7.1 Analytes that are also Endogenous Compounds 1024
For analytes that are also endogenous compounds, the accuracy of the measurement of the 1025
analytes poses a challenge when the assay cannot distinguish between the therapeutic agent and 1026
the endogenous counterpart.
1027
The endogenous levels may vary because of age, gender, diurnal variations, illness or as a side 1028
effect of drug treatment. If available, biological matrix with an adequate signal-to-noise ratio 1029
(i.e., endogenous level sufficiently low for the desired LLOQ, e.g., <20% of the LLOQ) should 1030
be used as blank matrix to prepare calibration standards and QCs since the biological matrix 1031
used to prepare calibration standards and QCs should be the same as the study samples (i.e., 1032
authentic biological matrix) and should be free of matrix effect and endogenous analyte at the 1033
level that causes interference.
1034
In those cases where matrices without interference are not available, there are four possible 1035
approaches to calculate the concentration of the endogenous analyte in calibration standards, 1036
QCs and, consequently, study samples: 1) the standard addition approach, 2) the background 1037
subtraction approach, 3) the surrogate matrix (neat, artificial or stripped matrices) approach and 1038
4) the surrogate analyte approach.
1039
1) Standard Addition Approach:
1040
Every study sample is divided into aliquots of equal volume. All aliquots, but one, 1041
are separately spiked with known and varying amounts of the analyte standards to 1042
construct a calibration curve for every study sample. The study sample concentration 1043
is then determined as the negative x-intercept of the standard calibration curve 1044
prepared in that particular study sample.
1045
2) Background Subtraction Approach:
1046
The endogenous background concentrations of analytes in a pooled/representative 1047
matrix are subtracted from the concentrations of the added standards, subsequently 1048
the subtracted concentrations are used to construct the calibration curve.
1049
3) Surrogate Matrix Approach:
1050
The matrix of the study samples is substituted by a surrogate matrix. Surrogate 1051
matrices can vary widely in complexity from simple buffers or artificial matrices that 1052
try to mimic the authentic one, to stripped matrices.
1053
4) Surrogate Analyte Approach:
1054
Stable-isotope labelled analytes are used as surrogate standards to construct the 1055
calibration curves for the quantification of endogenous analytes. In this method it is 1056
assumed that the physicochemical properties of the authentic and surrogates analytes 1057
are the same with the exception of molecular weight. However, isotope standards may 1058
differ in retention time and MS sensitivity, therefore, before application of this 1059
approach, the ratio of the labelled to unlabelled analyte MS responses (i.e., the 1060
response factor) should be close to unity and constant over the entire calibration range.
1061
If the response factor does not comply with these requirements, it should be 1062
incorporated into the regression equation of the calibration curve.
1063
Validation of an analytical method for an analyte that is also an endogenous compound will 1064
require the following considerations.
1065
7.1.1 Quality Control Samples 1066
The endogenous concentrations of the analyte in the biological matrix should be evaluated prior 1067
to QC preparation (e.g., by replicate analysis). The blank matrices with the minimum level of 1068
the endogenous analyte should be used. The concentrations of the QCs should account for the 1069
endogenous concentrations in the biological matrix (i.e., additive) and be representative of the 1070
expected study concentrations.
1071
The QCs used for validation should be aliquots of the authentic biological matrix unspiked and 1072
spiked with known amounts of the authentic analyte. In spiked samples, the added amount 1073
should be enough to provide concentrations that are statistically different from the endogenous 1074
concentration.
1075
7.1.2 Calibration Standards 1076
In the Surrogate Matrix and Surrogate Analyte Approaches, these surrogates should be used 1077
only for the preparation of the calibration standards.
1078
In the Standard Addition and Background Subtraction Approaches the same biological matrix 1079
and analyte as the study samples is used to prepare the calibration standards. However, when 1080
the background concentrations are lowered by dilution of the blank matrices before spiking 1081
with the standards (e.g., if a lower LLOQ is required in the Background Subtraction Approach) 1082
the composition of the matrices in the study samples and the calibration standards is different, 1083
which may cause different recoveries and matrix effects.
1084
7.1.3 Selectivity, Recovery and Matrix Effects 1085
The assessment of selectivity is complicated by the absence of interference-free matrix. For 1086
chromatography, peak purity should be investigated as part of method validation by analysing 1087
matrices obtained from several donors using a discriminative detection system (e.g., tandem 1088
mass spectrometry (MS/MS)). Other approaches, if justified by scientific principles, may also 1089
be considered.
1090
For the Standard Addition and Background Subtraction Approaches, as the same biological 1091
matrix and analyte are used for study samples and calibration standards, the same recovery and 1092
matrix effect occurs in the study samples and the calibration standards. For the Surrogate Matrix 1093
and Surrogate Analyte Approaches, the matrix effect and the extraction recovery may differ 1094
between calibration standards and study samples.
1095
• If the Surrogate Matrix Approach is used, demonstration of similar matrix effect and 1096
extraction recovery in both the surrogate and original matrix is required. This should be 1097
investigated in an experiment using QCs spiked with analyte in the matrix against the 1098
surrogate calibration curve and should be within ±15% for chromatographic assays and 1099
within ±20% for LBA assays.
1100
• If the Surrogate Analyte Approach is used, demonstration of similarity in matrix effect 1101
and recovery between surrogate and authentic endogenous analytes is required. This 1102
should be investigated in an experiment within ±15% for chromatographic assays and 1103
within ±20% for LBA assays.
1104
Since the composition of the biological matrix might affect method performance, it is necessary 1105
to investigate matrices from different donors, except in the Standard Addition Approach, where 1106
each sample is analysed with its own calibration curve.
1107
7.1.4 Parallelism 1108
Parallelism should be evaluated in the Surrogate Matrix and Surrogate Analyte Approaches by 1109
means of the Standard Addition approach, spike recovery or dilutional linearity.
1110
7.1.5 Accuracy and Precision 1111
In case of using a surrogate matrix or analyte, the assessment of accuracy and precision should 1112
be performed by analysing the QCs against the surrogate calibration curve. In certain cases, 1113
dilution of the QCs with surrogate matrix may be necessary. These experiments should be 1114
repeated with authentic biological matrices from different donors to address variability due to 1115
the matrix. Analysis of the unspiked QCs will give the mean endogenous background 1116
concentration and only precision and no accuracy can be determined for this QCs.
1117
The concentration of the endogenous substance in the blank sample may be determined and 1118
subtracted from the total concentrations observed in the spiked samples. Accuracy is 1119
recommended to be calculated using the following formula:
1120
𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 (%) = 100 ×(Measured concentration of spiked sample − endogenous concentration ) Nominal concentration
1121
7.1.6 Stability 1122
In order to mimic study samples as much as possible, stability experiments should be 1123
investigated with the authentic analyte in the authentic biological matrix and with unspiked and 1124
spiked samples. However, if a surrogate matrix is used for calibration standards, stability should 1125
also be demonstrated for the analyte in the surrogate matrix, as this could differ from stability 1126
in the authentic biological matrix.
1127
7.2 Parallelism 1128
Parallelism is defined as a parallel relationship between the calibration curve and serially 1129
diluted study samples to detect any influence of dilution on analyte measurement. Although 1130
lack of parallelism is a rare occurrence for PK assays, parallelism of LBA should be evaluated 1131
on a case-by-case basis, e.g., where interference caused by a matrix component (e.g., presence 1132
of endogenous binding protein) is suspected during study sample analysis. Parallelism 1133
investigation or the justification for its absence should be included in the Bioanalytical Report.
1134
As parallelism assessments are rarely possible during method development and method 1135
validation due to the unavailability of study samples and parallelism is strictly linked to the 1136
study samples (i.e., an assay may have perfectly suitable parallelism for a certain population of 1137
samples, yet lack it for another population), these experiments should be conducted during the 1138
analysis of the study samples. A high concentration study sample (preferably close to Cmax) 1139
should be diluted to at least three concentrations with blank matrix. The precision between 1140
samples in a dilution series should not exceed 30%. However, when applying the 30% criterion, 1141
data should be carefully monitored as results that pass this criterion may still reveal trends of 1142
non-parallelism. In the case that the sample does not dilute linearly (i.e., in a non-parallel 1143
manner), a procedure for reporting a result should be defined a priori.
1144
7.3 Recovery 1145
For methods that employ sample extraction, the recovery (extraction efficiency) should be 1146
evaluated. Recovery is reported as a percentage of the known amount of an analyte carried 1147
through the sample extraction and processing steps of the method. Recovery is determined by 1148
comparing the analyte response in a biological sample that is spiked with the analyte and 1149
processed, with the response in a biological blank sample that is processed and then spiked with 1150
the analyte. Recovery of the analyte does not need to be 100%, but the extent of recovery of an 1151
analyte and of the IS (if used) should be consistent. Recovery experiments are recommended to 1152
be performed by comparing the analytical results for extracted samples at multiple 1153
concentrations, typically three concentrations (low, medium and high).
1154
7.4 Minimum Required Dilution 1155
MRD is a dilution factor employed in samples that are diluted with buffer solution to reduce 1156
the background signal or matrix interference on the analysis using LBA. The MRD should be 1157
identical for all samples including calibration standards and the QCs and it should be 1158
determined during method development. If MRD is changed after establishment of the method, 1159
partial validation is necessary. MRD should be defined in the Validation Report of the analytical 1160
method.
1161
7.5 Commercial and Diagnostic Kits 1162
Commercial or diagnostic kits (referred to as kits) are sometimes co-developed with new drugs 1163
or therapeutic biological products for point-of-care patient diagnosis. The recommendations in 1164
this section of the guideline do not apply to the development of kits that are intended for point-1165
of-care patient diagnosis (e.g., companion or complimentary diagnostic kits). Refer to the 1166
appropriate guideline documents regarding regulatory expectations for the development of 1167
these kits.
1168
If an applicant repurposes a kit (instead of developing a new assay) or utilises “research use 1169
only” kits to measure chemical or biological drug concentrations during the development of a 1170
novel drug, the applicant should assess the kit validation to ensure that it conforms to the drug 1171
development standards described in this guideline.
1172
Validation considerations for kit assays include, but are not limited to, the following:
1173
• If the reference standard in the kit differs from that of the study samples, testing should 1174
evaluate differences in assay performance of the kit reagents. The specificity, accuracy, 1175
precision and stability of the assay should be demonstrated under actual conditions of use 1176
in the facility conducting the sample analysis. Modifications from kit processing 1177
instructions should be completely validated.
1178
• Kits that use sparse calibration standards (e.g., one- or two-point calibration curves) should 1179
include in-house validation experiments to establish the calibration curve with a sufficient 1180
number of standards across the calibration range.
1181
• Actual QC concentrations should be known. Concentrations of QCs expressed as ranges are 1182
not sufficient for quantitative applications. In such cases QCs with known concentrations 1183
should be prepared and used, independent of the kit-supplied QCs.
1184
• Calibration standards and QCs should be prepared in the same matrix as the study samples.
1185
Kits with calibration standards and QCs prepared in a matrix different from the study 1186
samples should be justified and appropriate experiments should be performed.
1187
• If multiple kit lots are used within a study, lot-to-lot variability and comparability should be 1188
addressed for any critical reagents included in the kits.
1189
• If a kit using multiple assay plates is employed, sufficient replicate QCs should be used on 1190
each plate to monitor the accuracy of the assay. Acceptance criteria should be established 1191
for the individual plates and for the overall analytical run.
1192
7.6 New or Alternative Technologies 1193
When a new or alternative technology is used as the sole bioanalytical technology from the 1194
onset of drug development, cross validation with an existing technology is not required.
1195
The use of two different bioanalytical technologies for the development of a drug may generate 1196
data for the same product that could be difficult to interpret. This outcome can occur when one 1197
platform generates drug concentrations that differ from those obtained with another platform.
1198
Therefore, when a new or alternative analytical platform is replacing a previous platform used 1199
in the development of a drug it is important that the potential differences are well understood.
1200
The data generated from the previous platform/technology should be cross validated to that of 1201
the new or alternative platform/technology. Seeking feedback from the regulatory authorities is 1202
encouraged early in drug development. The use of two methods or technologies within a 1203
comparative BA/BE study is strongly discouraged.
1204
The use of new technology in regulated bioanalysis should be supported by acceptance criteria 1205
established a priori based on method development and verified in validation.
1206
7.6.1 Dried Matrix Methods 1207
Dried matrix methods (DMM) is a sampling methodology that offers benefits such as collection 1208
of reduced blood sample volumes as a microsampling technique for drug analysis and ease of 1209
collection, storage and transportation. In addition to the typical methodological validation for 1210
LC-MS or LBA, use of DMM necessitates further validation of this sampling approach before 1211
using DMM in studies that support a regulatory application, such as:
1212
• Haematocrit (especially for spotting of whole blood into cards) 1213
• Sample homogeneity (especially for sub-punch of the sample on the card/device) 1214
• Reconstitution of the sample 1215
• DMM sample collection for ISR 1216
o Care should be taken to ensure sufficient sample volumes or numbers of 1217
replicates are retained for ISR 1218
o Should be assessed by multiple punches of the sample or samples should be 1219
taken in duplicate 1220
When DMM is used for clinical or nonclinical studies in addition to typical liquid approaches 1221
(e.g., liquid plasma samples) in the same studies, these two methods should be cross validated 1222
as described (Refer to Section 6.2). For nonclinical TK studies, refer to Section 4.1 of ICH S3A 1223
Q&A. Feedback from the appropriate regulatory authorities is encouraged in early drug 1224
development.
1225
8. DOCUMENTATION 1226
General and specific SOPs and good record keeping are essential to a properly validated 1227
analytical method. The data generated for bioanalytical method validation should be 1228
documented and available for data audit and inspection. Table 1 describes the recommended 1229
documentation for submission to the regulatory authorities and documentation that should be 1230
available at the analytical site at times of inspection. This documentation may be stored at the 1231
analytical site or at another secure location. In this case the documentation should be readily 1232
available when requested.
1233
All relevant documentation necessary for reconstructing the study as it was conducted and 1234
reported should be maintained in a secure environment. Relevant documentation includes, but 1235
is not limited to, source data, protocols and reports, records supporting procedural, operational, 1236
and environmental concerns and correspondence records between all involved parties.
1237
Regardless of the documentation format (i.e., paper or electronic), records should be 1238
contemporaneous with the event and subsequent alterations should not obscure the original data.
1239
The basis for changing or reprocessing data should be documented with sufficient detail, and 1240
the original record should be maintained. Transcripts/copies of data derived from analyses in 1241
biohazardous areas should be maintained if applicable.
1242
8.1 Summary Information 1243
Summary information should include the following items in Section 2.6.4/2.7.1 of the Common 1244
Technical Document (CTD) or reports:
1245
• A summary of assay methods used for each study should be included. Each summary 1246
should provide the protocol number, the assay type, the assay method identification 1247
code, the Bioanalytical Report code, effective date of the method, and the associated 1248
Validation Report codes.
1249
• A summary table of all the relevant Validation Reports should be provided for each 1250
analyte, including Partial Validation and Cross Validation Reports. The table should 1251
include the assay method identification code, the type of assay, the reason for the 1252
new method or additional validation (e.g., to lower the limit of quantification).
1253
Changes made to the method should be clearly identified.
1254
• A summary table cross-referencing multiple identification codes should be provided 1255
when an assay has different codes for the assay method, the Validation Reports and 1256
the Bioanalytical Reports.
1257
• Discussion of method changes in the protocol (e.g., evolution of methods, reason(s) 1258
for revisions, unique aspects) 1259
• For comparative BA/BE studies a list of regulatory site inspections including dates 1260
and outcomes for each analytical site if available.
1261
8.2 Documentation for Validation and Bioanalytical Reports 1262
Table 1 describes the recommended documentation for the Validation and Bioanalytical Reports.
1263
43 Table 1: Documentation and Reporting
Items Documentation at the Analytical Site Validation Report* Bioanalytical Report*
Chromatographic System Suitability
• Dates, times, and samples used for suitability testing
• Not applicable • Not applicable
Synopsis
Overview of Method Evolution
• History/evolution of methods (e.g., to explain revisions, unique aspects with supportive data, if available)
• Not applicable • Not applicable
Reference Standards • CoA or equivalent alternative to ensure quality (including purity), stability/expiration/retest date(s), batch number, and manufacturer or source
• Log records of receipt, use, and storage conditions.
• If expired, recertified CoA, or retest of quality and identity with retest dates
• A copy of the CoA or equivalent alternative including batch/lot number,
source, quality (including purity), storage conditions, and expiration/retest date, or table with this information.
• If expired, quality and stability at the time of use and retest dates and retested values.
• A copy of the CoA or equivalent alternative including batch /lot number, source, quality (including purity), storage conditions, and expiration/retest date or a table with this information.
• If expired, quality and stability at the time of use and retest dates and retested values.
Internal Standard • IS quality or demonstration of suitability
• Log records of receipt, use, and storage conditions
• Name of reagent or standard
• Origin
• Name of reagent or standard
• Origin
1264
44 Table 1 continued: Documentation and Reporting
Items Documentation at the Analytical Site Validation Report* Bioanalytical Report*
Critical Reagents • Name of reagent
• Batch/ Lot number
• Source/Origin
• Concentration, if applicable
• Retest date (expiry date)
• Storage conditions
• Name of reagent
• Batch/ Lot number
• Source/ Origin
• Retest date (expiry date)
• Storage conditions
• Name of reagent
• Batch/ Lot number
• Source/ Origin
• Retest date (expiry date)
• Storage conditions
Stock Solutions • Log of preparation, and use of stock solutions
• Storage location and condition
• Notation that solutions were used within stability period
• Stock solution stability
• Storage conditions
• Notation that solutions were used within stability period
• Stock solution stability †
• Storage conditions† Blank Matrix • Records of matrix descriptions, lot
numbers, receipt dates, storage conditions, and source/supplier
• Description, lot number, receipt dates
• Description, lot number, receipt dates††
Calibration
Standards and QCs
• Records and date of preparation
• Record of storage temperature (e.g., log of in/out dates, analyst, temperatures, and freezer(s))
• Description of preparation including matrix
• Batch number, preparation dates and stability period
• Storage conditions (temperatures, dates, duration,
etc.)
• Description of preparation†
• Preparation dates and stability period
• Storage conditions†
1265
45 Table 1 continued: Documentation and Reporting
Items Documentation at the Analytical Site Validation Report* Bioanalytical Report*
SOPs SOPs for all aspects of analysis, such as:
• Method/procedure (validation/analytical)
• Acceptance criteria (e.g., run, calibration curve, QCs)
• Instrumentation
• Reanalysis
• ISR
• Record of changes to SOP (change, date, reason, etc.)
• A detailed description of the assay procedure
• A list of SOPs/analytical protocols used for the assay procedure
Sample Tracking • Study sample receipt, and condition on receipt
• Records that indicate how samples were transported and received. Sample inventory and reasons for missing samples
• Location of storage (e.g., freezer unit)
• Tracking logs of QCs, calibration standards, and study samples
• Freezer logs for QCs, calibration standards, and study samples entry and exit
• Freezer logs for QCs, calibration standards, and study samples entry and exit