Benign Notochordal Cell Tumour

What it is. The benign notochordal cell tumour (BNCT) is a benign intraosseous tumour of notochordal remnant origin, arising along the path of the embryonic notochord: the clivus, the vertebral bodies and the sacrum. It sits in the notochordal tumour spectrum beside its malignant relative, chordoma, and the two share their brachyury (TBXT) marker.

How common. Notochordal rests are common autopsy findings. BNCTs have been reported in up to about 20% of spines at autopsy, a widely repeated figure for which no primary autopsy series was retrieved here, so treat it as an order of magnitude rather than a measured prevalence. The vast majority remain stable, asymptomatic and undetected; only a tiny fraction come to clinical attention, usually as an incidental finding on axial imaging.

Who. Biopsy-proven series are small, and the largest (Iorgulescu 2013) describes eight adults. There is no established sex or age predilection from cohort data; reported cases are adults.

Why it matters. The entity is a clinical problem almost entirely because of its resemblance to, and possible precursor relationship with, chordoma. Recognising it avoids the morbidity of unnecessary oncological resection, while a chordoma mislabelled as BNCT is a malignant, locally destructive tumour left to progress.

The notochord and its rests. The notochord is the embryonic axial midline structure. It normally regresses, and its only adult remnant is the nucleus pulposus of the intervertebral disc. Notochordal cell rests persist in the clivus, vertebral bodies and sacrococcygeal region, and BNCT, also called a giant notochordal rest or hamartoma, arises from these intraosseous rests.

The extraosseous relative. Ecchordosis physaliphora is a related benign notochordal remnant that is extraosseous and intradural, classically on the dorsal clivus in the prepontine region. It is the intracranial counterpart of the intraosseous BNCT.

Brachyury. Brachyury is the T-box transcription factor T (gene TBXT), the master regulator of notochord development, expressed in all notochordal cells: BNCT, chordoma and nucleus pulposus. Germline TBXT duplication and a common TBXT variant are chordoma susceptibility factors, which is the molecular basis of the shared marker and of the precursor link.

The precursor question. BNCT may be a precursor to chordoma. Most BNCTs remain stable, and only a small proportion behave as precursors, with some chordomas arising adjacent to a BNCT.

The usual story. BNCT is characteristically asymptomatic and incidental, found on CT or MRI performed for unrelated reasons. When patients do present, the reported context is non-specific back or axial pain that prompts imaging; every patient in the Iorgulescu series presented this way. The lesion is then incidental to the symptom rather than the proven pain generator, so "incidental" describes how the lesion relates to the pain, not a screening discovery.

Red flags. These point away from BNCT and toward chordoma:

• Progressive or night pain
• Neurological symptoms: radiculopathy, sacral plexus symptoms, bladder or bowel disturbance
• A palpable presacral mass on rectal examination
• Any imaging evidence of cortical destruction or a soft-tissue component

Symptoms do not decide it. Symptoms favour chordoma, but alone they do not separate the two. The Terzi cases all presented with subacute mild pain, including the one that harboured incipient chordoma; the imaging assessment of aggressiveness is what makes the call.

Examination. Usually normal in BNCT. Any neurological finding or palpable mass should be treated as chordoma until proved otherwise and referred to a specialist spine/tumour centre.

The benign pattern. BNCT consists of sheets of bland, adipocyte-like vacuolated notochordal cells filling the marrow space. It lacks the myxoid or chondroid matrix, the lobular architecture and the bone destruction of chordoma.

Immunohistochemistry. Brachyury, S100, EMA and cytokeratin confirm notochordal differentiation, but they do not by themselves separate BNCT from chordoma.

View larger

Treating destruction and soft-tissue extension as an absolute switch between observation and en-bloc surgery is clean but incomplete. A proposed intermediate group exists, and knowing it is what separates a safe answer from a confident one.

Atypical notochordal cell tumour. Carter and colleagues described four adults, aged 53 to 83, with lumbar or sacral notochordal tumours that fitted neither the BNCT nor the chordoma criteria. Three showed subtle cortical permeation with minimal soft-tissue extension on imaging, yet all four had the histology of a benign notochordal cell tumour, two with focal myxoid change.

How they behaved. Three were followed with serial imaging for 26 to 120 months: two showed no progression, and one grew 3.7 mm over ten years. The pathologists proposed the provisional label atypical notochordal cell tumour for lesions failing the criteria for either diagnosis, arguing that calling them chordoma "precipitates potentially overly aggressive surgical management". With four patients it remains a proposal rather than an established category.

Enhancement can mean real chordoma. The grey zone cuts both ways. In Terzi's three-case series the two classic lesions were observed without progression, while the third, with an atypical lytic pattern and contrast enhancement, underwent en-bloc resection with significant associated morbidity, and the specimen contained coexistent foci of incipient chordoma. An enhancing or lytic notochordal lesion is not over-called when it is treated seriously: the price of getting it wrong is a missed malignancy in one direction and a sacrectomy in the other.

Observe the benign lesion. A benign, asymptomatic BNCT is managed by observation with interval imaging to confirm stability. The support is the largest biopsy-proven series: 16 lesions in 8 patients followed radiographically for a median of 21.6 months (range 8.5 to 71.2), none of which progressed radiographically or symptomatically. The authors state plainly that their conclusion is limited by that short follow-up.

No validated schedule. No validated surveillance interval, imaging modality or stopping point exists. Published follow-up ranges from under a year to ten years and no schedule has been tested, so interval imaging is a reasonable practice justified by the natural history observed so far, not a protocol, and should be set with the specialist centre.

The atypical lesion. Benign histology with subtle cortical permeation or minimal soft-tissue extension, the provisional atypical notochordal cell tumour, was followed rather than resected in three of the four reported cases. Minor permeation alone should prompt specialist review, not a sacrectomy.

Resect the chordoma. Chordoma requires wide or en-bloc oncological resection, with consideration of adjuvant radiotherapy (e.g. proton therapy).

The lesion itself. In the available follow-up, biopsy-proven BNCTs do not progress (Iorgulescu 2013), and the atypical-category cases followed were similarly indolent (Carter 2017). The prognosis of a genuine BNCT is therefore excellent.

Transformation or coexistence with chordoma. This is the complication that matters. The Terzi case proves an atypical notochordal lesion can harbour incipient chordoma, but no series provides a rate: the cohorts are too small and too short to quantify transformation risk, which is precisely why surveillance has no validated stopping point.

The iatrogenic complications. En-bloc sacral or spinal resection carries significant morbidity, as the Terzi resected case illustrates: sacral nerve sacrifice, bladder, bowel and sexual dysfunction, and wound and reconstruction problems. Over-calling a BNCT as chordoma inflicts that morbidity for a lesion that would never have harmed the patient. Both error directions are complications of the diagnosis, and both are minimised by specialist-centre review.

Surveillance as the safety net. Interval imaging is both the management and the safety net: it confirms stability, and it catches the rare lesion that declares itself aggressive early enough for curative resection.