RELiZORB® by Alcresta Therapeutics, Inc.

RELiZORB® is a single-use, point-of-care digestive enzyme cartridge that connects in-line with existing enteral feeding circuits. RELiZORB is designed to hydrolyze (digest) fats contained in enteral formulas from triglycerides into fatty acids and monoglycerides to allow for their absorption and utilization by the body. This hydrolysis of fats by RELiZORB is intended to mimic the function of the digestive enzyme lipase in patients. RELiZORB is comprised of a clear cylindrical, plastic cartridge with a single inlet connection port and a single outlet connection port. Inside the cartridge, there are small white beads covalently bonded to the digestive enzyme, lipase. This lipase-bead complex, iLipase™ (immobilized lipase) is retained within the cartridge during use by filters on both ends of the cartridge. The fat in enteral formulas is hydrolyzed upon contact with iLipase as the formula passes through the cartridge.

AI/ML Overview

This document is a 510(k) summary for the RELiZORB® device, which is an enzyme-packed cartridge used to hydrolyze fats in enteral formula. It is a submission to the FDA seeking clearance for a modified version of an already cleared device.

The provided text focuses on demonstrating the substantial equivalence of the subject device (RELiZORB® K232784) to its predicate device (RELiZORB® K231156). It does not contain information about the acceptance criteria and study proving performance for an AI/Software as a Medical Device (SaMD).

Specifically, the document does not contain the following information typically found in acceptance criteria and study designs for AI/SaMD devices:

A table of acceptance criteria and reported device performance (for an AI/SaMD)
Sample sizes used for a test set (as would be used for an AI/SaMD model's performance evaluation)
Data provenance for a test set
Number of experts and qualifications for ground truth establishment
Adjudication method for a test set
MRMC comparative effectiveness study details
Standalone (algorithm-only) performance results
Type of ground truth used (e.g., pathology, outcomes data)
Sample size for the training set (for an AI/ML model)
How ground truth for the training set was established

Instead, the document details performance testing related to the physical and chemical characteristics of the RELiZORB® cartridge, which is a hardware medical device with an enzymatic function.

Here's a breakdown of the performance testing cited for this hardware device, which is different from "acceptance criteria" for an AI/SaMD:

Overview of Performance Testing for RELiZORB® (Hardware Device):

The performance testing listed is geared towards a physical device with a biological function (enzyme action). The "acceptance criteria" for this device would therefore relate to its physical integrity, chemical activity, and biocompatibility, rather than diagnostic accuracy or algorithmic performance.

1. Table of "Acceptance Criteria" (Inferred from Performance Testing) and Reported Device Performance:

The document lists categories of testing rather than specific numerical acceptance criteria or detailed results. However, we can infer some "acceptance criteria" that would be met by these tests:

Acceptance Criteria Category (Inferred)	Specific Tests Performed (from text)	Reported Device Performance (General Statement)
Biocompatibility	Biocompatibility testing (Sections 15, 19)	Device causes neither an adverse tissue response nor adverse performance.
Device Integrity/Durability	Internal and external appearance, Pouch integrity, Physical/Mechanical Testing (Filter tensile strength, Weld tensile strength, Filter integrity/bead retention), Ship testing	Demonstrated to withstand clinical forces, maintain integrity, and retain beads.
Enzymatic Function	Hydrolysis, Demonstration of enzymatic effect on intended macronutrient	Hydrolyzes fats in enteral formula from triglycerides into fatty acids and monoglycerides.
Enzyme Retention	Unconjugated protein, Amount of enzyme that exits the cartridge (Section 10)	Enzyme is retained within the cartridge during use by filters; amount exiting the cartridge is characterized.
Flow Dynamics	Flow rate and leakage testing, Validation that the device does not impede flow alarms on enteral feeding pumps (Sections 18 & 14)	Does not impede the flow of enteral formula or flow alarms on enteral feeding pumps.
Nutritional Impact	Validation that the device does not adversely impact the nutritional composition of enteral formula (Section 18)	Does not adversely impact the nutritional composition.
Shelf Life	Evaluation of product shelf life, Performance data to support shelf life by demonstrating package integrity and device functionality over the identified shelf life (Section 14)	Functionality and integrity maintained over identified shelf life.
Safety in Use	Human factors testing (Section 10), Preclinical Studies in porcine model	Risks characterized; safety and efficacy assessed, including impact on fat absorption and weaning from parenteral to enteral nutrition.
Labeling Requirements	Labeling (Section 13)	Labeling includes required summaries, instructions, warnings, and patient information.

2. Sample size used for the test set and the data provenance:

Test Set Sample Size: Not applicable in the context of an AI/ML test set. For physical and chemical tests, typically specific numbers of units are tested according to standard protocols (e.g., n=3, n=5, n=10 per batch), but these specific numbers are not provided in the summary.
Data Provenance: The preclinical studies were conducted in a porcine model. The document does not specify the country of origin for the studies, nor whether they were retrospective or prospective, though preclinical animal studies are typically prospective.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

Not applicable as this is not an AI/ML diagnostic device requiring human expert ground truth for interpretation of imaging or other complex data. The "ground truth" for this device's performance would be the scientific measurement of its physical properties and chemical activity (e.g., fat hydrolysis percentage).

4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:

Not applicable. Adjudication methods are relevant for human interpretation or consensus in diagnostic studies, not for the performance testing of a physical medical device.

5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance:

No. This is not an AI-assisted diagnostic device, so an MRMC study is not relevant or performed.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

No. This is a physical device, not an algorithm.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

For this device, the "ground truth" for its performance is established through standard laboratory testing for:
- Chemical Analysis: Measuring the hydrolysis of fats (e.g., quantifying fatty acids and monoglycerides produced).
- Physical Measurements: Tensile strength, flow rates, bead retention, appearance.
- Biological Testing: Biocompatibility tests (e.g., cytotoxicity, sensitization), and in vivo animal studies (porcine model) to assess safety and efficacy (fat absorption, impact on parenteral nutrition weaning).

8. The sample size for the training set:

Not applicable. This is not an AI/ML device that requires a "training set."

9. How the ground truth for the training set was established:

Not applicable.

In summary: The provided text details the regulatory clearance process for a physical medical device (enzyme-packed cartridge), not an AI/Software as a Medical Device (SaMD). Therefore, the criteria and study types described in the prompt (which are typical for AI/SaMD) are not present in this document. The "performance data" here refers to standard engineering, chemical, and biological testing of a hardware product.

Summary

Regulation Number and Section

§ 876.5985 Enzyme packed cartridge.

(a)
Identification. An enzyme packed cartridge is anex vivo prescription device that is used in enzymatic hydrolysis of macronutrients into their essential nutrient forms at the time of delivery. The device consists of an outer casing containing an inert polymer with a covalently bound enzyme through which nutritional formula is directed. The device fits in line with enteral feeding systems.(b)
Classification. Class II (special controls). The special controls for this device are:(1) The patient contacting components of the device must be demonstrated to be biocompatible.
(2)
In vivo testing must be performed and must demonstrate that the device causes neither an adverse tissue response nor adverse performance.(3) Non-clinical testing must demonstrate that the device performs as intended under anticipated conditions of use. The following performance characteristics must be demonstrated:
(i) Mechanical testing to demonstrate that the device can withstand clinical forces;
(ii) Flow rate and leakage testing to demonstrate that the device does not impede the flow of enteral formula;
(iii) Demonstration of enzymatic effect on intended macronutrient;
(iv) The amount of enzyme that exits the cartridge must be characterized;
(v) Validation that the device does not adversely impact the nutritional composition of enteral formula; and
(vi) Validation that the device does not impede flow alarms on enteral feeding pumps.
(4) Human factors testing must be performed to characterize use error risks.
(5) Performance data must support shelf life by demonstrating package integrity and device functionality over the identified shelf life.
(6) Labeling must include the following:
(i) A detailed summary of
in vivo testing pertinent to use of the device, including device-related adverse events;(ii) A detailed summary of compatible formulas that is supported by non-clinical testing, including the expected enzymatic conversion as a percentage;
(iii) Detailed instructions on how to place the device into an enteral feeding circuit;
(iv) A warning regarding the possibility for misconnections; and
(v) Expiration date or shelf life.
(7) Patient labeling must be provided and must include:
(i) Relevant warnings, precautions, adverse effects, and complications;
(ii) A description of the device and how it operates;
(iii) Instructions on how to correctly use the device; and
(iv) The benefits and risks associated with the use of the device.