The findings, published online in Gastroenterology (2023 Jun 3. doi: 10.1053/j.gastro.2023.05.042), come from the largest (Hemospray, Cook Medical), compared with standard endoscopic hemostatic interventions for tumor bleeding.
For their research, Rapat Pittayanon, MD, of Chulalongkorn University in Bangkok and her colleagues, randomized patients (60% male, mean age 63) with active malignant upper or lower GI bleeding and low disability levels related to their cancers (ECOG score 0-2). The study was conducted at nine hospitals in Thailand.
The 106 patients who passed screening underwent either TC-325 or standard endoscopic hemostasis, which could involve use of thermal or mechanical methods or adrenaline injection, alone or combined with another modality, at the endoscopist’s discretion. Crossover between treatment allocations was permitted if hemostasis was not achieved. Investigators assessed rates of immediate hemostasis and rebleeding at 30 days.
Dr. Pittayanon and colleagues found rebleeding to be significantly lower among TC-325 treated patients, at 2.1%, compared with 21.3% for standard care (odds ratio, 0.09; 95% confidence interval, 0.01-0.80; P = .03). Rates of immediate hemostasis were 100% for TC-325–treated subjects, compared with 68.6% in the conventional-treatment group (OR, 1.45; 95% CI, 0.93-2.29; P < .001).
None of the 55 patients in the TC-325 group underwent crossover treatment, but 15 patients in the standard care group were crossed over to TC-325 after their endoscopic treatment was deemed to have failed. One-fifth of patients who got TC-325 as a crossover treatment developed rebleeding at 30 days, which the investigators surmised was related to mucosal damage incurred during the endoscopic procedure.
The study was not powered to adequately assess survival outcomes. Seven patients in the TC-325 group and four in the conventional care group died before 30 days’ follow-up, and no death was directly related to recurrent tumor bleeding.
“To our knowledge, our trial is the first to show such significant findings in an RCT setting, which now provide a long-awaited efficacious hemostatic approach where one had been lacking when managing patients with malignant GI bleeding,” the investigators wrote in their analysis.
“Perhaps most importantly, this carefully controlled study also highlights the unreliable hemostatic effect of standard endoscopic modalities available for GI tumor hemostasis, with high 30-day rebleeding rates in our patient population.”
Dr. Pittayanon and colleagues noted several limitations of their study. These included the inability to blind patients to an endoscopist, which “may have influenced subsequent management decisions … including the decision to cross over.”
Only in 5 of 15 cases of crossover did the treating endoscopist provide photo evidence of treatment failure as required by the trial’s protocol. Also, the use of adrenaline injection alone was permitted in the study, in contrast to best practice guidelines for endoscopic hemostasis to treat peptic ulcer bleeding. Finally, the study was conducted in Thailand, potentially reducing the generalizability of the results.
The study was funded by King Chulalongkorn Memorial Hospital; the Thai Red Cross; and Chulalongkorn University. Cook Medical donated some of the TC-325 kits used in the study.
One study coauthor, Alan N. Barkun, disclosed consulting work for Medtronic and past paid work for Cook Medical. The remaining authors disclosed no conflicts of interest.