CLASSICA: Validating artificial intelligence in classifying cancer in real time during surgery

AIM

Treatment pathways for significant rectal polyps differ depending on the underlying pathology, but pre-excision profiling is imperfect. It has been demonstrated that differences in fluorescence perfusion signals following injection of indocyanine green (ICG) can be analysed mathematically and, with the assistance of artificial intelligence (AI), used to classify tumours endoscopically as benign or malignant. This study aims to validate this method of characterization across multiple clinical sites regarding its generalizability, usability and accuracy while developing clinical-grade software to enable it to become a useful method.

METHODS

The CLASSICA study is a prospective, unblinded multicentre European observational study aimed to validate the use of AI analysis of ICG fluorescence for intra-operative tissue characterization. Six hundred patients undergoing transanal endoscopic evaluation of significant rectal polyps and tumours will be enrolled in at least five clinical sites across the European Union over a 4-year period. Video recordings will be analysed regarding dynamic fluorescence patterns centrally as software is developed to enable analysis with automatic classification to happen locally. AI-based classification and subsequently guided intervention will be compared with the current standard of care including biopsies, final specimen pathology and patient outcomes.

DISCUSSION

CLASSICA will validate the use of AI in the analysis of ICG fluorescence for the purposes of classifying significant rectal polyps and tumours endoscopically. Follow-on studies will compare AI-guided targeted biopsy or, indeed, AI characterization alone with traditional biopsy and AI-guided local excision versus traditional excision with regard to marginal clearance and recurrence.

Robotic transanal minimally invasive surgery - technical, oncological and patient outcomes from a single institution

AIM

Robotic transanal minimally invasive surgery (R-TAMIS) is gaining traction around the globe as an alternative to laparoscopic conventional TAMIS for local excision of benign and early malignant rectal lesions. The aim was to analyse patient and oncological outcomes of R-TAMIS for consecutive cases in a single centre.

METHODS

A prospective analysis of consecutive R-TAMIS procedures over a 12-month period was performed. Data were collated from hospital databases and theatre registers.

RESULTS

Eleven patients (six men, five women), mean age 69.81 years (51-92 years), underwent R-TAMIS over 12 months utilizing a da Vinci Xi platform. The mean lesion size was 36 mm (20-60 mm) with a mean distance from the anal verge of 7.5 cm (3-14 cm). Five lesions were posterior in anatomical location, four anterior, one right lateral and one left lateral. All procedures were performed in the lithotomy position using a GelPOINT Path Platform. Mean operative time was 64 min (40-100 min). Complete resection was achieved in 10/11 patients with two patients being upgraded to a diagnosis of adenocarcinoma. Nine patients were diagnosed with dysplastic lesions. Four patients had a false positive diagnosis of an invasive tumour on MRI. Six patients required suturing for full-thickness resections. One patient had a postoperative bleed requiring repeat endoscopy and clipping. One patient (full-thickness resection of T3 tumour) proceeded to a formal resection without difficulty with no residual disease (T0N0, 0/22). One patient with a fully resected T2 tumour is undergoing a surveillance protocol. The mean length of stay was 1 day with two patients having a length of stay of 2 days and one patient of 4 days.

CONCLUSION

R-TAMIS could potentially represent a safe novel approach for local resection of rectal lesions.

Resection of rectal GIST using a novel technique: a report of two cases

Rectal gastrointestinal stromal tumours (GISTs) are uncommon tumours and usually present with large sizes. We present two cases of rectal GIST. Imatinib was used in the setting of neoadjuvant and adjuvant therapy. Both tumours were resected transanally by the transanal endoscopic operation (TEO) platform. Oncosurgeons are recommended to implement sphincter-sparing surgeries for these cases.

Is there a limit to transanal endoscopic surgery? A comparative study between standard and technically challenging indications among 168 consecutive patients

AIM

To assess the surgical outcome of transanal endoscopic surgery (TES) for rectal neoplasms in technically challenging indications.

METHOD

All patients who underwent TES for a rectal neoplasm from 2007 to 2014 were included. Technically challenging indications included a tumour with (i) diameter ≥ 5 cm, (ii) involving ≥ 50% of the rectal circumference and (iii) located ≥ 10 cm from the anal verge. Patients were divided into three groups according to how many of these features they had, as follows: Group 1, none; Group 2, one; Group 3, two or more.

RESULTS

Of the 168 patients (80 benign and 88 malignant tumours) included in the study, 73 (44%) were in Group 1, 46 (27%) in Group 2 and 49 (29%) in Group 3. There was no difference between Group 1 and Group 2 with regard to peritoneal perforation (P = 0.210), severe postoperative morbidity (P = 0.804), length of hospital stay (P = 0.444), incomplete resection (P = 0.441), piecemeal resection (P = 0.740), locoregional recurrence (P = 0.307) and long-term symptomatic rectal stenosis (P = 0.076). Conversely Group 3 showed significantly impaired results compared with Group 1 with regard to peritoneal perforation (P = 0.003), piecemeal resection (P = 0.005), incomplete resection (P = 0.025), locoregional recurrence (P = 0.035) and long-term symptomatic rectal stenosis (P < 0.001), but no difference in severe postoperative morbidity (P = 0.328).

CONCLUSION

Transanal endoscopic surgery for rectal neoplasms appears to be safe and effective, even in patients presenting with a technically challenging tumours. Although the short- and long-term outcomes after TES are worse in patients with highly challenging tumours, nevertheless the technique should still be considered in patients at high risk of requiring a proctectomy.

Structural proteins of bovine coronavirus strain L 9: effects of the host cell and trypsin treatment

The polypeptide profile of the cell-adapted strain of bovine coronavirus (Mebus BCV-L 9) is remarkably affected by the host cell and trypsin. We compared the structural proteins of virus purified from different cell lines and found cell-dependent differences in the virus structure. BCV was purified from four clones of human rectal tumour cells (HRT-18): 3F3, D2, 3E3, and 4B3. The structural profiles of BCV propagated in clones 3E3 and 3F3 were identical, consisting of proteins with molecular weights of 185, 160, 140, 125, 110, 100, 52, 46, 37, 31-34, and 26-28 kilodaltons (kd). BCV purified from clone D2 lacked the 100 kd species, and clone 4B3 yielded virus lacking the 46 kd protein. We compared the structures of BCV propagated in HRT-18 cells [BCV(HRT-18)] and virus raised in bovine fetal spleen cells [BCV(D2 BFS)]. The concentration of the 185 kd protein was higher in BCV (D2BFS), and it also contained a 200 kd species. Protein profiles of in vitro trypsin treated and untreated BCV(HRT-18) differed only under reducing conditions, suggesting that trypsin cleavage sites are located within disulfide-linked regions of affected proteins. Propagation of BCV in D2 BFS cells in the presence of trypsin resulted in cleavage of the 185 kd protein and a concomitant increase of the 100 kd protein. Activation of the fusion function probably depends on this cleavage process because fusion of BCV-infected D2 BFS cells is trypsin dependent.