Original article
Detection of mitotic figures in thin melanomas—Immunohistochemistry does not replace the careful search for mitotic figures in hematoxylin-eosin stain

https://doi.org/10.1016/j.jaad.2015.07.007Get rights and content

Background

The mitotic rate is an important prognostic criterion in patients with thin melanoma ≤ 1 mm.

Objective

The aim of this study was to investigate the reproducibility of the mitotic rate in thin melanoma in hematoxylin-eosin (H&E) stain and compare it with the detection of mitotic figures by immunohistochemistry.

Methods

The number of mitoses stated in the routine diagnostic report in 190 pT1 melanomas was compared with the number gained from re-evaluation of H&E sections and the number detected after staining with the mitotic marker, phosphohistone H3 (PHH3). Two different approaches were used for choosing the “hot spot” for evaluation (dermal vs epidermal/dermal).

Results

Comparing routine H&E-stained slides with re-evaluation slides, the number of mitotic figures was slightly variable. However, findings did not result in a change of the tumor stage. In 34% of the tumors with dermal mitotic figures on H&E, mitoses could not be found in the corresponding PHH3 slide anymore. In 4% of the cases, stage relevant mitoses could only be found by PHH3 immunohistochemistry.

Limitation

This is a single center study.

Conclusion

Immunohistochemical staining for mitotic figures does not replace a careful evaluation of H&E-stained slides. Immunohistochemical detection of mitosis is only an additional tool; the time-saving effect is therefore negligible.

Introduction

Tumor thickness is the most powerful independent prognostic criterion in primary melanoma. In thin melanoma, the mitotic rate was considered one major prognostic indicator by the American Joint Committee on Cancer (AJCC) guidelines.1, 2, 3, 4 The presence of ≥1 mitosis in the dermal tumor part results in a shift to pT1b.4, 5 Several studies have found that patients with melanoma less than 1 mm with additional risk factors, such as mitoses, might benefit from sentinel lymph node biopsy.6, 7, 8, 9 The evaluation of mitotic figures is still a matter of debate. In reviewing pathology sections, it may be difficult to unequivocally identify a mitotic figure in a melanocyte versus another cell in the dermis. Moreover, high interobserver variability has been reported.10, 11, 12 The antibody phosphohistone H3 (PHH3, ser10) has been identified as a reliable and sensitive marker to detect mitotic figures in every stage.13, 14, 15 To differentiate mitoses in melanocytes from mitoses in other cells with certainty, double staining for Melan A and PHH3 are helpful.11 The aim of our study was to evaluate the reproducibility of the mitotic count in hematoxylin-eosin (H&E)-stained slides and to determine the benefit of immunohistochemical staining with PHH3 in the subgroup of pT1 melanomas. Moreover, we looked into whether the extension of the “hot spot” to the epidermal tumor part has an influence on the number of stage-relevant mitoses.

Section snippets

Patients and methods

All anonymized clinical and histologic data were stored in a database. All examinations were performed in a blinded fashion. Our study was approved by the local ethics commission.

Patient and tumor characteristics: relation to the mitotic rate

We found a significant correlation between the number of dermal mitoses and Breslow thickness (Mann-Whitney U test; P < .001) and between the mitotic count and Clark level (Jonckheere-Terpstra test; P < .001). We found an equal distribution of mitoses between gender (χ2; P = .09) and age (Mann-Whitney U test; P = .620). The details of the patients and tumor characteristics are given in Table I.

Mitotic figures in H&E—reproducibility in re-evaluation

In routine histology (H&E) of the first cohort (150 pT1a and pT1b melanomas), dermal mitoses were

Discussion

Since the implementation of the mitotic rate in the AJCC classification for thin melanoma as a criterion with a high prognostic impact,2, 4 the correct evaluation and reproducibility of the detection of mitoses has remained a matter of debate.11, 12, 17, 18 In contrast to others,11, 12 we found a high interobserver reproducibility of the mitotic rate with H&E (routine diagnostics vs re-evaluation) with respect to the tumor stage. When comparing the mitotic rate gained by re-evaluation with that

References (26)

  • S.B. Kesmodel et al.

    Mitotic rate as a predictor of sentinel lymph node positivity in patients with thin melanomas

    Ann Surg Oncol

    (2005)
  • P. Donizy et al.

    Mitotic rate is a more reliable unfavorable prognosticator than ulceration for early cutaneous melanoma: a 5-year survival analysis

    Oncol Rep

    (2014)
  • R.A. Scolyer et al.

    Interobserver reproducibility of histopathologic prognostic variables in primary cutaneous melanomas

    Am J Surg Pathol

    (2003)
  • Cited by (18)

    • Iron oxide nanoparticle core-shell magnetic microspheres: Applications toward targeted drug delivery

      2020, Nanomedicine: Nanotechnology, Biology, and Medicine
      Citation Excerpt :

      The excess unabsorbed drug molecules present on the surface of the microspheres were removed while washing with deionized water. The Fe3O4@LEC-CUR-PLGA-MMS was collected in powder form for further analyses.47 An X-ray crystallographic study of Fe3O4NP, Fe3O4@LEC NP and Fe3O4@LEC-CUR-PLGA-MMS was performed using an X-ray diffractometer, D/Max-IIIC, Japan (Cu-kα radiation, λ = 1.5406 Å).

    • Molecular Pathology and Testing in Melanocytic Tumors

      2016, Diagnostic Molecular Pathology: A Guide to Applied Molecular Testing
    View all citing articles on Scopus

    Funding sources: None.

    Parts of this work were presented as a poster at the meeting of the Dermatohistopathologic Working Group German Society of Dermatology, 2014.

    Conflicts of interest: None declared.

    View full text