Regulatory Decision Summary for Lumblue

This New Drug Submission (NDS) was filed to obtain market authorization for Lumblue (methylene blue extended and delayed release tablets) as a visualization aid for the detection of mucosal lesions, including non-polypoid and small lesions, in patients undergoing colonoscopy.

Health Canada does not consider the benefit-harm-uncertainty profile of Lumeblue (methylene blue delayed and extended release tablets) to be favourable as a visualization aid for the detection of mucosal lesions, including non-polypoid and small lesions, in patients undergoing colonoscopy.

The proposed indication was supported by the results of a phase III multi-centre, multi-national, placebo controlled, randomized, double-blind, parallel-group study in patients undergoing screening or surveillance colonoscopy for colorectal cancer (Study CB-17-01/06). This study was conducted in 1,208 male and female patients aged 50-75 years scheduled for screening or surveillance after more than 2 years from the previous colonoscopy. Lumeblue tablets were administered orally approximately 12 hours before the colonoscopy procedure. The primary efficacy endpoint was achieved in the overall population and in subjects 50-64 years of age; there was a significantly higher percentage of subjects with at least one histologically proven Adenoma or Carcinoma (defined by the Vienna Classification Grade 3-5b) in all regions of the colon in the Lumeblue-enhanced white light endoscopy group compared to placebo (white light endoscopy alone). However, the number of subjects with at least one histologically proven Adenoma or Carcinoma in all regions of the colon with the exception of the rectosigmoid region was comparable, indicating that the efficacy of Lumeblue was demonstrated only in the rectosigmoid region of the colon, which was further supported by a subgroup analysis of bowel regions. No statistical significance was reached for subjects undergoing screening and surveillance ≤2 years after previous colonoscopy, in subjects older than 65 years of age, in Black subjects, and in Male subjects.

The safety profile of methylene blue is well known. The most critical risks include methylene blue’s status as a teratogen that causes developmental anomalies. In the pivotal study, the only adverse events with incidence ≥1% were chromaturia (48.0% in the 200 mg Lumeblue group vs 1.5% in the Placebo group), feces discoloured (19.5% vs 0%), nausea (5.9% vs 3.5%), vomiting (4.7% vs 2.7%), headache (2.7% vs 1.7%), abdominal pain (1.2 % vs 0.4%), and hypotension (1.0% vs 0.6%).

Two major concerns related to Lumeblue’s efficacy and safety prompted Health Canada to issue a Notice of Non-Compliance (NON) on August 16, 2019:

1. A significant bias was introduced in the study design that prevented a proper evaluation of the results. As a result, the efficacy of Lumeblue for the proposed indication could not be established. The number and type of lesions were unknown before enrollment and were not subsequently protected by randomization, resulting in an imbalance in the number of carcinoma/adenoma per subject between treatment group and placebo at baseline. Health Canada was concerned that this imbalanced distribution of lesions may have biased the results of the primary and secondary analyses, because the difference in observed detection rate could be erroneously attributed to the effect of Lumeblue versus the comparator.

2. The mutagenic potential of methylene blue under conditions of white light colonoscopy was insufficiently evaluated in the submitted data, which precluded Health Canada from making informed decisions about the safety of Lumeblue under conditions of use.

The Sponsor was given an opportunity to respond to the above concerns. In response to the first Major Objection, the Sponsor argued that the observed differences in the number of carcinoma/adenoma per subject between treatment group and placebo at baseline does not introduce a valid argument for suspecting a Type I error greater than the usual and planned 5% when interpreting the study results. Due to how Lumeblue is administered and used in the human body, a novel design approach was needed in order to test its efficacy in a manner that prevented investigator or selection bias. Overall, the response provided by the Sponsor was considered to satisfactorily address the statistical issue raised in the NON.

The second Major Objection was raised due to inconsistent results across in vivo/in vitro studies and usage of a non-validated assay by the Sponsor, which raised a concern about the safety of Lumeblue with respect to its mutagenic potential under conditions of use. The Sponsor was encouraged to confirm the safety of Lumeblue under conditions of use with respect to its mutagenic potential using internationally recognized and recommended assays. In response, the Sponsor conducted an in vivo mammalian alkaline comet assay performed on cells isolated from dog colon biopsies. The study was conducted in Beagle dogs mimicking conditions of use in the clinic with respect to the dose, administration, and colonoscopy procedure (the length of the procedure, biopsy collection procedure etc.). The methodology was derived from appropriate ICH and OECD guidelines with some changes to better mimic conditions of use, which was found to be acceptable. The results demonstrated the presence of statistically significant DNA damage. Overall, Health Canada concluded that the new study provided by the Sponsor was not able to resolve the safety concerns with respect to the potential DNA damage associated with the proposed procedure of Lumeblue enhanced white light colonoscopy.

Assessment of the benefits, harms, and uncertainties concluded that the benefits associated with detecting more adenomas in the rectosigmoid region of the colon do not outweigh the risks associated with the potential genotoxic action of Lumeblue under conditions of use (i.e., white light colonoscopy). The risks to the proposed patient population, including those undergoing colonoscopy for routine screening and those undergoing colonoscopy for surveillance, were deemed to be substantial and risk mitigation strategies were determined to prove ineffective. The risks that could not be overcome include:

• The risk of interval cancer potentially associated with the mutagenic properties of Lumeblue under conditions of use (i.e., white light colonoscopy) is undefined in the presented evidence.

• Alternative approaches to detect lesions in the colon exist, including conventional white light colonoscopy, as was demonstrated in the pivotal clinical trial where with the exception of the rectosigmoid segment of the colon, the white light colonoscopy performed similarly to Lumeblue enhanced colonoscopy.

• Post-market surveillance and action cannot be used to identify and track patients in which Lumeblue’s genotoxic action has caused the development of adenomas and carcinomas versus those who developed the lesions due to causes unrelated to Lumeblue.

• Risk mitigation by limiting administration of the product to the regions of the lumen where improved visualization is needed is impossible because Lumeblue is administered orally 12 hours prior to the colonoscopy procedure.

• The risk of white light exposure, which promotes Lumeblue’s genotoxicity, cannot be mitigated. Because Lumeblue is administered orally and is incorporated throughout the colonic lumen, all regions of the lumen regardless of the presence or absence of lesions are at risk of exposure to white light. White light is necessary for the safe conduct of a colonoscopy; therefore, minimizing the time of white light exposure of the lumen is impractical and not safe. Total exposure to white light is difficult to predict as it depends on the complexity of the procedure and the experience of the endoscopist, among other factors. The time dependency of the DNA damage was not studied and the minimum time to induce damage under conditions of use is unknown; however, the DNA damage demonstrated in the Beagle dog study took place within typical colonoscopy procedure timing.

In view of the unfavourable benefit-harm-uncertainty profile of Lumeblue for the proposed indication, a NON-Withdrawal Letter is recommended as per section C.08.004 of the Food and Drug Regulations and Health Canada’s Management of Drug Submissions and Applications guidance document.