Donald A. Johnson, PhD, RAC is senior scientific advisor at Eurofins BioPharma Product Testing.

Donald A. Johnson, PhD, RAC

	Submitted: August 23, 2019; Accepted: October 16, 2019

Soft gelatin (softgel) capsules can be a means of achieving bioavailability of highly lipophilic drugs that are practically water insoluble. The API is generally dissolved in edible oil. Typical oils used in softgel capsules are medium-chain triglycerides such as Miglyol (IOI Oleochemical) and mono- or di-glycerides such as Capmul (Abitec). For softgel capsules containing lipophilic drugs, the Division of Bioequivalence (DBE), in the Office of Generic Drugs, Center for Drug Evaluation and Research, FDA, will ask applicants to submit a “quantitative rupture”

 drug release test to measure the drug released in the dissolution medium after the capsule shell ruptures (1). The method will typically use a dissolution apparatus 1 (basket) or apparatus 2 (paddle) described in 

) <711> Dissolution. In this paper, quantitative rupture testing and dissolution testing will mean the same thing for softgel capsules. Even when the drug is already completely dissolved in the capsule fill, the capsule shell still needs to dissolve for drug release.

	A suitable surfactant at an appropriate concentration in the aqueous dissolution medium is a key element for quantitative rupture testing of these softgel formulations. Developing a quantitative rupture test for capsules containing lipophilic drugs formulated in oils presents many challenges. Cross-linking of the capsule shell is the most common challenge. With time, cross-linking will increase, and drug release values may fall below established specifications. Tier 2 release testing with digestive enzymes is an acceptable solution to cross-linking of gelatin capsules (2). During Tier 2 testing of softgel capsules that have cross-linked, an occasional aberrant high value for a capsule may be found. This could be, at times, more than 200% of label claim. This paper discusses what causes cross-linking, how cross-linking is addressed with addition of enzymes, and consideration for the occasional high results that can be obtained during release testing.

	Physicochemical properties of drugs in softgel products

	There are two important properties that must be taken into account for getting an oral drug absorbed through the walls of the gastrointestinal tract and into the body: solubility and permeability. If a drug is insoluble in aqueous media, the drug can have very limited bioavailability because there is not a significant driving force for partitioning of the drug from the aqueous environment of the intestinal lumen into the intestinal membrane. On the other hand, if a drug is too soluble in aqueous media, it may not want to partition into the lipid layers of the intestinal membrane. The solubility/permeability trade-off is the basis for the Biopharmaceutics Classification System (BCS) (3).

	For drugs that are highly lipophilic, a softgel capsule product can be a useful dosage form for increasing bioavailability of the drug. For the purpose of this paper, the softgel products discussed are those in which the capsule fill contains the drug completely dissolved in oil.  The drug will have a high concentration in the oil phase relative to an aqueous medium as indicated by the partition coefficients of the three APIs shown in 

. Without the oil phase, the concentration of the drug in the aqueous environment of the intestinal lumen is too low to provide a driving force for intestinal membrane permeability.

	With softgel products, the drug is already dissolved in the oil fill of the product and drug dissolution is not a physical process that occurs during rupture testing of softgel capsules. Many still refer to release of drug form softgel products as dissolution testing because the quantitative rupture testing generally follows 

 <711>. It is the capsule shell that dissolves in order for the drug to be released.

 shows typical physicochemical properties of three softgel capsule products that have the drug dissolved in an oil fill.

	It is seen in Table I that the aqueous solubilities of the three drugs are very low, ranging from nanograms per milliliter to micrograms per milliliter. The solubilities in lipid media relative to their solubilities in aqueous media are high for the three drugs, as indicated by their high partition coefficients, P. The partition coefficient measures the concentration ratio of a drug at equilibrium in a lipid medium and aqueous medium, where the two media are immiscible but in contact. Partition coefficients are commonly determined with octanol and water.  In 

, the partition coefficients show that these drugs have solubility in lipid media that range from approximately 100,000 to more than 10 million times the solubility in aqueous media. Drugs with large partition coefficients would have high permeabilities across the intestinal membrane, thus drug dissolved in an oil fill of a softgel would be an ideal dosage form for drugs with similar properties to those in 

	Impact of cross-linking on drug release of softgel products

	Gelatin is derived from the partial hydrolysis of collagen. Gelatin crosslinking occurs due to chemical reactions between the peptide chains of gelatin. Once formed in a softgel capsule shell, the capsule shell will only rupture in the presence of proteolytic enzymes if the cross-linking is significant.

	Cross-linking is a common problem encountered in the dissolution of gelatin capsules and is most commonly seen during stability testing. Low and incomplete dissolution may be observed while performing 

 release testing, as the capsule shell may dissolve slowly and incompletely, delaying full release beyond the specified sampling time point of the test method. If severe cross-linking occurs, bioavailability issues may also arise.

	Cross-linking may not occur evenly across a container of drug product. 

show typical results that may be seen during release testing. In both cases, only one out of six capsules tested has appreciable cross-linking. 

 shows a case where a single low value is obtained. This result is more common in cases where no enzyme is added to the dissolution medium (Tier 1 testing as explained in the following section). 

, an aberrant high result is obtained in one capsule. This result is commonly seen in testing with enzymes (Tier 2 testing).

	When a softgel product fails dissolution acceptance criteria due to cross-linking, 

 allows the use of enzymes in the dissolution medium and allows for two-tier dissolution testing. In Tier 2, pepsin is added to acidic or water dissolution medium to achieve an activity of 750,000 units or less per liter. Pancreatin 

 is added to a dissolution medium at or above pH 6.8 to achieve a protease activity of not more than 1750 units per liter.

	The two-tier dissolution testing is described in 

 <711>. Dissolution criteria for dissolution testing are expressed with a time point and a Q value. For example, a dissolution specification may be 75% (Q) at 45 minutes. Table IV shows the criteria for the two tiers in 

	In Tier 1, the test is performed with the normal dissolution medium of the test method. If the product fails Stage 3 criteria at any stage in Tier 1, testing progresses directly to Tier 2, where enzymes are added to the dissolution medium. For example, if cross-linking is significant, failure to meet Stage 3 criteria in Tier 1 may occur at Stage 1 testing. In this case, there is no need to perform Tier 1 stages 2 and 3. Testing should proceed  directly to Tier 2 testing. Most commonly, Tier 2 criteria are met at Stage 1 of Tier 2. Any failure to meet Tier 2 criteria at Tier 2 Stage 3 is a failure of the dissolution specification and indicates a problem for the product.

	Pretreatment of softgel capsules when using enzymes and surfactants

	During quantitative rupture testing of softgel capsules that contain an oil fill, a surfactant must be added to the dissolution medium in order to solubilize the oil once it is released from the capsule shell. Otherwise, when the capsule shell ruptures, the oil droplets will be released and simply float to the surface of the dissolution medium. The amount of surfactant added to the dissolution medium must be sufficient to form micelles. The lipophilic core of the micelle  is able to take up the oil-containing drug released from the capsule upon rupture.

	Although surfactants are a necessary component in quantitative rupture testing, they are known to inhibit the activity of enzymes used in the dissolution medium when Tier 2 testing is performed.

	To prevent inactivation of the enzymes in Tier 2 testing, pretreatment of the cross-linked gelatin capsules with the medium containing the enzyme but not the surfactant must be performed. Pretreatment is performed by adding the enzyme to a portion of the dissolution medium without the surfactant and stirring the medium in the dissolution vessel for a short period of time, usually not more than 15 minutes. After this period, the remaining portion of the dissolution medium containing the surfactant is added to the dissolution vessel. After addition of the remaining dissolution medium, the final concentration of the surfactant in the vessel and the final dissolution medium volume will be the same as specified in the test method. The pretreatment time is included in the total run of the dissolution test method. For example, if the sample time of the test method is 45 min and the pretreatment time was 15 minutes, the sample time will be 30 minutes after the addition of the surfactant to the medium, thus maintaining the sample time at 45 minutes as stated in the method.

	A few of the commonly used surfactants used in dissolution testing are sodium dodecyl sulfate, Triton-X-100 (polyethylene glycol tert-octylphenyl ether, Sigma Aldrich), and lauryldimethylamine N-oxide.

	Physical explanation of aberrant results

	When low release values are found in Tier 1 testing due to cross-linking, the addition of enzymes will almost always prove successful in obtaining results within specification. Only with the most severe degree of cross-linking will out-of-specification (OOS) results be obtained in Tier 2 testing. It should be noted, as shown in 

 criteria for two-tiered testing does not have upper limits. OOS results are for low release values below the specification limits. Any result obtained above 125%, for example, would be considered aberrant and may require an investigation.

 shows actual data obtained in quantitative rupture testing of a pharmaceutical product. The specification for this product is 75% (Q) at 45 minutes. It is seen that the criterion for Tier 1 Stage 3 was not met at Tier 1 Stage 1 because there are release values below 50% (Q-25%). There was no need to continue with stages 2 and 3 of Tier 1 because no additional testing could result in passing Stage 3 criteria. Tier 2 testing resulted in acceptable data at Stage 1.

 are data from another softgel product on stability after nine months at 25 °C/60% relative humidity (RH) conditions. The release specification of the product is 80% (Q) at 45 minutes. The product fails dissolution in Tier 1 testing at Stage 1. Although the product passes the criteria for Tier 2 testing at Stage 1 as shown in 

, there are some high aberrant values at the sampling time point of the test method (i.e., 45 minutes). An additional sample was taken at 75 minutes, and the results were acceptable values for the product. The capsules are not super-potent and non-uniform as might be suggested by the 45-minute sample. The high results obtained at 45 minutes are explained by the difference between the time the oil-containing drug is released from the ruptured capsule and the time the oil droplets are taken up by the surfactant micelles.

	Surfactants form micelles when added to the dissolution media above their critical micelle concentration. Micelles are spherical in shape and have diameters on the order of 10–50 nanometers.  During rupture testing, an oil droplet emerging from the capsule shell will have a diameter on the order of 0.25–0.5 millimeters. A micelle with a diameter of 30 nm has a volume of 1.4 x 10-20 mm

. An oil droplet with a diameter of 0.25 mm has a volume of 8.2 x 10-3 mm

. The volume of the oil droplet is on the order of a quadrillion times that of the micelle volume. However, there are many micelles capable of taking up all the oil released from the capsule. The oil in a typical softgel fill will be from 250–350 mg. The amount of surfactant added to the dissolution medium is on the order of 10s of grams. 

 shows examples of dissolution methods from the FDA database for softgel capsules.

	When an oil droplet is released from the capsule, there is a finite amount of time it takes for the tiny micelles to take up the large oil droplets. If the time when an oil droplet is released from the capsule shell is too close to the sampling time, the drug distribution in the dissolution medium may not be homogeneous because the oil droplets have not been fully taken up and absorbed by the micelles. In this case, sample taken may include an oil droplet not yet taken up by the micelles, resulting in high aberrant results.

	As an example, say the sample volume is 1 mL and the dissolution medium is 1000 mL. If a drug product contains 2 mg of API in 250 mg of oil fill, the even distribution of drug at full release in the dissolution medium would be 2 µg/mL.  But, if an oil droplet containing the drug has not been fully taken up by the micelles because the time between capsule rupture and sample time is insufficient, high release results will be obtained if any small oil droplet is included in the sample volume. Two mg of API in a 250 mg of capsule fill is 8 mg/mL. This means a droplet size of only 0.25 µL taken up during sampling will contain 2 µg of drug result in a drug release of approximately 200% label claim. A 0.25% droplet in a 1 mL sample size would be hard to visually detect in the sample.

	Dissolution testing of softgel capsules is better described as quantitative rupture testing because the drug is already dissolved in the oil fill of the drug product. The critical dissolution that occurs in quantitative rupture testing of softgel products is dissolution of the capsule gelatin shell. When cross-linking of the gelatin shell occurs, release of the drug from the capsule will be retarded. The impact of cross-linking on bioavailability is not significant if it can be shown that the drug release is overcome by addition of enzymes to the dissolution medium. If during Tier 2 release testing, drug release is below specifications, cross-linking is most likely severe enough to affect bioavailability. If, on the other hand, aberrant high dissolution results are obtained, the high results are indicative of release of drug from the capsule at a rupture time that is close to the sample time. Subsequent sampling beyond the sample time of the test method will result in assay values more in line with the product label claim. 

	It is important to understand results obtained from quantitative rupture testing of softgel capsules. Aberrant high values during quantitative rupture testing should not be interpreted as super-potent drug product or non-uniformity of dosage units. However, aberrant high results may still require careful consideration and possibly a laboratory investigation. Low results obtained during Tier 2 testing with enzymes may signal a possible problem with 

 bioavailability. Experience with quantitative rupture testing is especially important during stability testing in understanding the performance of the drug product.

, 13 (3), pp. 328-335 (September 2011).
	2. USP, <711> Dissolution, 

 July 26, 2013.
	3. FDA, The Biopharmaceutics Classification System (BCS) Guidance, 

, accessed Jan. 17, 2020.
	4. NIH PubChem Database, 

, accessed Jan. 24, 2020.
	5. NIH PubChem Database, 

, accessed Jan. 24, 2020.
	6. NIH PubChem Database, 

, accessed Jan. 24, 2020.
	7. FDA, Dissolution Methods, Database, 


	Vol. 44, No. 2
	February 2020
	Pages: 38-42

	When referring to this article, please cite it as D. A. Johnson, "Quantitative Rupture Testing of Soft Gelatin Capsules: Understanding Aberrant Results," 

Articles

This paper discusses what causes cross-linking, how cross-linking is addressed with addition of enzymes, and consideration for occasional high results that can be obtained during release testing.