Lamb’s tripe extract and vitamin B12 capsules for treatment of chronic atrophic gastritis: A multicenter, randomized, double-blind, placebo parallel-controlled clinical trial

Chronic atrophic gastritis (CAG), characterized by gastric mucosal atrophy and intestinal metaplasia (IM), is a precancerous condition requiring effective intervention.[1] Although Helicobacter pylori (H. pylori) eradication is the cornerstone therapy, options for non-H. pylori CAG remains limited. Traditional Chinese herbal medicine, particularly lamb’s tripe extract and vitamin B12 capsules (LTEVB12C), have shown promise due to their bioactive components (rennet, pepsin, and mucoprotein), which enhance digestion, mucosal repair, and microbiome regulation.[2] This multicenter trial evaluated the efficacy and safety of LTEVB12C in non-H. pylori CAG. Study design and participants: A randomized, double-blind, placebo-controlled trial (ClinicalTrials.gov, NCT01474044) was conducted across 16 centers in China. The number of patients enrolled in each center could be seen in the Supplementary File 1, https://links.lww.com/CM9/C657. Eligible patients (aged 18–70 years) had histopathologically confirmed CAG without H. pylori infection (initially screened using the C13/C14 breath test and confirmed by pathology). Key exclusion criteria were: (1) CAG with high-grade intraepithelial neoplasia; (2) severe gastric mucosal erosion or bleeding requiring treatment; (3) active peptic ulcer; (4) gastroesophageal reflux disease; (5) esophageal stricture; (6) history of upper gastrointestinal tract surgery, or malignant diseases; (7) inadequate organ function; (8) pregnancy or lactation; and (9) participation in another trial. The full eligibility criteria are provided in the protocol. This study was conducted in accordance with the Declaration of Helsinki, with ethics approval and patient consent (No. 20110804-1 from Fourth Military Medical University). Randomization and masking: Eligible patients were randomized (2:1) to receive LTEVB12C or placebo via centralized allocation. The sponsor, holding the patents to the invention, produced identical-appearing pills. Participants, investigators, and staff were blinded to the treatment assignment, with randomization codes accessible only to designated pharmacy personnel. Blinding was maintained using indistinguishable capsules and opaque packaging. The pills were sealed in sequentially numbered boxes (computer-generated randomization) and shipped to sites. After written informed consent was obtained, research nurses dispensed medication from the lowest-numbered box to ensure allocation concealment. Procedures: All randomized patients received 180 days of LTEVB12C or placebo, taken three times a day, three capsules per dose, orally after meals. Gastroscopy with mucosal marking biopsies (five standardized sites) was performed at baseline and on day 180, as evaluated by two blinded endoscopists per center. Histopathology was scored (0–3 points) for chronic or active inflammation, atrophy, and IM. Follow-ups occurred at 30, 60, 120, and 180 days; adverse events (AEs) were assessed via questionnaires and laboratory tests. Continuous use of proton pump inhibitors, H2 receptor antagonists, gastric mucosa protective agents, or prokinetic agents for ≤7 consecutive days and ≤21 cumulative days during treatment was allowed. All other medications for CAG were prohibited. An independent committee adjudicated serious adverse events (SAEs). Treatment was continued unless consent was withdrawn, drug-related SAEs, or investigator judgment discontinuation necessary; discontinuation did not trigger unblinding. Protocol-defined quality control was maintained throughout the study. Protocol details are provided in the Supplementary File, https://links.lww.com/CM9/C657. Outcomes: The primary endpoints were histological response (≥1-point grade reduction) and reversal rates (≥2-point reduction in moderate/severe cases). Evaluations excluded normal baseline cases, and mild cases were not assessed for reversal rates. Two analytical methods were used: (1) principal analysis tracking the worst baseline site (targeted site), (2) comparison of the worst sites before and after treatment. The following secondary endpoints were included: (1) histopathological score changes (sum/individual site scores), (2) lack of efficacy (LOE) incidence (non-responders after 4 months could exit early), (3) endoscopic mucosal changes, (4) symptom score changes, and (5) safety. The LOE cases followed the last observation carried forward (LOCF) rule. Statistical analysis: The sample size was determined on the basis of the primary efficacy endpoint (response rate at 180 days), with an assumed response rate of 10% in the placebo group and 18% in the treatment group (minimum clinically meaningful difference). Using a two-sided α = 0.05, β = 0.2 (80% power), and a 2:1 allocation ratio (experimental vs. placebo), the initial calculation yielded a minimum requirement of 711 patients (474 in the treatment group and 237 in the placebo group). To account for an estimated 17% dropout rate, the final planned enrollment was adjusted to 855 patients (570:285), ensuring sufficient statistical power even with potential attrition. An independent committee conducted analyses using the full-analysis set (FAS) and per-protocol set (PPS) populations, with the rules of LOCF and worst observation carried forward for missing data. Safety was assessed in the as-treated populations. Continuous variables were compared via t-tests; categorical variables via χ2 or Fisher’s exact tests. Primary endpoints were analyzed with the two-way center-adjusted unordered Cochran–Mantel–Haenszel χ2 test (CMH-χ2); secondary endpoints were analyzed with t-tests (score sums), Wilcoxon tests (individual sites), and log-rank tests (LOE incidence). All tests were two-sided (P <0.05 significant), performed using SAS software (version 8.3; SAS Institute Inc., Cary, NC, USA) under the oversight of an independent data review committee. Study populations: From September 2011 to June 2012, of the 961 patients screened, 850 participants were randomly assigned across 16 tertiary centers in China (567 in LTEVB12C group and 283 in placebo group, [Supplementary Figure 1, https://links.lww.com/CM9/C657]). Two activated hospitals did not recruit any participants. The baseline demographics, histopathological grades, clinical symptoms, and endoscopic findings were balanced between the groups in the FAS and PPS populations [Supplementary Tables 1–9, https://links.lww.com/CM9/C657]. Efficacy: In the FAS, Method 1 (targeted-site assessment) revealed significantly higher histopathological response rates with LTEVB12C vs. placebo for chronic inflammation (49.21% vs. 39.93%; P = 0.010), atrophy (59.61% vs. 36.17%; P <0.001), and IM (62.24% vs. 34.29%; P <0.001), but not for active inflammation (63.69% vs. 58.62%; P = 0.607) [Supplementary Figure 2A, https://links.lww.com/CM9/C657]. LTEVB12C also demonstrated superior reversal rates for chronic inflammation (6.10% vs. 1.59%; P = 0.006), atrophy (19.96% vs. 6.39%; P <0.001), and IM (47.86% vs. 11.82%; P <0.001) than the placebo group, however, it did not show a significant benefit in active inflammation (50.00% vs. 63.89%; P = 0.042) [Supplementary Figure 2B, https://links.lww.com/CM9/C657]. Using Method 2 (most severe-site assessment), LTEVB12C showed higher response rates for chronic inflammation (15.70% vs. 10.60%; P = 0.041), atrophy (16.01% vs. 7.80%; P = 0.001), and IM (32.29% vs. 16.57%; P <0.001), though active inflammation remained comparable (34.48% vs. 41.07%; P = 0.534), compared with the placebo group [Supplementary Figure 3A, https://links.lww.com/CM9/C657]. IM reversal rates were significantly higher with LTEVB12C after adjustment for baseline imbalance (P = 0.008) [Supplementary Figure 3B, https://links.lww.com/CM9/C657]. Secondary analyses favored LTEVB12C for improvements in the summed and worst histopathological grades of chronic inflammation, atrophy, and IM [Supplementary Tables 10 and 11, https://links.lww.com/CM9/C657], with subgroup analyses confirming benefits across affected sites [Supplementary Tables 12–16, https://links.lww.com/CM9/C657]. No between-group differences were observed in endoscopic outcomes, clinical symptom scores [Supplementary Tables 17 and 18, https://links.lww.com/CM9/C657], and the cumulative incidence of LOE (log-rank P = 0.740 [Supplementary Figure 4, https://links.lww.com/CM9/C657]). All of these results were consistent with the PPS analysis [Supplementary Tables 10–18, Supplementary Figures 5–7, https://links.lww.com/CM9/C657]. Safety: The mean duration of treatment was 170.9 ± 43.5 days in the placebo group, and 170.3 ± 42.3 days in the LTEVB12C group (P = 0.837). Compliance rates of 80–120% were achieved by 88.85% and 87.45% of patients in the placebo and LTEVB12C groups, respectively (P = 0.567 [Supplementary Table 19, https://links.lww.com/CM9/C657]). No significant between-group differences were observed in changes in vital signs [Supplementary Table 20, https://links.lww.com/CM9/C657], or in the use of concomitant drugs between these groups during the study period [Supplementary Table 21, https://links.lww.com/CM9/C657]. The AE and SAE rates were 17.81% and 1.76% in the placebo group, and 20.85% and 1.06% in the LTEVB12C group, respectively, during 180 days of follow-up [Supplementary Table 22, https://links.lww.com/CM9/C657]. Among the events related to the study agents, the most common AEs were skin and appendage damage (8/567, 1.41%), and diarrhea (5/567, 0.88%) in the LTEVB12C group, while skin and appendage damage (7/567, 2.47%) and head discomfort (3/283, 1.06%) in the placebo group. This multicenter, randomized, double-blind, placebo-controlled trial provides compelling evidence that LTEVB12C significantly improves histopathological outcomes in patients with non-H. pylori CAG, with superior response rates and reversal rates in chronic inflammation, atrophy, and IM compared with placebo. These robust findings, consistent across the FAS and PPS populations, were further confirmed using two distinct analytical methods, suggesting that LTEVB12C holds substantial promise as a novel therapeutic intervention for CAG. The therapeutic efficacy of LTEVB12C appears to stem from the complementary pharmacological actions of its key components.[2,3] Although further research is needed to fully elucidate its therapeutic potential of LTEVB12C in CAG, several mechanisms may be involved: (1) Supplying digestive enzymes (pepsin and rennin) to compensate for reduced gastric acid, (2) enhancing mucosal repair via growth factors and nutrients (e.g., vitamin B12), (3) modulating gastric acid/gastrin secretion, and (4) exerting anti-inflammatory and antioxidant effects to reduce oxidative stress. Although preclinical and limited clinical studies have suggested benefits, further rigorous human studies are required to confirm these mechanisms. A noteworthy finding was the dissociation between LTEVB12C’s significant histopathological benefits and its lack of superiority in symptom relief or endoscopic improvement. The study was powered primarily for histopathological improvement, while symptom and endoscopic endpoints—both subject to greater variability—were secondary. Histopathology offers an objective measure of mucosal repair, whereas symptoms are influenced by multifactorial physiological and psychosocial factors, and conventional endoscopy lacks sensitivity to early structural changes. These results challenge the notion that histological improvement must align with short-term clinical outcomes, underscoring the need for distinct evaluation frameworks in future trials. A particularly intriguing finding was LTEVB12C’s limited effect on active inflammation, with the placebo group showing a higher reversal rate. This may be due to the low baseline prevalence of moderate-to-severe cases, which limited statistical power, and the agent’s primary action targeting chronic atrophy repair rather than acute inflammation, allowing natural fluctuation to drive the placebo response.[2] The study has several limitations. Its 180-day duration precludes the assessment of cancer prevention. Residual effects from past infections could not be excluded due to the lack of serological testing, suggesting that future trials to incor-porate anti-H. pylori immunoglobulin G testing to better stratify participants. The baseline imbalance in IM severity, though addressed through statistical adjustment using logistic regression, highlights the value of stratification by histological grade in future studies. Conflicts of interest None. Acknowledgements The authors gratefully acknowledge the contributions of the investigators, research staff, and all committee members. We also extend our sincere thanks to the participants and their families.