Background Carcinoembryonic antigen is overexpressed in colorectal cancer (CRC), making it an optimal target for fluorescence imaging. A phase I/II study was designed to determine the optimal imaging dose of SGM-101 for intraoperative fluorescence imaging of primary and recurrent CRC. Methods Patients were included and received a single dose of SGM-101 at least 24 h before surgery. Patients who received routine anticancer therapy (i.e., radiotherapy or chemotherapy) also were eligible. A dedicated near-infrared imaging system was used for real-time fluorescence imaging during surgery. Safety assessments were performed and SGM-101 efficacy was evaluated per dose level to determine the most optimal imaging dose. Results Thirty-seven patients with CRC were included in the analysis. Fluorescence was visible in all primary and recurrent tumors. In seven patients, no fluorescence was seen; all were confirmed as pathological complete responses after neoadjuvant therapy. Two tumors showed false-positive fluorescence. In the 37 patients, a total of 97 lesions were excised. The highest mean intraoperative tumor-to-background ratio (TBR) of 1.9 (p = 0.019) was seen in the 10-mg dose. This dose showed a sensitivity of 96%, specificity of 63%, and negative predictive value of 94%. Nine patients (24%) had a surgical plan alteration based on fluorescence, with additional malignant lesions detected in six patients. Conclusions The optimal imaging dose was established at 10 mg 4 days before surgery. The results accentuate the potential of SGM-101 and designated a promising base for the multinational phase III study, which enrolled the first patients in June 2019.

Background: In the West, pre-treatment abnormal lateral lymph nodes (LLN+) in patients with a low locally advanced rectal cancer (AJCC Stage III), are treated with neoadjuvant (chemo)radiotherapy (nCRT), without a lateral lymph node dissection (LLND). It has been suggested, however, that LLN+ patients have higher local recurrence (LR) rates than similarly staged patients with abnormal mesorectal lymph nodes only (LLN−), but no comparative data exist. Therefore, we conducted this international multi-center study in the Netherlands and Australia of Stage III rectal cancer patients with either LLN+ or LLN− to compare oncological outcomes from both groups.Materials and Methods: Patients with Stage III low rectal cancer with (LLN+ group) or without (LLN− group) abnormal lateral lymph nodes on pre-treatment MRI were included. Patients underwent nCRT followed by rectal resection surgery with curative intent between 2009 and 2016 with a minimum follow-up of 2-years. No patient had a LLND. Propensity score matching corrected differences in baseline characteristics.Results: Two hundred twenty-three patients could be included: 125 in the LLN+ group and 98 in the LLN− group. Between groups, there were significant differences in cT-stage and in the rate of adjuvant chemotherapy administered. Propensity score matching resulted in 54 patients in each group, with equal baseline characteristics. The 5-year LR rate in the LLN+ group was 11 vs. 2% in the LLN− group (P = 0.06) and disease-free survival (DFS) was 64 vs. 76%, respectively (P = 0.09). Five-year overall survival was similar between groups (73 vs. 80%, respectively; P = 0.90).Conclusions: In Western patients with Stage III low rectal cancer, there is a trend toward worse LR rate and DFS rates in LLN+ patients compared to similarly staged LLN− patients. These results suggest that LLN+ patients may currently not be treated optimally with nCRT alone, and the addition of LLND requires further consideration.

PurposeIntraoperative identification of rectal cancer (RC) can be challenging, especially because of fibrosis after treatment with preoperative chemo- and radiotherapy (CRT). Tumor-targeted fluorescence imaging can enhance the contrast between tumor and normal tissue during surgery. Promising targets for RC imaging are carcinoembryonic antigen (CEA), epithelial cell adhesion molecule (EpCAM) and the tyrosine-kinase receptor Met (c-Met). The effect of CRT on their expression determines their applicability for imaging. Therefore, we investigated whether CRT modifies expression patterns in tumors, lymph node (LN) metastases and adjacent normal rectal tissues.Patients and methodsPreoperative biopsies, primary tumor specimens and metastatic LNs were collected from 38 RC patients who did not receive CRT (cohort 1) and 34 patients who did (cohort 2). CEA, EpCAM and c-Met expression was determined using immunohistochemical staining and was semiquantified by a total immunostaining score (TIS), consisting of the percentage and intensity of stained tumor cells (0–12).ResultsIn both cohorts CEA, EpCAM and c-Met were significantly highly expressed in >60% of tumor tissues compared with adjacent normal epithelium (T/N ratio, P<0.01). EpCAM showed the most homogenous expression in tumors, whereas CEA showed the highest T/N ratio. Most importantly, CEA and EpCAM expression did not significantly change in normal or neoplastic RC tissue after CRT, whereas levels of c-Met changed (P=0.02). Tissues of eight patients with a pathological complete response after CRT showed expression of all biomarkers with TIS close to normal epithelium.ConclusionHistological evaluation shows that CEA, EpCAM and c-Met are suitable targets for RC imaging, because all three are significantly enhanced in cancer tissue from primary tumors or LN metastases compared with normal adjacent tissue. Furthermore, the expression of CEA and EpCAM is not significantly changed after CRT. These data underscore the applicability of c-Met and especially, CEA and EpCAM as targets for image-guided RC surgery, both before and after CRT.

No abstract