Table 2 Summary of reported advantages and limitations when utilizing DIA systems to assess Ki-67 for PanNENs.

From: Ki-67 assessment of pancreatic neuroendocrine neoplasms: Systematic review and meta-analysis of manual vs. digital pathology scoring

AUTHOR, YEAR

ADVANTAGES

LIMITATIONS

Bagci, 2012

NR

Highest impact on turnaround time, depending on technician availability; low practicality and moderate accuracy

Remes, 2012

Quick, precise and reliable; not influenced by changes in cell size or growth patterns

NR

Goodell, 2012

Efficient method

Can be influenced by counting hotspot vs. randomly selected fields; low reproducibility if standardized thresholds are lacking

Tang, 2012

Ki67 quantification by MC and DIA demonstrate comparable accuracy

Inability to evaluate each tumor cell

Cimic, 2014

Reproducible

NR

van Velthuysen, 2014

Reproducible

NR

Reid, 2015

Pathologist independent

Dependent upon laboratory technician availability and instrument accessibility; high cost

Kroneman, 2015

Almost perfect correlation between MC and DIA

Difficulty with cell counting due to inability to separate individual cells because of indistinct cell borders

Mejias, 2015

NR

Inability to distinguish infiltrating lymphocytes and other non-neoplastic cells

Neely, 2016

Accurate for cytology

Risk of counting non-tumor contaminants (lymphocytes, pigmented macrophages)

Burdette, 2016

Accuracy

NR

Jin, 2016

NR

Non-tumor cell contamination and insufficient sampling

Dere, 2019

Reduction of time for Ki67 evaluation

Expensive

Saadeh, 2020

Accurate, efficient, reliable and reproducible

Inability to evaluate each tumor cell

Satturwar, 2020

Excellent reliability

NR

Lea, 2021

Improved reliability and reproducibility of grading

NR

Boukhar, 2021

Non-inferiority and substantial time savings

Expert morphologic assessment required for quantitative evaluation

  1. Abbreviations: PanNENs Pancreatic neuroendocrine neoplasms; NR Not reported; MC Manual count; DIA Digital image analysis.