An Emerging Ancillary Technique in Risk Stratification
The diagnosis of dysplasia in the setting of Barrett’s esophagus
is one of the most challenging areas of GI pathology and is associated
with high levels of inter-observer variability amongst general pathologists.
In one recent study, 85% of Barrett’s low grade dysplasia (LGD)
cases diagnosed in a community setting were found to be over-diagnoses
(i.e., not true dysplasia but unrecognized variants of non-dysplastic
Barrett’s).1 Specialized GI pathology training, a high level of
GI pathology experience, and confirmation by one or more additional GI
pathologists significantly improve the accuracy of diagnoses. The GI pathology
team at Inform Diagnostics takes great pride in employing various approaches
to provide the highest quality of diagnosis in Barrett’s biopsies,
including diagnostic consensus criteria, numerous didactic and microscopic
review conferences, and daily multi-headed scope conferences, where an
unprecedented number of GI pathology experts convene to review the most
complex and challenging cases from our nearly 500,000 patients seen each year.
This expertise is exemplified by the appropriately low frequency of “LGD”
among Barrett’s biopsies interpreted at Inform Diagnostics (2.0%),
which is essentially equivalent to the frequencies found following expert
review in recent clinical studies.1 Further, and perhaps more importantly,
we very infrequently render the diagnosis of “indefinite for dysplasia”
among Barrett’s patients (< 2%).
Even with an accurate diagnosis of LGD, however, the time interval for
progression to high grade dysplasia (HGD) and/or invasive esophageal adenocarcinoma
(EAC) cannot be accurately predicted. In addition, a small subset of biopsies
cannot be definitively classified. These shortcomings have driven a large
body of research towards the development of predictive biomarkers. Amongst
these candidates, analysis of p53 has emerged as the most practical for
application to routine biopsy specimens, and is now being regularly utilized
by Inform Diagnostics pathologists.
p53 is one of the most commonly altered genes in human cancer, with mutations
in 70–90% of EAC, 50–70% of HGD, 30–50% of LGD, 10–30%
of lesions classified as indefinite for dysplasia, and in a small percentage
of biopsies considered to be negative for dysplasia. Importantly, the
presence of p53 alterations in biopsies classified as negative, indefinite,
or positive for LGD predicts a significantly increased risk of progression
to HGD and EAC (Table 1). Similarly, mutations in HGD predict a significantly
increased risk of progression to EAC.
Table 1. Risk of progression to HGD/EAC stratified by p53 status.
*Complete references available upon request: firstname.lastname@example.org
The major limitations to utilization of p53 as a genetic biomarker are
technical: conventional genetic analysis requires sophisticated tissue
microdissection, DNA extraction, sequencing, and genomic copy number studies.
As a result, attempts have been made to utilize p53 immunohistochemistry
(IHC) as a surrogate marker of p53 mutation status. Interpretation of
p53 IHC requires implementation of strict scoring criteria, however, and
inexperienced pathologists are at significant risk of misinterpreting
p53 staining. At Inform Diagnostics, our pathologists have developed
and implemented a set of criteria that increase the sensitivity of detection
of abnormal p53 expression by 20%, making it a much more useful marker
for routine clinical practice. In this system, p53 IHC results are categorized
into one of three separate patterns. In normal tissue and non-neoplastic
Barrett’s mucosa, p53 is expressed in the nucleus of a subset of
proliferating cells (Figure 1). In contrast, point mutations in p53 result
in nuclear accumulation of mutant p53 protein (Figure 2). Finally, inactivating
(null) mutations are associated with absent nuclear p53 protein expression
(Figure 3) and are often misinterpreted as normal (wild type).
Figure 1. Wildtype p53 IHC pattern. Left panel, normal squamous mucosa has mostly
1–2+ and scattered 3+ nuclear positivity (brown) confined to the
basal layer. Right panel, non-dysplastic Barrett’s mucosa has mostly
1+ and scattered 2–3+ nuclear positivity (brown) in crypt epithelium.
Figure 2. p53 IHC point mutation pattern. Left panel, Barrett’s LGD with 100%
3+ nuclear p53 positivity (brown) in crypt epithelial cells. Right panel,
Barrett’s LGD (right of field) with 100% 2–3+ nuclear p53
positivity (brown). The patterns in both cases are indicative of underlying
p53 point mutations.
Inform Diagnostics pathologists are beginning to utilize p53 IHC in three
different settings. In biopsies with a diagnosis of LGD or HGD, p53 mutation
patterns may be used to indicate that a lesion has an increased risk of
progression to HGD and/or EAC, and is therefore in need of more vigilant
follow-up/management. In biopsies that are indefinite for dysplasia, the
presence of a p53 mutation pattern may be used to identify those patients
who, despite uncertain histology, are more likely to have a truly neoplastic
process. And in biopsies that are thought to be negative for dysplasia,
p53 IHC may assist in identification of very small microscopic foci of
dysplasia that may have been otherwise missed by routine H&E examination
(Figure 4), or perhaps identify that a patient might benefit from either
closer observation and/or increased biopsy sampling at the next surveillance
encounter. In each of these settings, p53 IHC may help generate a more
accurate and clinically useful report. Analogous to other ancillary stains
that pathologists utilize in the evaluation of GI biopsies, p53 IHC does
not require a special order by the treating physician and is routinely
reimbursed by Medicare and third-party insurance payers.
Figure 3. p53 IHC null mutation pattern. A single non-dysplastic Barrett’s
crypt (center of field) has mostly 1+ and rare 2+ nuclear p53 positivity
(brown). Surrounding Barrett’s LGD has no nuclear p53 positivity,
indicative of an underlying null mutation involving p53.
Figure 4. Focal p53 IHC point mutation pattern. A single crypt and two small surface
epithelial foci reveal 2–3+ nuclear p53 positivity (brown), consistent
with point mutation pattern. Multiple additional deeper sections were
obtained because of the p53 result, and a definite focus of LGD was diagnosed.
p53 IHC may increase the sensitivity of detection of microscopic foci of LGD.
The GI pathology team at Inform Diagnostics strives to provide the highest
quality diagnoses and the most useful ancillary test results. The addition
of p53 IHC provides an important advance in the quality of Barrett’s
diagnoses by moving beyond traditional H&E diagnostics. We are optimistic
that this will be the first of many Barrett’s innovations that our
team will utilize in improving the care of our patients.
– Mark Redston, MD, Director of GI Molecular Pathology
Curvers WL et al. Low-grade dysplasia in Barrett’s esophagus: overdiagnosed
and underestimated.American Journal of Gastroenterology. 2010;105:1523-30.