Impact of Preanalytical Factors on the Measurement of Tumor Tissue Biomarkers Using Immunohistochemistry.
Author(s): Bagchi A, Madaj Z, Engel KB, Guan P, Rohrer DC, Valley DR, Wolfrum E, Feenstra K, Roche N, Hostetter G, Moore HM, Jewell SD
Publication: J Histochem Cytochem, 2021, Vol. , Page 22155421995600
PubMed ID: 33641490 PubMed Review Paper? No
Purpose of Paper
The purpose of this paper was to assess the potential effects introduced by delay to fixation (DTF) and time in formalin (TIF) on immunohistochemical staining of cancer biomarkers in prospectively collected formalin-fixed, paraffin-embedded (FFPE) colon, kidney, lung, and ovarian tumor specimens under the National Cancer Institute's Pre-analytical Variable (BPV) program.
Conclusion of Paper
DTF induced effects on immunohistochemical staining were influenced by the tissue and antigen evaluated. While significant reductions in H-score (staining score x the percentage of stained cells/nuclei) relative to 1 h DTF controls were observed for paired box gene 8 (PAX8) and epidermal growth factor receptor (EGFR )after 2 h, reductions in PAX2, p53, and phosphorylated protein kinase B (pAKT) were not observed until a 3 h DTF, and reductions in mesenchymal epithelial transition factor (c-Met) and survivin were not significant until a 12 h DTF. Notably, renal cell carcinoma 1 (RCC1) displayed a 57% increase in H-score after a 12 h DTF relative to controls fixed after a 1 h delay. Similar, albeit nonsignificant, patterns of effect on H-score were also observed for affected antigens in other tissue types examined with smaller sample sizes. For example, similar onset and magnitudes of change were observed for PAX8 in ovary, c-Met in lung, and EGFR in colon and lung. For the most part, the DTF-induced alterations in H-score observed in BPV specimens translated to high estimated probabilities of a change in intensity score for the same antigens and DTFs. Nine antigens that were examined in DTF timecourse specimens as fresh FFPE blocks and those stored for 1 year displayed similar results.
Of the 13 antigens evaluated for TIF-associated effects, results were limited to reductions in H-scores for c-Met and cluster of differentiation 44 (CD44) of 23% and 15%, respectively, after a TIF of 72 h relative to 1 h DTF controls. Fixation for 72 h rather than 12 h also translated to an estimated probability of a reduction in intensity score of 51% and 25% for c-Met and CD44, respectively. Similar results were observed for RCC1 staining when TIF timecourse specimens were evaluated as fresh FFPE blocks and after 1 year of storage.
Studies
-
Study Purpose
The purpose of this study was to assess the potential effects introduced by delay to fixation (DTF) on immunohistochemical staining of cancer biomarkers in prospectively collected FFPE tumor specimens under the National Cancer Institute's Pre-analytical Variable (BPV) program. Tumors from 111 kidney, 29 colon, 28 ovarian, and 8 lung resections were collected at four medical institutions using a common set of standard operating procedures (SOPs). Each surgical resection was dissected to yield tumor segments of approximately equivalent size (0.33 cm x 0.5 cm) which were stored for 1 h (control), 2 h, 3 h, or 12 h in a humidified chamber at room temperature before fixation. Tumor segments were then fixed in 10% neutral buffered formalin (NBF) at room temperature for 10-12 h before paraffin-embedding. Depending on the tissue, between 5 and 13 biomarkers were examined in FFPE sections (5 µm) by immunohistochemistry. Positive immunostaining was quantified using Aperio software after calibration using slides manually scored by a pathologist. H-score was calculated by multiplying the percentage of positively stained cells/nuclei by the intensity of the stain (0-3). Statistical analysis included the mean and percent difference in H-scores for experimental (2, 3, 12 h DTF) specimens relative to corresponding 1 h DTF controls. H-score data collected from BPV tumor specimens was used to calculate the estimated probability of a change in intensity score (0-3) for each DTF timepoint relative to the 1 h DTF control. Statistical significance was set to p<0.05. A subset of specimens stored as FFPE blocks were re-evaluated one year post-collection.
Summary of Findings:
DTF induced effects on immunohistochemical staining were influenced by the tissue and antigen evaluated. The earliest effects of DTF were observed in kidney tumors with significant (p<0.05) reductions in H-scores compared to case-matched 1 h DTF controls after a 2 h DTF for PAX2 (21%) and EGFR (22%). Reductions in H-scores relative to 1 h controls also occurred after a 3 h DTF for PAX8 (51%), PAX2 (48%), p53 (56%), EGFR (24%), and pAKT (57%); and after a 12 h DTF for PAX8 (93%), PAX2 (68%), p53(94%), c -Met (21%), and pAKT (120%). Kidney tumors fixed for 12 h also display an increase in H-scores for RCC1 and CD10 by 57% and 11%, respectively, compared to 1 h DTF controls. Similar, albeit nonsignificant, patterns of effect on H-score were also observed for affected antigens in other tissue types examined with smaller sample sizes. For example, similar onset and magnitudes of change were observed for PAX8 in ovary, c-Met in lung, and EGFR in colon and lung.
The estimated probability of a change in intensity score (0-3) was calculated for each DTF and tissue type based on data collected with BPV specimens. Antigens with significantly altered H-scores in BPV kidney tumor specimens during the DTF timecourse also had high calculated probabilities of a change in immunostaining intensity score for a similar DTF (25-79%). Based on data collected in BPV ovarian tumor specimens, a DTF ≥3 h in ovarian specimens carries an estimated probability of a reduction in immunostaining score of 40% for PAX8, 30-40% for Wilms tumor 1 (WT1), and 30-40% for estrogen receptor (ER). Data collected in BPV colon tumor specimens indicate that a DTF ≥2 h carries an estimated probability of an increase in immunostaining intensity score of 8-17%. The nine antigens that were examined in DTF timecourse specimens as fresh FFPE blocks and after 1 year of storage displayed similar results. When directly compared, differences between fresh and stored FFPE blocks were limited to EGFR H-scores in 2 h DTF colon tumor specimens (ΔH-score = 12.06 lower in stored specimens).
Biospecimens
Preservative Types
- Formalin
Diagnoses:
- Neoplastic - Carcinoma
Platform:
Analyte Technology Platform Protein Immunohistochemistry Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Immunohistochemistry Specific Targeted peptide/protein PAX2
PAX8
EGFR
p53
pAKT
c-Met
RCC1
CD10
WT1
ER
Biospecimen Acquisition Cold ischemia time 1 h
2 h
3 h
12 h
-
Study Purpose
The purpose of this study was to assess the potential effects associated with time in fixative (TIF) on immunohistochemical staining of cancer biomarkers in prospectively collected FFPE kidney tumor specimens under the National Cancer Institute's Pre-analytical Variable (BPV) program. Tumors from 60 kidney resections were collected at four medical institutions using a common set of standard operating procedures (SOPs). Each surgical resection was dissected to yield tumor segments of approximately equivalent size (0.33 cm x 0.5 cm) that included a quality control for histopathology, a snap-frozen control, and four experimental tumor segments. After a DTF of 1 h, experimental segments were placed in 10% NBF at room temperature for 6, 12 (control), 23, or 72 h. A total of 13 biomarkers were examined in FFPE sections (5 µm) by immunohistochemistry. Positive immunostaining was quantified using Aperio software after calibration using slides manually scored by a pathologist. H-score was calculated by multiplying the percentage of positively stained cells/nuclei by the intensity of the stain (0-3). Statistical analysis included the mean and percent difference in H-scores of experimental (6, 23, 72 h TIF) specimens relative to corresponding 12 h TIF controls. H-score data collected from BPV tumor specimens was used to calculate the estimated probability of a change in intensity score for each TIF timepoint relative to the 12 h TIF control. Statistical significance was set to p<0.05. A subset of specimens stored as FFPE blocks were re-evaluated one year post-collection for RCC1 immunostaining.
Summary of Findings:
Of the 13 antigens evaluated, TIF-associated effects were limited to reductions in c-Met and CD44 after a TIF of 72 h. Relative to case-matched control kidney tumor specimens that were fixed for 12 h, specimens with a TIF of 72 h had significantly lower H-scores for c-Met and CD44 (23% and 17% , respectively; p<0.05 for both). Fixation for 72 h rather than 12 h also translated to an estimated probability of a reduction in intensity score of 51% and 25% for c-Met and CD44, respectively. Similar results were observed for RCC1 staining when TIF timecourse specimens were evaluated as fresh FFPE blocks and after 1 year of storage.
Biospecimens
Preservative Types
- Formalin
Diagnoses:
- Neoplastic - Carcinoma
Platform:
Analyte Technology Platform Protein Immunohistochemistry Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Immunohistochemistry Specific Targeted peptide/protein c-Met
CD44
Biospecimen Preservation Time in fixative 6 h
12 h
23 h
72 h