5. Discussion
Chordomas are low‑to‑intermediate‑grade, midline primary bone tumors, with dual epithelial-mesenchymal differentiation. The origin is attributed to embryonal remnants of notochord, with the strongest evidence being the expression of transcription factor brachyury in undifferentiated notochord as well as in chordomas9. Upon reaching 11 mm in embryo development, the notochord undergoes maturation, leading to its obliteration and displacement from the centre towards both cranial and caudal ends. Microscopic foci persist within the vertebral bodies at these locations, which undergo malignant transformation in the third to fourth decade for spheno-occipital lesions and in the fifth to sixth decades for sacrococcygeal lesions. Despite the ectodermal origin, these tumors are usually classified and approached as sarcomas due to being bony tumors. World Health Organization (WHO) classification of tumours of soft tissue and bone 2020 classified chordomas histologically into conventional, dedifferentiated and poorly differentiated, with chondroid chordoma being a subtype of conventional type. Conventional type has characteristic physaliphorous cells, with the matrix appearing like hyaline cartilage in case of chondroid subtype. The latter arise almost always in the base of skull. Dedifferentiated type has a high-grade sarcoma component apart from the conventional chordoma component and are usually diagnosed post RT and in recurrent setting. Role of aberrant p53 inactivation is suggested. Poorly differentiated type is the rarest, with around 60 reported cases, most commonly affecting children and young adults. It usually involves skull base and cervical spine, and rarely the sacrococcygeal region. These tumors usually have loss of INI1 expression, with absence of physaliphorous cells. The prognosis is worse with poorly differentiated and dedifferentiated types10. Chordomas are slow-growing tumors and usually clinically silent till late stages, with pain being the most common manifestation, followed by location-specific neurological symptoms like radiculopathy or bowel-bladder dysfunction. There is associated bony destruction with surrounding soft-tissue mass. Location-wise, sacral chordomas are usually regarded as the commonest ones, with maximum involvement of fourth and fifth sacral vertebrae.
National Comprehensive Cancer Network (NCCN) guidelines 2020 recommend extensive resection of operable chordomas11. Positive resection margins and soft tissue mass larger than 7 cm mandate postoperative RT. No response to chemotherapy has been reported4. Molecular targeted inhibitors and Brachyury vaccine are upcoming promising options. Complete resection is the cornerstone of treatment, with the extent of resection largely determining the recurrence rate. Poor anatomical accessibility, size of tumor, extent of intraoperative blood loss, and high chances of sphincteric and sexual dysfunction due to sacral nerve roots sacrifice, make wide resection feasible in only small number of patients, culminating in high recurrence risk in sacral chordomas12. Among 56 chordomas of sacrum and mobile spine, 33% recurrence rate after extensive resection and 67% after intracapsular or marginal resection have been reported13. Similarly, 66% of 48 patients of mobile spine failed locally in an analysis, with those treated with intralesional excision or RT alone having 100% recurrence in a period of 17-20 months, whereas post margin-free en-bloc resection, only 33% recurred at 56-94 months. Even Intralesional extracapsular resection followed by RT had 75% recurrence at 30 months5.
These tumors are resistant to conventional radiation doses less than 60 Gy14. Dose escalation upto 78Gy attempted with image-guided intensity modulated radiation therapy (IG-IMRT) was shown to decrease five-year local failure rates to 48%, compared to 83% in patients with surgery alone or low-dose RT15. Various studies have highlighted the role of higher doses in skull-base chordomas, especially with particle radiation (alone or in combination with photons), though the dose-response relationship is not consistent across all series14. Following resection of spinal chordoma, adjuvant RT with proton therapy was shown to have 40% recurrence risk, compared to 88% with photons, though the ratio of patients in both arms was skewed16. Five-year local-progression-free survival rate 53%-85% has been reported with proton RT even in sacral and mobile spine chordomas17. The median survival as per the analysis of400 patients, published in 2001, is 6.29 years with 5-year, 10-year and 20-year survival rates decreasing significantly to 67.6%, 39.9%, and 13.1%, respectively, irrespective of race and sex2. Another review of 682 patients, published in 2017 showed the median survival as 9.5 years, with 5-year and 10-year rates as 76% and 47% respectively, likely indicative of the improvement in survival due to advances in treatment modalities. Their multivariate analyses showed that age less than 18 years increased the likelihood of recurrence as well as death, sacral origin was negative predictor of progression-free survival (PFS) and the overall survival (OS) was decreased with recurrence and metastases3. Local recurrence has been regarded as the most significant determinant of mortality, other important factor being distant metastases. A review of 53 cases of poorly differentiated chordomas revealed the mean age of presentation as 9.7 years, the range being 3 months to 42 years old. 100% of the patients’ specimen stained positive for brachyury and panCK, and 96% had INI1 loss. Around 96% had origin in clivus, skull base or cervical spine, and only 3 cases (6%) were reported in sacral region. Out of the 41 patients whose follow-up data was available, only 7 patients (17%) were alive and free of disease. The median survival was reported as 46 months, much lower than that by other studies18.
Due to the usually late diagnosis of chordomas, distant metastases are likely, but less than 5% are detected at presentation, with most being detected way further in the disease course7. Metastases to lung, bone, soft tissues and liver have been reported in upto 43% patients on prolonged follow-up of 16 years6. As per an analysis of 219 patients of chordoma by Young et al, the median time from initial diagnosis to metastasis was 4.8 years, with the median survival after the detection of metastases being 1.7 years, varying with the location of the primary and the metastases8. Others have reported the mean interval from surgery to metastases as six years, with a mean survival of 0.2 years post detection of metastases19. Young et al also showed that 17.8% patients developed metastases, out of which 53.8% did to the lungs. However, 11.8% metastases occurred in the setting of local recurrence and only 5.9% occurred with locally controlled disease, showing statistically significant correlation between local recurrence and metastasis8. Yang et al corroborated the significant difference in the rate of metastases with (33.7%) and without (6.7%) local recurrence20.
The mainstay of treatment of such metastatic as well as advanced or relapsed patients are molecular targeted inhibitors, selected based on gene mutation screening and IHC21. Tyrosine kinase inhibitors (TKI) monotherapy can be used as first-line therapy with Imatinib being the most widely used one22. In case of drug-resistant chordoma, two TKIs or TKI plus mammalian target of rapamycin (mTOR) inhibitor is preferred23. Brachyury vaccine (recombinant Saccharomyces cerevisiae encoding brachyury) followed by RT could not show any difference in overall response, compared to placebo arm, yet remains a good target for developmental therapeutics in chordoma24. Role of Immune checkpoint inhibitors looks promising as per few case series, and multiple trials are ongoing regarding the same25. Enrolment of such patients in clinical trials will pave the way for generation of evidences regarding these newer modalities. Unfortunately, our patient could not be given any systemic therapy due to financial constraints.
In our patient, the rapid growth of local and distant disease can hence be attributed to paediatric age group, sacral origin and the poorly differentiated histology. The peculiar characteristics of our case include poorly differentiated histology in sacral disease, presence of lung metastases upfront, diagnosis of chordoma in paediatric age group and a chordoma progressing rapidly. Ours is probably the only case reported till date with these unique features. The diagnosis at an early age is justified by the poorly differentiated histology, which also justifies rapid local progression and development of early lung metastasis. However, association of this histology with sacral chordoma is very rare. The review of literature shows that paediatric chordomas with poor differentiation have really bad prognosis, hence such cases should be started on definitive treatment without any undue delay. The quest for an effective targeted therapy for INI1 deficient tumors has remained elusive.