Noa Davis1, Alexander Shtabsky2, Sylvia Lew3, Ronny Rona4, Ilan Leibovitch5, Ofer Nativ6, Michael Cohen7, Yoram Mor8, Uri Lindner9, Yael Glickman10, Haim Matzkin11, Alexander Tsivian12, Ofer Gofrit13, Ofer Yossepovitch14. 1. Micromedic Technologies Ltd., Tel Aviv, Israel. Electronic address: noa@m-medic.com. 2. Institute of Pathology, Tel Aviv Sourasky Medical Center, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel. 3. Patho-Lab Diagnostics Ltd., Ness Ziona, Israel. 4. Kfar Saba, Israel. 5. Department of Urology, Meir Medical Center, Kfar Saba, Israel. 6. Department of Urology, Bnai Zion Medical Center, Haifa, Israel. 7. Department of Urology, HaEmek Medical Center, Afula, Israel. 8. Department of Urology, Sheba Medical Center, Tel Aviv, Israel. 9. Department of Urology, Kaplan Medical Center, Rehovot, Israel. 10. Micromedic Technologies Ltd., Tel Aviv, Israel. 11. Department of Urology, Sourasky Medical Center, Tel Aviv, Israel. 12. Department of Urology, Wolfson Medical Center, Holon, Israel. 13. Department of Urology, Hadassah Medical Center, Jerusalem, Israel. 14. Department of Urology, Rabin Medical Center, Petach Tikva, Israel.
Abstract
BACKGROUND: CellDetect is a unique histochemical stain enabling color and morphological discrimination between malignant and benign cells based on differences in metabolic signature. OBJECTIVE: The objective of the present study was to validate the performance of this assay in a controlled, blinded, multicenter study. DESIGN, SETTING, AND PARTICIPANTS: The study, conducted in nine hospitals, included patients with documented history of bladder cancer, monitored for urothelial carcinoma (UCC) or scheduled for bladder cancer surgery. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Cystoscopy and/or biopsy were used as a reference standard to determine sensitivity and specificity. Smears were stained by CellDetect and interpreted by two cytologists blinded to the patient's final diagnosis. The findings were compared with those of standard urine cytology and BTA stat. RESULTS AND LIMITATIONS: Two hundred and seventeen voided urine specimens were included. Ninety-six (44%) were positive by histology and 121 (56%) were negative by either cystoscopy or histology. The overall sensitivity of CellDetect was 84%. Notably, the sensitivity for detecting low-grade nonmuscle-invasive bladder cancer tumors was greater than this of BTA stat (78% vs 54%) and more than two-fold higher compared with standard cytology (33%, p ≤ 0.05). The specificity was 84% in patients undergoing routine surveillance by cystoscopy. At a median follow-up of 9 mo, 21% of the patients with positive CellDetect and negative reference standard developed UCC, which was significantly higher compared with the 5% of the true negative cases. Limitations include the lack of instrumental urine samples and the lack of patients with nongenitourinary cancers in the study population. CONCLUSIONS: This study validates the performance of CellDetect as a urine-based assay to identify UCC in patients with history of bladder cancer. The high sensitivity was maintained across all cancer grades and stages without compromising the assay specificity. Further studies are required to test whether this novel stain can be incorporated in routine bladder cancer surveillance as a noninvasive alternative to cystoscopy. PATIENT SUMMARY: Surveillance of bladder cancer requires frequent invasive procedures. In the present study, we validate the ability of a novel biomarker to accurately identify early-stage tumors in urine specimens for the noninvasive monitoring of patients with history of bladder cancer.
RCT Entities:
BACKGROUND:CellDetect is a unique histochemical stain enabling color and morphological discrimination between malignant and benign cells based on differences in metabolic signature. OBJECTIVE: The objective of the present study was to validate the performance of this assay in a controlled, blinded, multicenter study. DESIGN, SETTING, AND PARTICIPANTS: The study, conducted in nine hospitals, included patients with documented history of bladder cancer, monitored for urothelial carcinoma (UCC) or scheduled for bladder cancer surgery. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Cystoscopy and/or biopsy were used as a reference standard to determine sensitivity and specificity. Smears were stained by CellDetect and interpreted by two cytologists blinded to the patient's final diagnosis. The findings were compared with those of standard urine cytology and BTA stat. RESULTS AND LIMITATIONS: Two hundred and seventeen voided urine specimens were included. Ninety-six (44%) were positive by histology and 121 (56%) were negative by either cystoscopy or histology. The overall sensitivity of CellDetect was 84%. Notably, the sensitivity for detecting low-grade nonmuscle-invasive bladder cancer tumors was greater than this of BTA stat (78% vs 54%) and more than two-fold higher compared with standard cytology (33%, p ≤ 0.05). The specificity was 84% in patients undergoing routine surveillance by cystoscopy. At a median follow-up of 9 mo, 21% of the patients with positive CellDetect and negative reference standard developed UCC, which was significantly higher compared with the 5% of the true negative cases. Limitations include the lack of instrumental urine samples and the lack of patients with nongenitourinary cancers in the study population. CONCLUSIONS: This study validates the performance of CellDetect as a urine-based assay to identify UCC in patients with history of bladder cancer. The high sensitivity was maintained across all cancer grades and stages without compromising the assay specificity. Further studies are required to test whether this novel stain can be incorporated in routine bladder cancer surveillance as a noninvasive alternative to cystoscopy. PATIENT SUMMARY: Surveillance of bladder cancer requires frequent invasive procedures. In the present study, we validate the ability of a novel biomarker to accurately identify early-stage tumors in urine specimens for the noninvasive monitoring of patients with history of bladder cancer.
Authors: Jan Gleichenhagen; Christian Arndt; Swaantje Casjens; Carmen Töpfer; Holger Gerullis; Irina Raiko; Dirk Taeger; Thorsten Ecke; Thomas Brüning; Georg Johnen Journal: PLoS One Date: 2022-06-29 Impact factor: 3.752