Skip to main content

Pre-analytical processing protocol of breast biopsies affects multigene panel results.

Lima VR, da Costa Aguiar Alves B, Fonseca FLA, Zveibil DK, Del Giglio A

Publication: Int J Exp Pathol (2022) • Vol 103 • Page 112-120PMID: 35569033View on PubMed

Purpose

This paper explored the potential influence of a change in workflow on an observed difference between mRNA signatures and progression-free survival rates that were in two specimen cohorts. Specifically, the potential influence of a change in unbuffered formalin and paraffin manufacturers and the duration of formalin-fixed paraffin-embedded (FFPE) block storage were also investigated.

Conclusion

The authors state that the levels of the 21 genes targeted were not correlated with time to relapse, but when 8 of the 16 classifier mRNAs were normalized only to GAPDH and RPLP0 there was a trend toward correlation. When those 8 classifier mRNAs were scored, specimens that  scored below the 75’th percentile had higher progression-free survival  than those with scores above the cut-off in one set of specimens but not the other. Despite similar clinical pathological characteristics, the authors noted average gene scores were higher in the second set of specimens than the first. The authors noted that there was a change in workflow during the period of specimen collection that entailed a change in the manufacturer of the unbuffered formalin/paraffin used and changed in humidity.  Authors reported that the proportion of specimens collected before versus after the change in workflow differed between the two specimen sets.   When specimens were classified based on the change in workflow, those collected after the change had higher gene scores. Further analysis showed that a the percentage of specimens with a 260/280 nm ratio between 1.7-2.0 was higher in specimen set 1 compared to set 2, but there was no effect of block storage time.

Studies (1)

Purpose

This study explored the potential influence of a change in workflow on an observed difference between mRNA signatures and progression-free survival rates that were in two specimen cohorts. Specifically, the potential influence of a change in unbuffered formalin and paraffin manufacturers and the duration of FFPE block storage were also investigated. Tumor specimens from 208 breast cancer patients were fixed in 10% unbuffered formalin overnight, dehydrated in alcohol (70%, 90% and 4 changes of 100%), cleared with three changes of xylene, and paraffin embedded at 60°C. Specimens were collected from 2005-2010 and a change in paraffin and unbuffered formalin manufacturers occurred in 2009. The authors note that a change in humidity during block storage also occurred in 2009, but further details were not provided. Specimens were divided into two sets: set 1 (117 specimens) and set 2 (66 specimens); each specimen set included specimens collected before and after the change in workflow. RNA was extracted from two 10 µm thick sections using the RNeasy FFPE kit and quantified using a NanoDrop Spectrophotometer. RNA was reverse transcribed using QuantiNova Reverse Transcription Kit and levels of 21 mRNAs were quantified by real-time PCR.

Summary of Findings

When the first set of specimens (117 specimens) were analyzed, levels of the 21 genes targeted did not correlate with time to relapse and several of the 5 reference genes amplified poorly.  Of the 117 specimens collected, amplification of at least 8 of the 16 classifier mRNAs were successful in 71 specimens; when these 8 genes were normalized to GAPDH and RPLP0, normalized expression displayed a trend to correlation with progression free survival (P=0.1270). When the cutoff for the mRNA score of the 8 genes was set at or above the 75’th percentile, there was a significantly longer progression free survival among patients with scores below the cut-off than above (P=0.0054). In the second set of specimens, progression-free survival and risk score based on the expression for the 8 genes were not correlated, but gene scores were higher in this set than the first specimen set (P=0.02141). When specimens were classified based on whether they were collected before or after the workflow (change in formalin/paraffin manufacturer and the humidity of block storage), those collected after the change had higher gene scores (P=0.02314). Further analysis showed that a higher percentage of specimens collected before the workflow change had a 260/280 nm ratio of 1.7-2.0 than after the change (85% versus 63%, P=0.0040), but there was no effect of block storage.

  • Biospecimens

    TissueBreast
  • Preservation Types

    Formalin
  • Diagnoses

    NeoplasticCarcinoma
  • Platforms

    RNAReal-time qRT-PCR
    RNASpectrophotometry
  • Pre-analytical Factors

    ClassificationPre-Analytical FactorValue(s)
    Biospecimen PreservationEmbedding medium
    Multiple manufacturers evaluated
    Paraffin
    Biospecimen PreservationType of fixation/preservation
    Formalin (unbuffered)
    Multiple manufacturers evaluated
    StorageStorage conditions
    Unspecified change in humidity
    StorageStorage duration
    11-16 years
    Real-time qRT-PCR SpecificTargeted nucleic acid
    21 gene panel