| Article Access Statistics | | | Viewed | 5003 | | | Printed | 211 | | | Emailed | 4 | | | PDF Downloaded | 79 | | | Comments | [Add] | | | Cited by others | 5 | | |
|
Click on image for details.
|
|
 |
| ORIGINAL ARTICLE |
|
|
|
| Year : 2012 | Volume
: 60
| Issue : 1 | Page : 61-65 |
Concordance between local, institutional, and central pathology review in glioblastoma: Implications for research and practice: A pilot study
Tejpal Gupta1, Vimoj Nair1, Sridhar Epari2, Torsten Pietsch3, Rakesh Jalali1
1 Department of Radiation Oncology, Advanced Centre for Treatment Research and Education in Cancer (ACTREC) and Tata Memorial Hospital, Tata Memorial Centre, Mumbai, India
2 Department of Pathology, Advanced Centre for Treatment Research and Education in Cancer (ACTREC) and Tata Memorial Hospital, Tata Memorial Centre, Mumbai, India
3 Department of Neuro-Pathology, University of Bonn, Germany
| Date of Submission | 04-Aug-2011 |
| Date of Decision | 18-Sep-2011 |
| Date of Acceptance | 08-Dec-2011 |
| Date of Web Publication | 7-Mar-2012 |
Correspondence Address:
Tejpal Gupta
Department of Radiation Oncology, Advanced Centre for Treatment Research and Education in Cancer (ACTREC), Tata Memorial Centre, Kharghar, Navi Mumbai-410-210, Maharashtra
India
 Source of Support: None, Conflict of Interest: None  | Check |
DOI: 10.4103/0028-3886.93594
Background: There is significant inter-observer variation amongst the neuro-pathologists in the typing, subtyping, and grading of glial neoplasms for diagnosis. Centralized pathology review has been proposed to minimize this inter-observer variation and is now almost mandatory for accrual into multicentric trials. We sought to assess the concordance between neuro-pathologists on histopathological diagnosis of glioblastoma. Materials and Methods: Comparison of local, institutional, and central neuro-oncopathology reporting in a cohort of 34 patients with newly diagnosed supratentorial glioblastoma accrued consecutively at a tertiary-care institution on a prospective trial testing the addition of a new agent to standard chemo-radiation regimen. Results: Concordance was sub-optimal between local histological diagnosis and central review, fair between local diagnosis and institutional review, and good between institutional and central review, with respect to histological typing/subtyping. Twelve (39%) of 31 patients with local histological diagnosis had identical tumor type, subtype and grade on central review. Overall agreement was modestly better (52%) between local diagnosis and institutional review. In contrast, 28 (83%) of 34 patients had completely concordant histopathologic diagnosis between institutional and central review. The inter-observer reliability test showed poor agreement between local and central review (kappa statistic=0.12, 95% confidence interval (CI): -0.03-0.32, P=0.043), but moderate agreement between institutional and central review (kappa statistic=0.51, 95%CI: 0.17-0.84, P=0.00003). Agreement between local diagnosis and institutional review was fair. Conclusions: There exists significant inter-observer variation regarding histopathological diagnosis of glioblastoma with significant implications for clinical research and practice. There is a need for more objective, quantitative, robust, and reproducible criteria for better subtyping for accurate diagnosis.
Keywords: Agreement, central review, concordance, glioblastoma, grade
How to cite this article:
Gupta T, Nair V, Epari S, Pietsch T, Jalali R. Concordance between local, institutional, and central pathology review in glioblastoma: Implications for research and practice: A pilot study. Neurol India 2012;60:61-5 |
How to cite this URL:
Gupta T, Nair V, Epari S, Pietsch T, Jalali R. Concordance between local, institutional, and central pathology review in glioblastoma: Implications for research and practice: A pilot study. Neurol India [serial online] 2012 [cited 2023 Nov 30];60:61-5. Available from: https://www.neurologyindia.com/text.asp?2012/60/1/61/93594 |
| » Introduction | |  |
Gliomas are not only the most common [1],[2] but also the most heterogeneous group of primary brain tumors with wide variation in clinical presentation, biological behavior and outcomes. Histological grade as defined by the World Health Organization (WHO) [3] is the most important factor that determines outcomes in tumors of glial origin. Contemporary clinical management relies heavily on accurate histological grading and typing for selection of therapeutic regimen as well as prognostication. There is ample evidence regarding the existence of significant inter-observer variation in the typing, subtyping, and grading of brain tumors [4] including glial neoplasms [5] in the peer-reviewed indexed medical literature. Centralized pathology review by an experienced neuro-oncopathologist utilizing supplementary diagnostic techniques is highly recommended to minimize this inter-observer variation amongst different pathologists and almost mandatory for accrual into multicentric drug trials. Pretreatment central review may not be necessary whenever there are well-established diagnostic criteria that can be applied uniformly and reported consistently by local and/or institutional pathologists. However, the review may be considered appropriate even with well-established criteria in the presence of significantly variability in interpretation and reporting. [6] There is lack of published data highlighting inter-observer variation in glioblastoma amongst pathologists within India and their concordance with central pathology review. In this context, we sought to assess the concordance and extent of agreement on histopathological diagnosis in a cohort of patients with newly diagnosed glioblastomas referred to us from local community hospitals. In a pilot study, we compared the degree of concordance and agreement in the histopathological diagnosis between the neuro-oncopathologists of the local community hospital, tertiary-care institution, and central pathology review carried out by a dedicated and experienced neuro-pathologist, considering central review as the reference standard for final and definitive diagnosis.
| » Materials and Methods | | |
Patients with newly diagnosed supratentorial glioblastomas referred to a tertiary-care centre for postoperative adjuvant therapy were considered for inclusion in the study. Patients who had undergone surgical resection at any outside community hospital were asked to submit at least two hematoxylin-and-eosin stained slides and preferably one representative formalin-fixed paraffin-embedded (FFPE) tissue block for pathology review by our staff neuro-pathologist. Initial histopathology report (type, subtype, and grade of tumor) from the referring hospital was documented and categorized as 'local', while our in-house tertiary-care centre pathology review was categorized as 'institutional'. In the large majority of the cases (>80%), local histological diagnosis was based on light microscopy alone, although in a smaller proportion of cases (<20%), immuno-histochemistry (IHC) was also used by local pathologist(s) depending upon available infrastructure. Institutional pathology review was also largely based on morphological criteria, although IHC was employed in 60% of patients for better pathologic characterization. Based on our institutional pathologist's report of Grade IV glioma (glioblastoma or its variants), eligible patients were accrued to a prospective drug trial, testing the efficacy of the addition of a new pharmacological agent as a supplement to standard chemo-radiation regimen, after written informed consent of the patient or legal guardian. Representative FFPE block(s) were subsequently sent for 'central' pathology review to be evaluated by a single, dedicated, and experienced neuro-oncopathologist who was not blinded to the institutional pathologist's diagnosis. A fresh set of slides was prepared from the representative block at the central laboratory for diagnosis of glioblastoma based on morphological criteria supplemented with appropriate IHC in all the samples. Molecular pathology tools, however, were not used either by the institutional or central neuro-pathologist for diagnostic confirmation. Concordance between local, institutional, and central histopathological diagnosis was analyzed considering 'central pathology review' as the reference standard with appropriate statistical methodology. Inter-observer reliability analysis was performed using the unadjusted kappa statistic (test of agreement) and reported with 95% confidence intervals (CI).
| » Results | | |
In a one-year period (January to December 2010), 34 patients with newly diagnosed glioblastoma (or its variants) were accrued consecutively into the study and their FFPE tissue blocks were sent to a single experienced neuro-pathologist in Europe for central pathology review. Thirty-one patients had undergone surgical resection in the community hospitals and were referred along with a local histopathological diagnosis for further evaluation and management. Three patients had undergone primary surgery at our tertiary-care institute. Institutional and central pathology review reports were available for the entire study cohort [Table 1]. All tumors were reported as glial in origin, excepting for one patient who had a histopathological diagnosis of atypical teratoid/rhabdoid tumor from the local hospital, which was reported as giant-cell glioblastoma both by the institutional pathologist and at the central pathology review. Histological grade was reported as WHO Grade III (anaplastic glioma, either astrocytoma or oligodendroglioma) in five (16%) patients by the local referring pathologist. However, both the institutional as well as central pathology review interpreted them as WHO Grade IV tumors. There was sub-optimal overall agreement between the local and central review, while there was good overall agreement between the institutional and central review, with respect to histological typing/subtyping. Only 12 (39%) of 31 cases had a concordance in the type/subtype and grade of glial tumor between local histopathological diagnosis and the central review. There was modestly better overall agreement of 52% (16 of 31 patients) between the local pathologist's diagnosis and institutional review. In contrast, 28 (83%) of 34 tumor samples revealed concordance in histopathologic diagnosis (type/subtype and grade) between institutional and central review. The inter-observer reliability test (unadjusted kappa test) showed poor agreement between local and central pathology review (kappa statistic=0.12, 95%CI: -0.03-0.32, P=0.043), but moderate agreement between institutional and central review (kappa statistic=0.51, 95% CI: 0.17-0.84, P=0.00003). Agreement between local diagnosis and institutional review was fair (kappa statistic=0.22, 95% CI: 0.06-0.38, P=0.002). | Table 1: Local histological diagnosis, institutional review, and central pathology review
Click here to view |
| » Discussion | | |
Given the existence of cellular heterogeneity both within and between primary brain tumors, histopathological diagnosis can be highly variable and extremely challenging. Of all histological types of primary brain tumors, gliomas are the most difficult ones to categorize into discrete groups as they lie on a continuous spectrum of histology and malignancy. The currently accepted standard for brain tumor classification, the WHO system [3] uses morphological criteria such as cellularity, mitoses, anaplasia, nuclear pleomorphism, necrosis, microvascular and endothelial proliferation for histological grading. Evaluation of these criteria can be subjective at times leading to variable interpretation. Challenges in making accurate diagnoses in gliomas have received widespread attention from the neuropathology [7] and brain tumor epidemiology community. [8]
Apart from differences in training, expertise, and experience, several other factors can have an impact on the agreement and concordance (or the lack of it) between pathologists. One important reason is sample heterogeneity i.e. exactly the same material may not be available to two pathologists resulting in differential interpretation. Another commonly cited cause is technical quality, [5] especially when non-representative or poor-quality material is submitted for review process. Some of the other known factors that can influence the degree of agreement [8] include histological tumor type (glial or non-glial); type of review (uniform or consensus); setting (community or academic) and number of histological subtypes within one tumor (single or more than one). For example, the disagreement is expected to be larger in tumors that have combined characteristics of more than a single histology such as glioblastoma with oligodendroglial differentiation, gliosarcoma, or mixed anaplastic oligoastrocytoma as compared to single histological subtypes.
In a large systematic study [4] on a cohort of 500 brain tumor samples, some disagreement between pathologists was noted in 42.8% of instances (serious disagreement 8.8%). Another large study on gliomas reported discordant diagnoses in 23% of 457 patients (serious discordance 16%); higher discordance rates being noted in cases referred by local community hospitals compared to academic institutions. [7] The first attempt at exploring the impact of central pathology review on malignant high-grade gliomas was pioneered by the Radiation Therapy Oncology Group (RTOG) in a prospective clinical trial. [9] The study reported excellent concordance (96%) for locally diagnosed glioblastoma, but low concordance (66%) for astrocytoma with anaplastic foci. Thus, some degree of disagreement is expected in nearly one-third of patients (range 20-50%) of which 10% (range 8-20%) may be considered serious or clinically significant with implications for modified management and prognosis. [5]
The precision of diagnosis can be increased by integrating molecular biology with conventional pathology techniques. Molecular genetic analysis has provided deeper insight with an improved fundamental understanding of the biology of several primary brain tumors including glioblastoma. The aim of molecular classification of neoplasms is to identify sub-groups with distinct biological and clinical behavior. Evidence from clinico-patho-biological studies [10] has increasingly shown that glioblastoma is a heterogeneous neoplasm with subtypes having distinct phenotypes, diverse biological behavior, and varying clinical outcomes. The caveat is that currently there are no standardized and validated molecular panels for widespread adoption.
To the best of our knowledge, this is the first report from India that highlights the presence of significant inter-observer variation amongst neuro-pathologists in the diagnosis of glioblastoma. Although, agreement between local diagnosis and central review was poor, it was reassuring to note the presence of good overall concordance and moderate agreement between a tertiary-care institution from India and central pathology review from Europe. It may also be pertinent to note that the discordance between institutional and central review was largely with respect to histological subtyping, viz. disagreement on the presence of oligodendroglial component, mesenchymal component, or giant-cell predominance in the brain tumor material submitted for review.
However, several limitations remain that need to be accounted for proper interpretation. Firstly, our sample size is relatively small for statistical robustness. Secondly, these were not consecutively referred patients, hence a selection bias cannot be entirely ruled out. Only patients with our institutional diagnosis of glioblastoma (or its variants) were included, hence we cannot comment on lack of concordance between anaplastic glioma (WHO Grade III) and glioblastoma, a fact well-documented in medical literature. Thirdly, several neuro-pathologists from different community hospitals with variable infrastructure and expertise referred patients to the tertiary-care centre with a local histological diagnosis, whereas institutional and central pathology review was limited to a single dedicated neuro-pathologist each, with access to modern pathology tools. It is essential to realize these issues and evolve a relatively uniform protocol for diagnosis based on accepted criteria by general consensus.
The existence of significant inter-observer variation amongst the neuropathology community has significant implications for the design, conduct, and interpretation of prospective clinical therapeutic trials in contemporary neuro-oncologic practice. Most multicentric clinical trials now mandate central pathology review either prior to or after recruiting for the study to minimize inter-observer variation and ensure a more homogenous patient population for statistical evaluation of the outcome and prognosis. Diagnostic reliability and consistency in neuropathology has considerably improved over time and hopefully will continue to evolve further to help clinicians in choosing the treatments and optimizing the outcomes.
| » Conclusions | | |
Significant inter-observer variation exists amongst the neuropathology community regarding histopathological diagnosis of glioblastoma with significant implications for clinical research and practice. Molecular genetic approaches may help improve classification, but are unlikely to supplant conventional morphologic criteria in the near future. There is a need for more objective, quantitative, robust, and reproducible histopathological criteria for better subtyping of gliomas for accurate diagnosis to aid in patient selection for optimizing therapeutic outcomes.
| » Acknowledgments | | |
All neurosurgeons from local community hospitals for referring the patients postoperatively for further management and all pathologists for providing paraffin-blocks of those patients for institutional and central pathology review. A special acknowledgment to Ms. Sadhana Kannan for helping with the statistical analysis.
| » References | | |
| 1. | Central Brain Tumor Registry of the United States (CBTRUS) Statistical Report 2005-2006. Primary brain tumors in the United States:1998-2002: Hinsdale, IL. 
|
| 2. | Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer 2010;127:2893-917.
[PUBMED] [FULLTEXT] |
| 3. | Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, Burger PC, Jouvet A, et al. The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol 2007;114:97-109.
[PUBMED] [FULLTEXT] |
| 4. | Bruner JM, Inouye L, Fuller GN, Langford LA. Diagnostic discrepancies and their clinical impact in a neuropathology referral practice. Cancer 1997;79:796-803.
[PUBMED] |
| 5. | van den Bent MJ. Interobserver variation of the histopathological diagnosis in clinical trials on glioma: A clinician's perspective. Acta Neuropathol 2010;120:297-304.
[PUBMED] [FULLTEXT] |
| 6. | Teot LA, Sposto R, Khayat A, Qualman S, Reaman G, Parham D. The problems and promise of central pathology review: Development of a standardized procedure for the Children's Oncology Group. Pediatr Dev Pathol 2007;10:199-207.
|
| 7. | Aldape K, Simmons ML, Davis RL, Miike R, Wiencke J, Barger G, et al. Discrepancies in diagnoses of neuroepithelial neoplasms: The San Francisco Bay Area Adult Glioma Study. Cancer 2000;88:2342-9.
[PUBMED] |
| 8. | Davis FG, Malmer BS, Aldape K, Barnholtz-Sloan JS, Bondy ML, Brannstrom T, et al. Issues of diagnostic review in brain tumor studies: From the Brain Tumor Epidemiology Consortium. Cancer Epidemiol Biomarkers Prev 2008;17:484-9.
|
| 9. | Scott CB, Nelson JS, Farnan NC, Curran WJ Jr., Murray KJ, Fischbach AJ, et al. Central pathology review in clinical trials for patients with malignant glioma. A Report of Radiation Therapy Oncology Group 83-02. Cancer 1995;76:307-13.
|
| 10. | Verhaak RG, Hoadley KA, Purdom E, Wang V, Qi Y, Wilkerson MD, et al. Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1. Cancer Cell 2010;17:98-110.
[PUBMED] [FULLTEXT] |
[Table 1]
| This article has been cited by | | 1 |
Inter-pathologist agreement on diagnosis, classification and grading of canine glioma |
|
| Gregory A. Krane, Keith R. Shockley, David E. Malarkey, Andrew D. Miller, C. Ryan Miller, Debra A. Tokarz, Heather L. Jensen, Kyathanahalli S. Janardhan, Matthew Breen, Christopher L. Mariani | | Veterinary and Comparative Oncology. 2022; | | [Pubmed] | [DOI] | | | 2 |
Histological features associated with diagnostic agreement in atypical ductal hyperplasia of the breast: illustrative cases from the B-Path study |
|
| Kimberly H Allison,Mara H Rendi,Sue Peacock,Tom Morgan,Joann G Elmore,Donald L Weaver | | Histopathology. 2016; 69(6): 1028 | | [Pubmed] | [DOI] | | | 3 |
Brain tumour differentiation: rapid stratified serum diagnostics via attenuated total reflection Fourier-transform infrared spectroscopy |
|
| James R. Hands,Graeme Clemens,Ryan Stables,Katherine Ashton,Andrew Brodbelt,Charles Davis,Timothy P. Dawson,Michael D. Jenkinson,Robert W. Lea,Carol Walker,Matthew J. Baker | | Journal of Neuro-Oncology. 2016; | | [Pubmed] | [DOI] | | | 4 |
The Glioma International Case-Control Study: A Report From the Genetic Epidemiology of Glioma International Consortium |
|
| E. Susan Amirian,Georgina N. Armstrong,Renke Zhou,Ching C. Lau,Elizabeth B. Claus,Jill S. Barnholtz-Sloan,Dora Ilæyasova,Joellen Schildkraut,Francis Ali-Osman,Siegal Sadetzki,Christoffer Johansen,Richard S. Houlston,Robert B. Jenkins,Daniel Lachance,Sara H. Olson,Jonine L. Bernstein,Ryan T. Merrell,Margaret R. Wrensch,Faith G. Davis,Rose Lai,Sanjay Shete,Christopher I. Amos,Michael E. Scheurer,Kenneth Aldape,Irina Alafuzoff,Thomas Brännström,Helle Broholm,Peter Collins,Caterina Giannini,Marc Rosenblum,Tarik Tihan,Beatrice S. Melin,Melissa L. Bondy | | American Journal of Epidemiology. 2015; : kwv235 | | [Pubmed] | [DOI] | | | 5 |
Understanding diagnostic variability in breast pathology: lessons learned from an expert consensus review panel |
|
| Kimberly H Allison,Lisa M Reisch,Patricia A Carney,Donald L Weaver,Stuart J Schnitt,Frances P OæMalley,Berta M Geller,Joann G Elmore | | Histopathology. 2014; : n/a | | [Pubmed] | [DOI] | |
|
|
|
|
|
|
|
|
