Incurred Sample Reanalysis (ISR)
Bioanalytical assays are widely used to quantify drugs and their metabolites in a variety of biological matrices. Incurred sample reanalysis (ISR) has steadily gained popularity as a quality assessment tool for bioanalytical assays. ISR is repeated measurement of analyte within a selected sample set from dosed subjects in separate runs on different days to establish result reproducibility.
Method validation in pharmacokinetic and bioanalytical studies ensures high quality findings by minimizing bias and error. The validation process includes assessment of selectivity, specificity, sensitivity, accuracy, precision, recovery, and stability of analyte. We conduct these validations on quality control (QC) samples prepared by spiking drug-free matrix from human volunteers or animals with reference compound of known identity and concentration.
QC samples, however, may not always mimic test compound activity in incurred samples due to several reasons. For example, lack of compound recovery, sample non-homogeneity, metabolite presence, protein binding, and matrix suppression effects etc.
First discussions for the need of ISR guidelines, prompted by wide variability of reanalyzed samples compared to original results in some studies as observed by FDA, were held in May 2006 at the AAPS / FDA Bioanalytical Workshop entitled “Quantitative Bioanalytical Methods Validation and Implementation: Best Practices for Chromatographic and Ligand Binding Assays”. Afterwards, industry leaders met to discuss general procedures for ISR in Feb 2008 at the AAPS/FDA Bioanalytical Workshop entitled “AAPS workshop on current topics in GLP bioanalysis: Assay reproducibility for incurred samples – implications of Crystal City recommendations”, and subsequently published expectations for ISR conduct, documentation, and reporting in Jun 2009. European Medicines Agency first provided regulatory guidance on ISR. FDA followed in 2013 through publishing of its own draft document.
FDA’s ISR Mandate
- Require ISR for all pivotal PK / PD and in vivo human BE studies; ISR at least once for each method and species during non-clinical safety studies
- Incorporate ISR into method development and validation pilot to allow for remediation prior to pivotal study; monitor ISR in larger number and variety for pivotal PK / PD studies
- Select up to 10% of study samples for ISR
- Sample selection to ensure adequate coverage of entire PK / PD profile of all subjects, including assessments from time points around Cmax and end of elimination phase
- Analyze using the original bioanalytical method (including number of replicates and any applied dilution factor), and compare ISR samples to freshly prepared calibrators
- At least two-thirds (67%) of ISR results should be within 20% of original results for small molecules and within 30% of original results for large molecules; percentage difference calculated as (Repeat – Original)*100 / Mean
- No regulatory requirement for additional analyses, if an individual sample does not meet the above acceptance criteria; scientific judgment expected in case difference is substantial
- Halt analyses and perform / document investigation, if overall ISR does not meet acceptance criteria
- Document all ISR evaluations to reconstruct the study conduct and any follow up investigations
We conducted ISR analysis on a chemotherapy drug Capecitabine used in treatment of many types of cancers, such as breast, colon, rectal, stomach, oesophageal and pancreatic cancers. The subject drug is rationally designed 5-fluorouracil (FU) prodrug that mimics continuous infusion of 5-FU while avoiding complications / inconvenience of intravenous administration. This antimetabolite drug gets absorbed intact from the gastrointestinal tract, converted enzymatically to active 5-FU, and released directly into the tumour. Therefore, accurate measurement of 5-FU is an essential study.
We measured both Capecitabine and its metabolite 5-FU. For sample preparation, we put together standards and QCs containing both the analytes and internal standards. Our analysts ran separate LC-MS/MS for the two analytes due to significant retention time difference. Initially, ISR analysis passed for the parent but failed for the metabolite
During the ISR analysis, we noticed that the current bioanalytical method can accurately measure Capecitabine but the observed 5-FU levels were highly variable. Therefore, we initiated an investigation to find out the cause for this high variability and increased concentration of 5-FU. Note that the prodrug does not metabolizes to 5-FU directly, but goes through intermediate metabolism, as noted.
Capecitabine à 5’-deoxy-5-fluorocytidine (5’-DFCR) à 5’-deoxy-5-fluorourindine (5’-DFUR) à 5-FU
Additionally, there are specific enzymes involved in all the transformations
- Carboxylesterase in the liver (5’–DFCR)
- Cytidine deaminase in the liver and tumor tissue (5’-DFUR)
- Thymidine phosphorylase (TP) in tumor tissue (5-FU)
Hence, it was necessary to measure other intermediate metabolites to understand the issue. We performed several experiments to deduce the reason for 5-FU ISR failure. This was primarily due to further conversion of other intermediate metabolites to 5-FU as well, resulting in its way higher concentrations during ISR.