Interobserver variability among expert orthopedic pathologists for diagnosis, histologic grade, and determination of the necessity for chemotherapy in osteosarcoma

S. E. Kilpatrick, F. W. Abdul-Karim, J. B. Renner, Tonya King, M. J. Klein, A. E. Rosenberg, G. C. Steiner, P. G. Bullough, A. L. Schiller, H. D. Dorfman

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Eight expert orthopedic pathologists voluntarily, reviewed the clinicoradiologic and histologic data from 12 selected cases coded as osteosarcoma over the past seven years (procured from a total of 40 cases) from the files of the University of North Carolina and Wake Forest University. The study sample was chosen to include a full spectrum of histologic grades and subtypes including conventional, low-grade central, chondroblastoma-like, osteoblastoma-like, small cell, telangiectatic, parosteal, and periosteal variants. Participants were asked to render a diagnosis, provide a histologic grade (if applicable), and offer an opinion as to whether the patient could benefit from chemotherapy. Uniform agreement regarding the diagnosis of osteosarcoma was observed in 7 cases (58 hr %), including the most common, conventional subtypes, and the telangiectatic, parosteal, and periosteal variants. Among the remaining 5 cases, the consensus majority (≥ 4) also was osteosarcoma. However, additional diagnoses ranged from osteoblastoma to high-grade osteosarcoma in two tumors, mesenchymal chondrosarcoma versus small cell osteosarcoma in one, fibrosarcoma and malignant fibrous histiocytoma versus fibroblastic osteosarcoma in another, and clear cell chondrosarcoma versus chondroblastoma versus chondroblastic osteosarcoma in a fifth. Among osteosarcoma diagnoses, histologic grading ranged from low- to high-grade osteosarcoma in 6 tumors (50%). Regarding the necessity for chemotherapy, there was disagreement from at least one consultant in 7 cases (58 hr %). Evaluation of the intraclass correlation coefficient revealed modest agreement with respect to histologic grade and therapy but poor agreement for diagnosis, the latter reflecting the wide range of diagnoses in a minority of cases. Interobserver variability exists among expert orthopedic pathologists for determining the diagnosis, histologic grade, and necessity of chemotherapy in less common forms of osteosarcoma. A more objective histologic grading and scoring system for osteosarcoma should be established, followed by multi-institutional studies assessing the usefulness of the system as it relates to the necessity of chemotherapy and survival.

Original languageEnglish (US)
Pages (from-to)337-358
Number of pages22
JournalPediatric Pathology and Molecular Medicine
Volume19
Issue number5
DOIs
StatePublished - Jan 1 2000

Fingerprint

Observer Variation
Osteosarcoma
Orthopedics
Drug Therapy
Chondroblastoma
Osteoblastoma
Pathologists
Mesenchymal Chondrosarcoma
Malignant Fibrous Histiocytoma
Chondrosarcoma
Fibrosarcoma
Consultants
Neoplasms

All Science Journal Classification (ASJC) codes

  • Pediatrics, Perinatology, and Child Health
  • Pathology and Forensic Medicine

Cite this

Kilpatrick, S. E. ; Abdul-Karim, F. W. ; Renner, J. B. ; King, Tonya ; Klein, M. J. ; Rosenberg, A. E. ; Steiner, G. C. ; Bullough, P. G. ; Schiller, A. L. ; Dorfman, H. D. / Interobserver variability among expert orthopedic pathologists for diagnosis, histologic grade, and determination of the necessity for chemotherapy in osteosarcoma. In: Pediatric Pathology and Molecular Medicine. 2000 ; Vol. 19, No. 5. pp. 337-358.
@article{f8ba58a5c9e346df92370c14d6061737,
title = "Interobserver variability among expert orthopedic pathologists for diagnosis, histologic grade, and determination of the necessity for chemotherapy in osteosarcoma",
abstract = "Eight expert orthopedic pathologists voluntarily, reviewed the clinicoradiologic and histologic data from 12 selected cases coded as osteosarcoma over the past seven years (procured from a total of 40 cases) from the files of the University of North Carolina and Wake Forest University. The study sample was chosen to include a full spectrum of histologic grades and subtypes including conventional, low-grade central, chondroblastoma-like, osteoblastoma-like, small cell, telangiectatic, parosteal, and periosteal variants. Participants were asked to render a diagnosis, provide a histologic grade (if applicable), and offer an opinion as to whether the patient could benefit from chemotherapy. Uniform agreement regarding the diagnosis of osteosarcoma was observed in 7 cases (58 hr {\%}), including the most common, conventional subtypes, and the telangiectatic, parosteal, and periosteal variants. Among the remaining 5 cases, the consensus majority (≥ 4) also was osteosarcoma. However, additional diagnoses ranged from osteoblastoma to high-grade osteosarcoma in two tumors, mesenchymal chondrosarcoma versus small cell osteosarcoma in one, fibrosarcoma and malignant fibrous histiocytoma versus fibroblastic osteosarcoma in another, and clear cell chondrosarcoma versus chondroblastoma versus chondroblastic osteosarcoma in a fifth. Among osteosarcoma diagnoses, histologic grading ranged from low- to high-grade osteosarcoma in 6 tumors (50{\%}). Regarding the necessity for chemotherapy, there was disagreement from at least one consultant in 7 cases (58 hr {\%}). Evaluation of the intraclass correlation coefficient revealed modest agreement with respect to histologic grade and therapy but poor agreement for diagnosis, the latter reflecting the wide range of diagnoses in a minority of cases. Interobserver variability exists among expert orthopedic pathologists for determining the diagnosis, histologic grade, and necessity of chemotherapy in less common forms of osteosarcoma. A more objective histologic grading and scoring system for osteosarcoma should be established, followed by multi-institutional studies assessing the usefulness of the system as it relates to the necessity of chemotherapy and survival.",
author = "Kilpatrick, {S. E.} and Abdul-Karim, {F. W.} and Renner, {J. B.} and Tonya King and Klein, {M. J.} and Rosenberg, {A. E.} and Steiner, {G. C.} and Bullough, {P. G.} and Schiller, {A. L.} and Dorfman, {H. D.}",
year = "2000",
month = "1",
day = "1",
doi = "10.1080/15513810009168644",
language = "English (US)",
volume = "19",
pages = "337--358",
journal = "Fetal and Pediatric Pathology",
issn = "1551-3815",
publisher = "Informa Healthcare",
number = "5",

}

Interobserver variability among expert orthopedic pathologists for diagnosis, histologic grade, and determination of the necessity for chemotherapy in osteosarcoma. / Kilpatrick, S. E.; Abdul-Karim, F. W.; Renner, J. B.; King, Tonya; Klein, M. J.; Rosenberg, A. E.; Steiner, G. C.; Bullough, P. G.; Schiller, A. L.; Dorfman, H. D.

In: Pediatric Pathology and Molecular Medicine, Vol. 19, No. 5, 01.01.2000, p. 337-358.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Interobserver variability among expert orthopedic pathologists for diagnosis, histologic grade, and determination of the necessity for chemotherapy in osteosarcoma

AU - Kilpatrick, S. E.

AU - Abdul-Karim, F. W.

AU - Renner, J. B.

AU - King, Tonya

AU - Klein, M. J.

AU - Rosenberg, A. E.

AU - Steiner, G. C.

AU - Bullough, P. G.

AU - Schiller, A. L.

AU - Dorfman, H. D.

PY - 2000/1/1

Y1 - 2000/1/1

N2 - Eight expert orthopedic pathologists voluntarily, reviewed the clinicoradiologic and histologic data from 12 selected cases coded as osteosarcoma over the past seven years (procured from a total of 40 cases) from the files of the University of North Carolina and Wake Forest University. The study sample was chosen to include a full spectrum of histologic grades and subtypes including conventional, low-grade central, chondroblastoma-like, osteoblastoma-like, small cell, telangiectatic, parosteal, and periosteal variants. Participants were asked to render a diagnosis, provide a histologic grade (if applicable), and offer an opinion as to whether the patient could benefit from chemotherapy. Uniform agreement regarding the diagnosis of osteosarcoma was observed in 7 cases (58 hr %), including the most common, conventional subtypes, and the telangiectatic, parosteal, and periosteal variants. Among the remaining 5 cases, the consensus majority (≥ 4) also was osteosarcoma. However, additional diagnoses ranged from osteoblastoma to high-grade osteosarcoma in two tumors, mesenchymal chondrosarcoma versus small cell osteosarcoma in one, fibrosarcoma and malignant fibrous histiocytoma versus fibroblastic osteosarcoma in another, and clear cell chondrosarcoma versus chondroblastoma versus chondroblastic osteosarcoma in a fifth. Among osteosarcoma diagnoses, histologic grading ranged from low- to high-grade osteosarcoma in 6 tumors (50%). Regarding the necessity for chemotherapy, there was disagreement from at least one consultant in 7 cases (58 hr %). Evaluation of the intraclass correlation coefficient revealed modest agreement with respect to histologic grade and therapy but poor agreement for diagnosis, the latter reflecting the wide range of diagnoses in a minority of cases. Interobserver variability exists among expert orthopedic pathologists for determining the diagnosis, histologic grade, and necessity of chemotherapy in less common forms of osteosarcoma. A more objective histologic grading and scoring system for osteosarcoma should be established, followed by multi-institutional studies assessing the usefulness of the system as it relates to the necessity of chemotherapy and survival.

AB - Eight expert orthopedic pathologists voluntarily, reviewed the clinicoradiologic and histologic data from 12 selected cases coded as osteosarcoma over the past seven years (procured from a total of 40 cases) from the files of the University of North Carolina and Wake Forest University. The study sample was chosen to include a full spectrum of histologic grades and subtypes including conventional, low-grade central, chondroblastoma-like, osteoblastoma-like, small cell, telangiectatic, parosteal, and periosteal variants. Participants were asked to render a diagnosis, provide a histologic grade (if applicable), and offer an opinion as to whether the patient could benefit from chemotherapy. Uniform agreement regarding the diagnosis of osteosarcoma was observed in 7 cases (58 hr %), including the most common, conventional subtypes, and the telangiectatic, parosteal, and periosteal variants. Among the remaining 5 cases, the consensus majority (≥ 4) also was osteosarcoma. However, additional diagnoses ranged from osteoblastoma to high-grade osteosarcoma in two tumors, mesenchymal chondrosarcoma versus small cell osteosarcoma in one, fibrosarcoma and malignant fibrous histiocytoma versus fibroblastic osteosarcoma in another, and clear cell chondrosarcoma versus chondroblastoma versus chondroblastic osteosarcoma in a fifth. Among osteosarcoma diagnoses, histologic grading ranged from low- to high-grade osteosarcoma in 6 tumors (50%). Regarding the necessity for chemotherapy, there was disagreement from at least one consultant in 7 cases (58 hr %). Evaluation of the intraclass correlation coefficient revealed modest agreement with respect to histologic grade and therapy but poor agreement for diagnosis, the latter reflecting the wide range of diagnoses in a minority of cases. Interobserver variability exists among expert orthopedic pathologists for determining the diagnosis, histologic grade, and necessity of chemotherapy in less common forms of osteosarcoma. A more objective histologic grading and scoring system for osteosarcoma should be established, followed by multi-institutional studies assessing the usefulness of the system as it relates to the necessity of chemotherapy and survival.

UR - http://www.scopus.com/inward/record.url?scp=0033677278&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033677278&partnerID=8YFLogxK

U2 - 10.1080/15513810009168644

DO - 10.1080/15513810009168644

M3 - Article

VL - 19

SP - 337

EP - 358

JO - Fetal and Pediatric Pathology

JF - Fetal and Pediatric Pathology

SN - 1551-3815

IS - 5

ER -