Method Validation for Quantitative Heavy Metals Testing – USP Proposed <232> and <233> in an Injectable Drug Substance
Authors: Tania Russell, Assistant Manager and Nikki Schopp, Team Leader, Analytical Services Department, SGS Life Science Services, Lincolnshire, USA
In 2008 the US Pharmacopeia (USP) began proposing changes to the existing USP <231> Heavy Metals Tests. These changes are necessary as the current USP <231> and its methodology are out of date. USP <231> is a color-based method that is only sensitive enough for certain elements (Pb, Hg, Bi, As, Sb, Sn, Cd, As, Cu and Mo). The test is not specific, nor does it provide adequate recovery of the elements being tested.
The USP is continuing to modify a proposal for using inductively coupled plasma (ICP) and ICP-mass spectroscopy (ICP-MS) to detect contaminating heavy metals in drug products. The new USP <233> describes testing methods to detect elements of interest specifically, accurately and with high sensitivity, enabling detection at levels of ppb and lower. Although incorporation of these changes into the USP has been postponed, the current goal is to harmonize USP, European Pharmacopoeia (EP) and ICH methods. The proposed USP methods would test for Class 1 (As, Cd, Hg and Pb) and Class 2 (Cu, Cr, Ir, Mo, Ni, Os, Pd, Pt, Rh, Ru and V) elemental impurities. The suggested EP methods would test for class 1A (Pt and Pd), Class 1B (Ir, Rh, Ru, Os), Class 1C (Mo, Ni, Cr, B), Class 2 (Cu and Mn) and Class 3 (Fe and Zn) elemental impurities. Lastly, the proposed International Conference on Harmonization (ICH) methods would test for Class 1 (As, Pb, Cd, and Hg), Class 2A (V, Mo, Se and Co), Class 2B (Ag, Au,Tl, Pd, Pt, Ir, Os, Rh and Ru), Class 3 (Sb, Ba, Li, Cr, Cu, Sn and Ni) and Class 4 (B, Fe, Zn, K, Ca, Na, Mn, Mg, W and Al) elemental impurities.
USP testing can either be validated as a limit test or quantitative test. Limit test validation involves performing accuracy and repeatability testing at the limit and specificity. Quantitative validation involves assessing accuracy at 50%, 100% and 150% of the limit, repeatability at the limit, specificity and limit of quantitation (at 50% of the limit).
Determining sample preparation and analysis on the instrument raises certain challenges that need to be addressed. Sample preparation needs to be complete and the sample must contain relatively low levels of dissolved solids. Instrument methods need to take into consideration possible interferences and strategies for overcoming any interferences encountered. Sample preparation is crucial when validating a quantitative method.
Determining a Sample Preparation
The USP proposes neat, direct or indirect testing methods. The direct testing method is used for assessing typical injectable drug products; however, multiple sample preparations are occasionally necessary. For the direct method, the sample is prepared in a dilute acid and analyzed directly on the instrument. In contrast, the indirect method involves first a closed vessel microwave digestion step, which is then followed by analysis on the instrument. Nitric acid and hydrochloric acid are the most commonly used acids.
When choosing acids for direct or indirect sample preparation, a few considerations are necessary. Using ICP-MS to detect arsenic at low levels in chloride matrices is often difficult because of the 40Ar35Cl+ interference in the plasma. Because 40Ar35Cl+ has the same mass-to-charge ratio as arsenic, its formation causes a positive interference for the detection of arsenic. During microwave digestion, nitric acid is the best oxidizer for arsenic and thus is most commonly used; however, some elements are more compatible with hydrochloric acid than nitric acid. Osmium is difficult to work with due to its high volatility and conversion to osmium tetroxide in the presence of nitric acid, which may result in false readings. Additionally, when testing for mercury, a stabilizer such as gold (III) chloride (AuCl3) is typically added to maintain mercury as the mercuric ion and to prevent reduction to elemental mercury. Hence, while a single sample preparation method would be ideal, it is generally unlikely that only one will suffice. For the particular injectable drug product discussed here, As, Hg, Ru, Cd, Pb, Cr, Mo, Pd, and V were tested using a direct method with nitric acid; osmium was tested using a direct method with hydrochloric acid; and Ni, Cu, Ir, Pt and Rh were tested using an indirect microwave digestion.
Validation Requirements and Results for a Quantitative Test
Table 1 summarizes the parenteral limits as defined in the proposed USP <232> and <233> chapters.
The method validation data for the ICP-MS experiments performed in our laboratory demonstrate accuracy (Table 2), precision (Table 3), ruggedness (Table 4), and specificity (Table 5).
Limit of quantitation (LOQ), range and linearity are demonstrated by meeting the accuracy requirements.
The USP proposed <232> and <233> changes will require ICP and ICP-MS instrumentation and methodology expertise, as well as a broad depth of sample preparation knowledge. Instrument set-up, the system for sample introduction and plasma discharge are integral aspects of successful method validation. Also important is formulating strategies to overcome sample interferences and achieve acceptable elemental recovery criteria. SGS has the level of expertise necessary to overcome multiple interference types, including high arsenic and osmium recoveries, unstable mercury values and difficult sample preparations, including the use of Hydrofluoric Acid (HF) in the digestion process, just to name a few.
SGS has successfully validated specificity, linearity, accuracy, precision and ruggedness in the methods used for several types of samples, including injectable drug products. SGS’s extensive experience with the USP-proposed methodology will ultimately ensure a smooth transition for implementation of the new USP, EP and ICH requirements.
- USP General Chapter <232>, “Elemental Impurities – Limits”, USP 36 / NF 31, pp. 151, 6136
- USP General Chapter <233>, “Elemental Impurities – Procedures”, USP 36 / NF 31, pp. 153, 6138