Mass balance correlates the measured loss of a parent drug to the measured increase

in the amount of degradation products. It is a good quality control check on analytical

methods to show that all degradation products are adequately detected and do not
interfere with quantitation of the parent drug (i.e., stability-indicating methods).
Regulatory agencies use mass balance to assess the appropriateness of the analytical
method as a stability-indicating method and determine whether all degradants have
been accounted for (1–4).

In mass balance calculations, the loss of parent drug or the amount of drug remaining is
determined from a sample assay, and the measured increase in degradation products is
determined by a degradation method. The fundamental approach for determining mass
balance is to quantitate the degradation peaks using degradation methods and then
reconcile the measured loss in the parent drug with the amount of degradation products.
If the loss in potency can be reasonably accounted for by the amount of degradants
measured, then mass balance is achieved.

It has come to our attention that under certain circumstances it is not appropriate to
estimate mass balance directly from the amount of degradation products formed but
rather the percent parent drug lost. This percentage, which corresponds to the amount
of degradant formed, is the ultimate quantity that should be used to reconcile mass
balance. The amount of degradant is converted to the corresponding percent parent
drug degraded by means of the ratio of molecular weight of the parent drug relative to
that of the degradant. Thus, the operative entity used in reconciling mass balance is the
percent of parent drug degraded. In this article, we present two scenarios to illustrate
the necessity of making this conversion.

Scenario 1: Molecule degrades to form a lower molecular weight degradant

Because of the paucity of structural information about the degradation products of drugs
and their unavailability in early phases of drug development, scientists quantitate
degradants using area-percent normalization (area %). There is an underlying
assumption of an identical response factor (area relative to amount) for parent drug and
degradant. Discrepancies in mass balance when degradants are measured by area
normalization have been rightly explained by citing differences in detector response for
the parent drug and degradant. This is particularly true for analyses involving ultraviolet
(UV) detection. Appropriate correction often is applied to area percent, including using
relative UV response to convert area percent to weight–weight percent (5).

An overlooked cause for possible discrepancies in mass balance is the difference in

molecular weight between the parent drug and the degradant. To get the percent of
parent drug degraded, which is the useful term in a mass balance calculation, one often
assumes, sometimes unknowingly, that the percent of degradant formed corresponds to
the percent parent drug degraded. This assumption would be invalid if the degradant
had a molecular weight that differed significantly from that of the parent drug.
Consider the case in which drug P degrades to form A, a degradant with a lower
molecular weight:

One example is the hydrolysis reaction shown in Figure 1, in which the ester parent
drug could degrade by hydrolysis to produce an acid hydrolysis product with a different
molecular weight relative to the parent drug. The assumption that the amount (i.e.,
percent) of degradant is equal in numerical value to the corresponding percent of drug
degraded is incorrect, as explained in the following section.

Differing detector response. Consider the case in which the amount of degradant is
determined by area normalization and the detector response of degradant A differs from
that of the parent drug P. To convert the percent degradant to percent parent drug
degraded, two corrections must be applied to the former result. First, a relative
response factor is needed to convert area percent results to weight–weight (wt/wt)
percent results. The second correction, which often is overlooked, is the need to correct
for the difference in molecular weight between the parent and the degradant as follows:

percent drug degraded = percent degradant (wt/wt) X (Mp ÷ MD)

in which Mp is the molecular weight of parent drug and MD is the molecular weight of
degradant.

For example, if the amount of degradant quantitated is 0.5% (wt/wt), then the
corresponding amount of parent drug degraded is:

percent parent drug degraded = 0.5% × (Mp ÷ MD)

and not 0.5%. Therefore, the number that should be reconciled with the potency of the
drug to determine mass balance is (0.5% X [Mp ÷ MD]). That is, if the loss in potency as
determined from the assay is equal to (0.5% X [Mp ÷ MD]), then there is perfect mass
balance.

This second correction is necessary because the response factors calculated on the
basis of concentration assume equal moles for the same mass of compound. Because
the number of moles is a function of molecular weight, a significant difference in
molecular weight can lead to errors in the estimation of percent degraded. When the
molecular weights of a parent drug and a degradant are similar, then the ratio of
molecular weights (Mp ÷ MD) is unity, and the percent degraded is equal to the percent
degradant. Thus, the amount of degradant is equal to percent drug degraded.

Table I illustrates the effect of molecular-weight differences between the parent drug
and the degradant on the error in a mass balance calculation. The analysis is a
simulation of a drug with a molecular weight of 300 Da and degradants with molecular
weights of 300, 250, and 350 Da. Error analyses are completed assuming degradants at
levels of 0.5, 1.0, and 5.0%.

Table I: The effect of differences in molecular weight of drug and degradant on the error
in mass balance calculation.

Table I shows that the error in mass balance calculation increases as the difference in
molecular weight between the parent drug and the degradant increases. It also shows
that the percent error is almost independent of the amount of degradant (at typical
levels expected in pharmaceutical drugs). Thus, even in cases in which the level of
degradant is low, the correction advocated in this article is appropriate and warranted.

Equivalent detector response. When the detector response is equivalent between the
parent drug and the degradants, then area percent results can translate directly into
weight–weight results. Thus, only a correction for molecular weight must be done as
previously described.
Scenario 2: Molecule degrades by forming an adduct with a high molecular
weight excipient

Consider the example in which the parent drug P forms an adduct with excipient E, as
follows:

in which P – E is the degradant.

For example, consider the condensation reaction between

a drug with a primary amine functionality and citric acid
(see Figure 2). In this case, the parent drug undergoes a condensation reaction with
citric acid to form an amide citrate adduct.

In cases in which the molecular weight of the adduct is significantly greater than the
parent drug (as is the case just described), the percent of degradant calculated on a
weight–weight basis is not equal to the percent parent degraded. Rather, the calculated
percent degradant should be corrected for the difference in molecular weight as
illustrated below:

percent degraded = percent degradant (wt/wt) X (Mp ÷ MD)

To further illustrate this point, consider the case in which the molecular weight of P – E
is twice that of P and P degrades in a stability study to give 2% degradant (P – E). Mass
balance is reconciled after converting the amount of degradant to percent parent
degraded as follows:

percent parent degraded = 2% X (Mp ÷ MD) = 1.0%

For mass balance to hold, the assay result (amount of drug remaining) is 100 – 1.0%
and not (100 – 2.0%).
These scenarios demonstrate that when the molecular weight of the parent drug differs
significantly from that of the degradant, proper attention should be paid to the
implications of this difference and appropriate corrections should be carried out to
obtain a good mass-balance result. This process involves converting the percent
degradant result to the corresponding percent parent degraded, because percent parent
degraded is the operative number that should be accounted for in a mass balance
calculation.

Acknowledgment

The authors gratefully acknowledge the valuable insights provided by Bruce Johnson
and Steve Hagen of Pfizer, Inc. in their review of this article. We also thank Brian
Weekley, Peter Angus, and Derek Jackson of Pfizer, Inc. and Loren Wrisley of Wyeth
for valuable discussions over the years on the subject of mass balance.