Pathology of Chordoma and Chondrosarcoma of the Axial Skeleton

© 2018 Elsevier Inc. All rights reserved. Please note that the copyright for the original figures submitted by the contributors is owned by Contributors.

Chordoma

Epidemiology

The true incidence of chordoma is unknown. In several large studies, it accounted for 1%–4% of all primary malignant bone tumors. In more recent data, collected by the National Cancer Institute’s Surveillance, Epidemiology and End Results (SEER) program, which included 12,931 primary malignant tumors of the bone diagnosed between the years 1973 and 2012, 8.3% or 1080 cases were chordoma, and chordoma followed osteosarcoma, chondrosarcoma, and Ewing sarcoma in frequency. Updated SEER data report an overall incidence rate of chordoma of 0.84 per 1,000,000. SEER data from the years 1973 to 2003 show that chordoma follows chondrosarcoma and accounts for 26% of primary bone malignancies of the spine in patients who present with a nonmetastatic primary bone malignancy. In a report from the Leeds Regional Bone Tumor Registry, chordoma was the most common primary malignant tumor of the axial skeleton. Information from the Swedish Registry documented the annual incidence of chordoma in that country to be 0.5% per million individuals and indicated that chordoma was responsible for 17.5% of all primary malignant bone tumors and 20% of those arising in the spine. Similar epidemiologic characteristics were found in Finland.

Chordoma occurs in all age groups, but the majority arises during adulthood, with the median age of affected individuals being 58 years and 65% of patients being younger than 65 years at the time of diagnosis. Specifically, children, young adults, adults, and the elderly have incidence rates of 0.14, 0.43, 1.08, and 2.62 per million population, respectively. Chordomas originating in the base of the skull present a decade earlier than those that develop in the spine, especially the sacrum. The occurrence of chordoma in children is unusual, representing <5% of all chordomas, but in our experience and from data in the literature, it usually originates in the skull base and cervical spine and rarely in the sacrum. Most studies of chordomas have shown that there is a male predominance of approximately 2:1 for tumors arising in the sacrum and mobile spine, whereas in the skull base, the gender distribution is nearly equal. In the SEER data, the vast majority of tumors arose in whites, whereas only 2.2% affected African Americans.

The pathogenesis of chordoma is not fully elucidated, but numerous lines of evidence suggest that T ( brachyury ), a gene that encodes brachyury, a transcription factor that regulates notochord development, is crucial for the initiation and progression of chordoma. Although it is generally accepted that these tumors arise from persistent rests of notochord within rigid bony structures, and not intervertebral disks, this hypothesis has never been proven. Evidence supporting this hypothesis linking notochordal cell rests to chordoma include similar topographical distribution, morphological overlap at the light and electron microscopic levels, shared immunophenotype, and molecular phenotyping studies. Chordoma has no known association with irradiation or other environmental factors. A very small percentage of cases have a familial pattern of inheritance, and in at least one case, the mode of inheritance was probably autosomal dominant. A subset of both sporadic and hereditary chordomas has somatic chromosomal gain at the T locus. In four families with familial chordoma, high-resolution array comparative genomic hybridization showed unique duplications in 6q27, leading to T gene duplication. A study sequencing the T gene and measuring copy number variants in 24 familial cases, 103 sporadic cases, and 160 unrelated controls highlighted the importance of genetic variations in the T gene for both familial and sporadic chordoma. It demonstrated that germline T duplication is relatively common in families with chordoma (44%) but extremely rare in sporadic cases and suggested a complex susceptibility related to T . Broad genotyping of individuals with sporadic chordoma has highlighted recurrent single nucleotide polymorphisms (SNPs) in the T DNA–binding domain, suggesting that misregulation of genes controlled by T may be important in the pathogenesis of chordoma. Pathways shown to be important in chordoma tumorigenesis include cell cycle regulatory pathways and activated receptor kinase pathways; additionally, DNA methylation of tumor suppressor genes and microRNA’s has been demonstrated to modulate relevant pathways (see Chapter 4 ).

The anatomic distribution of chordoma is largely restricted to the axial skeleton, and this is in keeping with the purported tumor origin from persistent notochord. The overwhelming majority of chordomas arise within bone; they have never been reported to originate in the intervertebral disk. Historical studies reported that approximately 50% of chordomas develop in the sacrococcygeal region, whereas more modern data with larger numbers of patients suggest an approximately even distribution of tumors between the sacral region, mobile spine, and sphenooccipital region. Within the mobile spine, one series of 40 cases showed that 48% of tumors arose in the cervical spine, 33% in the lumbar spine, and 17% in the thoracic spine.

Classification and Pathology

Pathologically, chordoma is classified into the conventional, chondroid, and dedifferentiated variants. A component of the conventional type is virtually always found in the chondroid and dedifferentiated variants.