total) y que desarrollaron un rebrote local, se compararon con los pacientes restantes de la estrategia de observación y espera.PRINCIPALES MEDIDAS DE VALORACION:Supervivencia global entre los grupos, incidencia de rebrote y resultados de la cirugía de rescate.RESULTADOS:Fueron 67 pacientes. El rebrote local ocurrió en 20 (29,9%) pacientes tratados con una estrategia de observación y espera. El seguimiento medio fue de 62,7 meses. El rebrote se produjo a la media de 14,2 meses después de la quimiorradiación, la mitad de ellos dentro de los primeros 12 meses. Los pacientes se presentaron con una estadificación inicial comparable, metástasis en los ganglios linfáticos pélvicos laterales e invasión venosa extramural. El grupo de rebrote tuvo una mayor incidencia estadísticamente no significativa de afectación de la fascia mesorrectal (35,0 vs 13,3%, p = 0,089). Todos los rebrotes se sometieron a cirugía de rescate, en su mayoría (75%) con procedimiento de preservación del esfínter. La supervivencia global a 5 años fue del 71,1% en pacientes con rebrote y del 91,1% en pacientes con una respuesta clínica completa sostenida (p = 0,027).LIMITACIONES:Evaluación retrospectiva de la selección de pacientes para una estrategia y resultados de observar y esperar, tamaño de muestra pequeño.CONCLUSIONES:El rebrote local es un evento frecuente después de la política de observación y espera (29,9%), sin embargo los pacientes podrían someterse a un tratamiento quirúrgico de rescate con un adecuado control pélvico. En esta serie, la supervivencia global mostró una diferencia estadísticamente significativa de los pacientes manejados con una estrategia de observación y espera que experimentaron un rebrote local, en comparación con los que no lo hicieron. Consulte Video Resumen en http://links.lww.com/DCR/B773. (Traducción-Dr. Fidel Ruiz Healy).

High-risk human papillomavirus test in anal smears: can it optimize the screening for anal cancer?

OBJECTIVE: The current method for screening anal cancer is anal cytology, which has low sensitivity. Since high-risk human papillomavirus (HR-HPV) is associated with almost 90% of cases of anal cancer, the objective of this study is to evaluate whether testing for HR-HPV can optimize the screening. DESIGN: Prospective study with patients enrolled in a screening program for anal dysplasia. Considering high-resolution anoscopy (HRA)-guided biopsy as the gold standard for diagnosis of high-grade squamous intraepithelial lesions, the diagnostic performance of anal cytology, HR-HPV testing, and the combination of both was calculated. SETTINGS: A single center for anal dysplasia. PARTICIPANTS: A total of 364 patients (72% males, 82% HIV-positive). INTERVENTION: Patients underwent anal cytology, HR-HPV test, and HRA-guided biopsy of the anal canal. MAIN OUTCOME MEASURES: Ability of cytology and HR-HPV test (individually and combined) to detect high-grade squamous intraepithelial lesions, and analysis of the cost of each diagnostic algorithm. RESULTS: Cytology alone was the cheapest approach, but had the lowest sensitivity [59%, 95% confidence interval (CI) 46-71%], despite of highest specificity (73%, 95% CI 68-78%). Cotesting had the highest sensitivity (85%, 95% CI 74-93%) and lowest specificity (43%, 95% CI 38-49%), and did not seem to be cost-effective. However, HR-HPV testing can be used to triage patients with normal and atypical squamous cells of undetermined significance cytology for HRA, resulting in an algorithm with high sensitivity (80%, 95% CI 68-89%), and specificity (71%, 95% CI 65-76%), allied to a good cost-effectiveness. CONCLUSION: HR-HPV testing is helpful to optimize the screening in cases of normal and atypical squamous cells of undetermined significance cytology.

Two-step ESD: an option for en-bloc resection of extensive colorectal laterally spreading tumors.

Background and study aims Endoscopic submucosal dissection (ESD) is considered feasible and safe for treatment of colorectal laterally spreading tumors (LST), However it remains a challenge in case of extensive lesions even for experts. This study aimed to describe a new method to facilitate ESD of extensive colorectal LSTs. Between July 2010 and January 2018, 140 patients underwent ESD for colorectal LSTs. Four of them were submitted to two-step ESD and were included in this retrospective study. The submucosal dissection of lesions larger than 12 cm started and continued until the medical team decided to pause the procedure and continue it in a second step. The second procedure was performed 2 days after to finish the en-bloc resection.Three patients were male, with mean age of 67.2y (±â€Š2.2). All lesions were located in the rectum, with a mean size of 153.7 mm (±â€Š33.8). En-bloc and curative resection were successfully achieved in all cases. Mean duration of the first step of the procedure was 255 minutes (±â€Š61.8), and mean duration of the second step was 205 minutes (±â€Š205). Overall mean duration of both steps was 460 minutes (±â€Š168). Mean dissected area in the first step of the procedure was approximately 55 % of the lesion. No adverse events were observed. In conclusion, our results suggest that performing ESD in two steps could be a feasible and safe option for exceptional cases in which is not possible to finish the procedure in one step, avoiding the morbidity of surgical treatment.

Role of Magnification Chromoendoscopy in the Management of Colorectal Neoplastic Lesions Suspicious for Submucosal Invasion.

BACKGROUND: Correctly predicting the depth of tumor invasion in the colorectal wall is crucial for successful endoscopic resection of superficial colorectal neoplasms. OBJECTIVE: The aim of this study was to assess the accuracy of magnifying chromoendoscopy in a Western medical center to predict the depth of invasion by the pit pattern classification in patients with colorectal neoplasms with a high risk of submucosal invasion. DESIGN: This single-center retrospective study, from a prospectively collected database, was conducted between April 2009 and June 2015. SETTINGS: The study was conducted at a single academic center. PATIENTS: Consecutive patients with colorectal neoplasms with high risk of submucosal invasion were included. These tumors were defined by large (≥20 mm) sessile polyps (nonpedunculated), laterally spreading tumors, or depressed lesions of any size. INTERVENTIONS: Patients underwent magnifying chromoendoscopy and were classified according to the Kudo pit pattern. The therapeutic decision, endoscopic or surgery, was defined by the magnification assessment. MAIN OUTCOME MEASURES: Sensitivity, specificity, and positive and negative predictive values of magnifying chromoendoscopy for assessment of these lesions were determined. RESULTS: A total of 123 lesions were included, with a mean size of 54.0 ± 37.1 mm. Preoperative magnifying chromoendoscopy with pit pattern classification had 73.3% sensitivity, 100% specificity, 100% positive predictive value, 96.4% negative predictive value, and 96.7% accuracy to predict depth of invasion and consequently to guide the appropriate treatment. Thirty-three rectal lesions were also examined by MRI, and 31 were diagnosed as T2 lesions. Twenty two (70.1%) of these lesions were diagnosed as noninvasive by magnifying colonoscopy, were treated by endoscopic resection, and met the curative criteria. LIMITATIONS: This was a single-center retrospective study with a single expert endoscopist experience. CONCLUSIONS: Magnifying chromoendoscopy is highly accurate for assessing colorectal neoplasms suspicious for submucosal invasion and can help to select the most appropriate treatment. See Video Abstract at http://links.lww.com/DCR/A920.

Transanal endoscopic microsurgery (TEM): a minimally invasive procedure for treatment of selected rectal neoplasms / Microcirurgia endoscópica transanal (TEM): um procedimento minimamente invasivo para o tratamento de neoplasias selecionadas do reto

Transanal endoscopic microsurgery (TEM) provides a minimally invasive alternative to radical surgery for excision of benign and malignant rectal tumors. TEM aims to provide an alternative to conventional abdominal surgery (low anterior resection or abdominoperineal amputations), which carries not inconsiderable morbidity and mortality. Based on review of the literature and in the authors experience, this review present the method and indications for TEM.

A microcirurgia endoscópica transanal (TEM) é procedimento alternativo minimamente invasivo ao tratamento cirúrgico radical para excisão de tumores benignos e malignos do reto. Ela oferece possibilidade operatória aos procedimentos cirúrgicos convencionais (ressecção anterior baixa ou amputações abdominoperineais), as quais acarretam alta morbimortalidade. Baseada na revisão da literatura e na experiência própria dos autores, esta revisão tem por objetivo apresentar o método e as indicações para a TEM.

Screening anal dysplasia in HIV-infected patients: is there an agreement between anal pap smear and high-resolution anoscopy-guided biopsy?

PURPOSE: The purpose of this study was to analyze the agreement between anal Pap smear and high-resolution anoscopy-guided biopsy in diagnosing anal dysplasia in HIV-infected patients. METHODS: We conducted cross-sectional analysis of HIV-infected patients receiving anal dysplasia screening as part of routine care. Agreement between measures was estimated by weighted kappa statistics, using a three-tiered cytologic and histologic grading system (normal, low-grade dysplasia, and high-grade dysplasia). Estimates of sensitivity, specificity, and predictive values were calculated using a two-tiered cytologic and histologic grading system ("without dysplasia" and "with dysplasia of any grade"). Estimates were also calculated for the detection of high-grade dysplasia. RESULTS: During a one-year period, 222 patients underwent 330 anal Pap smears followed by high-resolution anoscopy-guided biopsies. There were 311 satisfactory Pap smears with concurrent biopsies. Considering histology the standard, the frequency of anal dysplasia was 46%. Kappa agreement between anal Pap smear and biopsy was 0.20. For detection of anal dysplasia of any grade, anal Pap smear showed sensitivity of 61%, specificity of 60%, positive predictive value of 56%, and negative predictive value of 64%. For high-grade dysplasia, anal Pap smear showed sensitivity of 16% and specificity of 97%. CONCLUSION: Anal Pap smears alone were not sensitive enough to rule out anal dysplasia. We recommend that high-resolution anoscopy-guided biopsy be incorporated as a complementary screening test for anal dysplasia in high-risk patients. Following baseline high-resolution anoscopy, these individuals could be followed with serial anal cytology to dictate the need for future high-resolution anoscopy-guided biopsies.

Positron emission tomography detection of distant metastatic or synchronous disease in patients with locally advanced rectal cancer receiving preoperative chemoradiation.

BACKGROUND: Patients with locally advanced rectal cancer may present with synchronous distant metastases. Choice of optimal treatment--neoadjuvant chemoradiation versus systemic chemotherapy alone--depends on accurate assessment of distant disease. We prospectively evaluated the ability of [18F]fluorodeoxyglucose ([18F]FDG) positron emission tomography (PET) to detect distant disease in patients with locally advanced rectal cancer who were otherwise eligible for combined modality therapy (CMT). METHODS: Ninety-three patients with locally advanced rectal cancer underwent whole-body [18F]FDG PET scanning 2-3 weeks before starting CMT. Sites other than the rectum, mesorectum, or the area along the inferior mesenteric artery were considered distant and were divided into nine groups: neck, lung, mediastinal lymph node (LN), abdomen, liver, colon, pelvis, peripheral LN, and soft tissue. Two nuclear medicine physicians blinded to clinical information used PET images and a five-point scale (0-4) to determine certainty of disease. A score greater than 3 was considered malignant. Confirmation was based on tissue diagnosis, surgical exploration, and subsequent imaging. RESULTS: At a median follow-up of 34 months, the overall accuracy, sensitivity, and specificity of PET in detecting distant disease were 93.7%, 77.8%, and 98.7% respectively. Greatest accuracy was demonstrated in detection of liver (accuracy = 99.9%, sensitivity = 100%, specificity = 98.8%) and lung (accuracy = 99.9%, sensitivity = 80%, specificity = 100%) disease; PET detected 11/12 confirmed malignant sites in liver and lung. A total of 10 patients were confirmed to have M1 stage disease. All 10 were correctly staged by pre-CMT PET; abdominopelvic computed tomography (CT) scans accurately detected nine of them. CONCLUSION: Baseline PET in patients with locally advanced rectal cancer reliably detects metastatic disease in liver and lung. PET may play a significant role in defining extent of distant disease in selected cases, thus impacting the choice of neoadjuvant therapy.

Squamous-cell carcinoma of the rectum: a rare but curable tumor.

PURPOSE: This study was designed to evaluate one institution's experience with treatment outcomes for rectal squamous-cell carcinoma. METHODS: Using our prospective Colorectal Database, we identified patients diagnosed with rectal squamous-cell carcinoma at our institution between 1983 and 2005. Pathology was rereviewed, tumor immunophenotype was compared to control cases of anal squamous-cell carcinoma and rectal adenocarcinoma, treatment modalities and outcomes were analyzed. RESULTS: Twelve patients were identified (10 females median age, 58 years). Median distal extent of tumors was 7 (range, 5-8) cm from the anal verge. Treatment included chemotherapy only (n = 1), chemoradiation only (n = 2), induction chemotherapy followed by chemoradiation and surgery (n = 2), chemoradiation followed by surgery (n = 5), and surgery followed by chemoradiation (n = 2). The chemotherapy regimen was 5-fluorouracil-based. Radiotherapy total dose was 50.4 Gy (1.8 Gy/day, daily x 5) external iliac and inguinal nodes were not included in the radiation field. Complete clinical responders to chemoradiation (n = 2) received no further treatment. All seven partial responders underwent surgery; six had complete pathologic response; nodal status in two of six was unknown because they had local excision. Immunophenotypical analysis showed similar keratin expression profile between rectal squamous-cell carcinoma (n = 5) and rectal adenocarcinoma (n = 5), which is different from anal squamous-cell carcinoma (n = 10). All patients were alive without evidence of disease at follow-up (median follow-up, 2.6 (range, 0.5-16) years). CONCLUSIONS: Our data suggest that most patients treated with upfront chemoradiation therapy followed by surgery did well. Sphincter-preserving surgery is usually feasible. Clinical judgment of tumor response after chemoradiation is not completely reliable. Immunohistochemistry suggests a common cellular origin for rectal squamous-cell carcinoma and rectal adenocarcinoma, which is different from anal squamous-cell carcinoma.

Prevalence of perianal intraepithelial neoplasia in HIV-infected patients referred for high-resolution anoscopy.

PURPOSE: This study was designed to describe perianal disease in a cohort of HIV-infected patients referred for high-resolution anoscopy. METHODS: A retrospective chart review was performed on 52 HIV-infected patients referred for high-resolution anoscopy from 2001 to 2005. All patients underwent anal canal and perianal high-resolution anoscopy in the office with biopsy of suspicious areas. Patients with high-grade intraepithelial perianal lesions underwent multiple biopsies under general anesthesia in the operating room to rule out malignancy. RESULTS: Of the 52 patients, 19 (37 percent) had perianal abnormalities noted on high-resolution anoscopy and underwent punch biopsy. The mean duration of known HIV infection in these 19 patients (15 males) was 10.6 years, with 17 on highly active antiretroviral therapy for the last 3-month period. Mean CD4 count was 371 cells/microl. Office perianal biopsies diagnosed two patients with invasive squamous-cell carcinoma and nine with high-grade squamous intraepithelial lesion. Seven of the nine patients with perianal high-grade squamous intraepithelial lesion on office biopsy were submitted to multiple biopsies under general anesthesia. One of these seven had an occult perianal squamous-cell carcinoma. CONCLUSIONS: Perianal disease was common in this group of HIV-infected patients; 11 patients (21 percent of total) were diagnosed with squamous-cell carcinoma or high-grade squamous intraepithelial lesion. Because only 19 patients had clinically suspicious perianal lesions biopsied, this may be an underestimate. Our data suggest that anal canal neoplasia often is accompanied by perianal disease and illustrates the need for biopsy of any suspicious perianal lesions.