Clival chordoma research
Home › Tissue and biopsy

The case for tissue

Prepared 3 September 2026 by Kya for a friend, to share with her partner if useful. This is a summary of the published evidence and expert guidance, checked against the original sources. It is not medical advice and it does not replace the team looking after the patient; it is meant to make the conversation with them sharper.

The short version

  1. Nobody treats a suspected chordoma without tissue. Every guideline and consensus document we found (the European (ESMO) bone-sarcoma guideline, the 2026 Global Chordoma Consensus, the Chordoma Foundation, NCCN) says the diagnosis is made by a pathologist, not a scan. Surgery is planned around it. Proton beam therapy needs it. Chemotherapy (for the aggressive subtype) needs it. There is no treatment route that skips it.
  2. "Not having a biopsy" does not mean "not having a procedure". For a skull-base tumour the tissue either comes from a small endoscopic biopsy through the nose, or from the start of a full resection (also through the nose). The only real choice is when the tissue is obtained, and whether the patient commits to a big irreversible operation before anyone knows what the lump is.
  3. the patient's case is exactly the one where the experts say "biopsy first". The 2026 consensus: "If your tumor does not have the usual chordoma appearance on scans or you may have poorly differentiated chordoma, a biopsy before surgery is recommended." Her team has already called it "atypical". She is under 35, one of the Chordoma Foundation's two triggers for testing for the aggressive (INI1-lost) subtype, which is treated differently and sometimes with chemotherapy before surgery. (The other trigger, unusually fast growth, now looks doubtful, see below, but the test costs nothing extra once tissue exists.)
  4. Scans get this wrong a lot. In one big series, 37% of skull-base chondrosarcomas (a much less dangerous tumour) arrived labelled "chordoma". In surgical series of cavernous sinus masses, the scan diagnosis was wrong in roughly 15 to 30%. Some of the things that look like this on MRI are treated with tablets, or steroids, or chemotherapy, or nothing, not skull-base surgery and 70+ Gy of radiation.
  5. The biopsy costs her far less than the disease is costing her now. Typically 1 to 2 hours under anaesthetic, one night in hospital (day-case in some units), pain around 1 to 3 out of 10, up and about within days, nasal crusting for a few weeks. Anaesthetic death risk for a healthy 34-year-old is about 1 in 100,000. The real procedural risk, a bleed or fluid leak needing a second look, is in the low single digits of percent in expert hands, and we show the honest small-series numbers below. Meanwhile 67 to 91% of tumour-related head and face pain improves once the lesion is treated, and nothing tumour-directed can start until it is named.

1. What "a biopsy" actually means here

This is not a needle through the cheek in an outpatient room. For a lesion in the cavernous sinus / clivus the standard route is endoscopic endonasal: under general anaesthetic, a camera and instruments go through the nostril into the sphenoid sinus at the back of the nose, which sits directly in front of the clivus. The surgeon takes a small piece of the lesion and a pathologist looks at it under the microscope and stains it for brachyury (chordoma yes/no), INI1/SMARCB1 (conventional vs poorly differentiated chordoma), and whatever else the appearance suggests (lymphoma markers, pituitary hormones, etc.).

The important point for her partner: this is the same corridor a full resection would use. If she later has the tumour removed, the biopsy track is inside the tissue that gets removed anyway, which is precisely what the guidelines require ("the biopsy track needs to be included in the surgical resection", ESMO 2021).

There is also a percutaneous needle route (through the foramen ovale under X-ray guidance) used by some centres for cavernous sinus lesions. It has a lower yield (86% diagnostic vs roughly 90% or better endoscopic), gives cells rather than a tissue block (which limits the stains that can be done) and a small permanent-deficit rate (3 of 75 in a systematic review), so endoscopic is generally preferred where the lesion is reachable.

Who should do it: every source says the same thing: at the specialist skull-base centre (Derriford's MDT discussed her on 3 September; Southampton's Wessex Neurological Centre is where the family has taken her), by the team that would do any later surgery. The Chordoma Foundation: "Whenever possible, avoid having a biopsy outside of a referral center." This is a legitimate thing for the family to insist on.

2. What treatment would actually look like: the location does not rule out surgery

A natural assumption is that a lump in the cavernous sinus "can't be operated on". That is half right, and the half that is wrong matters for the biopsy decision.

3. What it costs her: the honest numbers

This was her partner's specific worry: the energy to recover, and the risk. Here is what the literature actually shows.

Time and recovery. No paper reports operating time or length of stay for a diagnostic cavernous-sinus biopsy alone, so we used the closest data, full pituitary tumour resections through the same corridor, which are a bigger operation than a biopsy: - Ambulatory (same-day) transsphenoidal surgery protocol: median operating time 82 minutes; 18 of 20 patients went home the same afternoon; no readmissions (Sci Rep 2020). - 50 consecutive pituitary resections without nasal packing: 92% discharged next day, average stay 1.16 days (Thomas 2014). - Pain after endoscopic nasal surgery: mean 1.2/10 on the day, 1.05/10 the next day (124 patients, 2022). - Return to daily activities after transsphenoidal surgery (114 patients): walking median day 1, showering day 5, >97% back to basic daily activities within a month. Nose-blowing and lifting >10 kg are restricted for ~6 to 8 weeks to protect the surgical site; that is a rule, not exhaustion. - Nasal quality of life (SNOT-22 scores) after full endoscopic skull-base surgery: worst at 2 weeks, back to baseline by 3 months (multicentre, 218 patients, Little 2015). In the one series of biopsy-only cavernous sinus procedures (Zhang 2023, 11 patients), there was no measurable change in nasal quality of life at all.

Anaesthetic risk. Royal College of Anaesthetists / NAP7 (2023): about 1 in 100,000 risk of death from anaesthesia for a healthy adult having routine surgery; across 14,201 elective UK cases in the audit there was one death (in an ASA 2 patient), none in ASA 1. A 1.37-million-patient study of healthy patients: anaesthesia-attributable death or serious complication 7.3 per million (Schiff 2014). The anaesthetic is not the risk.

Procedural risk: this is the real one, and it is not zero. The biopsy-specific series are small and some are old, so we report every event: - Zhang 2023 (11 endoscopic cavernous sinus biopsies, two tertiary centres): diagnosis in 10/11; 1 nosebleed requiring return to theatre; no deaths. - Liu & Di 2009 (10 endoscopic cavernous sinus biopsies, one centre): diagnosis in 10/10; 1 internal carotid artery injury (treated with balloon occlusion) and 1 short-lived CSF leak. - Kelley 1999 (7 sphenoid/clival biopsies): 1 CSF leak, 1 bleed controlled at the time; no deaths. - Kang 2022 (39 endonasal biopsies of a nearby target): no CSF leaks or infections.

Pooled crudely that is roughly one "needs further intervention" event per 10 patients in small, partly 15 to 25-year-old series, with zero deaths. For scale, the numbers for a full endoscopic chordoma resection, a far bigger operation, from the 2026 meta-analysis of 1,197 patients are: CSF leak 12%, new cranial nerve palsy 7.5%, stroke 1.2%, major bleed 0.3%; carotid injury runs 0.4% across 8,477 endoscopic pituitary cases and about 2 to 3% in chordoma series, almost always managed by the radiologists without lasting deficit (a 0.4% death rate in the largest cavernous-sinus series is the honest floor). Meningitis essentially only follows a CSF leak. A biopsy sits well below all of these. The family should ask the operating team for their own numbers, carotid injury, CSF leak, cranial nerve damage, return to theatre. That is a fair question and a good team will answer it. Experience matters more than technique: in the 11,000-case carotid-injury study, "operator inexperience may be a more important risk factor than choice of surgical technique."

Does a biopsy make later surgery worse? We could find no evidence that it does. Tumour "seeding" (cells spreading along the surgical route) is a real chordoma phenomenon, but every documented skull-base case followed a full resection, not a biopsy: 1.1% in the largest endoscopic series (Pittsburgh, 173 patients), 2.8 to 7.3% in older open-skull series. No published skull-base biopsy-tract seeding rate exists. The consensus advice is simply that the biopsy track be within the later resection, which the endonasal route achieves.

In one sentence: a night in hospital and a couple of weeks of a blocked nose, with a small but real chance of a bleed or leak needing a second procedure, and negligible anaesthetic risk.

4. What not knowing costs her

(a) The diagnosis may simply be wrong. The team's own label is "atypical". In the radiology literature, "atypical" is the word that triggers widening the differential: - Skull-base chondrosarcoma vs chordoma: "no reliable distinguishing clinical or radiologic features" (Almefty 2007, 109 patients); "no conventional CT or MRI feature reliably distinguishes" them (2025 review). In the Massachusetts General series of 200 skull-base chondrosarcomas, 37% had arrived labelled chordoma. Chondrosarcoma has a 10-year disease-specific survival of 99% versus roughly 35% for chordoma, and low-grade chondrosarcoma may not need radiotherapy at all. On conventional MRI, experienced neurosurgeons picked chordoma vs chondrosarcoma correctly only 40 to 80% of the time (median 60%). - Cavernous sinus masses more broadly: in one surgical series 29% could not be correctly diagnosed on imaging; in an endoscopic clival-tumour series only 38% of "clival tumours" turned out to be chordoma (19% were metastases, 12% chondrosarcoma, plus plasmacytoma and others). - Things that mimic chordoma on MRI and are treated completely differently: lymphoma (chemotherapy, not surgery, not protons), pituitary adenoma extending into the clivus (a "presumed chordoma" encasing both carotid arteries turned out to be a prolactinoma treated with tablets, Karras 2016), plasmacytoma (radiotherapy alone), metastasis (in 36% of reported clival metastases the clivus was the first sign of the cancer), inflammatory disease (Tolosa-Hunt, IgG4, sarcoid: steroids), infection (aspergilloma was 13% of one resected cavernous-sinus series), and chordoid meningioma (separable from chordoma only by the brachyury stain). - In the published series of endoscopic biopsies of cavernous-sinus lesions that imaging could not classify (Zhang 2023), the result changed treatment in more than half of patients: 6 of 11 got something other than radiotherapy: immunotherapy, antibiotics, steroids, chemotherapy or observation. An earlier series of 10 found three nasopharyngeal cancers, a lung metastasis and a fungal infection among the diagnoses. - The growth story, corrected by the timeline. her partner's account (3 Sep): right-sided pain in the forehead, eye, cheek, jaw and neck, starting mildly 12 to 14 months ago, spreading and intensifying, daily for the last 4 to 5 months. Forehead/eye/cheek/jaw is the territory of the three branches of the right trigeminal nerve, which runs through the wall of the cavernous sinus; the trigeminal pain nucleus extends into the upper neck, so neck pain is a normal referral pattern. A lesion in that spot is the obvious explanation for that pain, which means it was almost certainly already there when the "clear" head MRI was done ~10 months ago, either outside the scan's coverage, without contrast, or small enough to be missed. "Nothing to 2 x 3 cm in 8 months" was probably never true. That matters in two directions: it removes much of the pressure toward the fast-growing possibilities (lymphoma, metastasis, inflammatory mass, poorly differentiated chordoma) and moves things toward the slow-growing ones, conventional chordoma, chondrosarcoma (which classically arises off-midline at the petroclival junction, so a right-sided lesion with strictly right-sided symptoms is the textbook chondrosarcoma pointer; chordoma is a midline tumour that can extend sideways), and the nerve-sheath and meningeal tumours of the cavernous sinus. The earlier scan must be formally re-read side by side with the new one; if it was genuinely clean on a proper skull-base protocol, the fast-growth reasoning above comes back.

(b) Nothing tumour-directed can start without tissue. - Radiotherapy/protons: the 2026 consensus: "radiation can be given as the main treatment after a biopsy confirms the diagnosis." The NHS England proton policy for skull-base tumours presupposes maximal safe resection at a specialist unit (i.e. tissue). - Surgery: a chordoma resection, a lymphoma biopsy, a prolactinoma prescription and a steroid course are four different plans. The surgeon cannot plan the operation without knowing which. - The aggressive subtype: poorly differentiated (INI1-lost) chordoma is "exceedingly rare" and mostly a childhood tumour, but adult clival cases with cavernous sinus extension and rapid growth exist (Harada 2024: a 32-year-old). If it is that, the 2026 consensus says "chemotherapy should be considered, possibly before surgery" with response rates above 50%, an option that only exists once INI1 status is known. The Chordoma Foundation's advisory board recommends INI1 testing for all patients under 35 or with unusually fast growth. The patient is under 35; whether the growth was truly fast depends on the earlier scan.

(c) Treatment is what relieves the pain. In every series that reports it, tumour-related head and facial pain improves once the lesion is treated: 80% of cranial neuropathies improved or resolved after endonasal chordoma removal (Fatemi/Kelly 2008); 67 to 75% in a 2023 clival series; 73% adequate relief of tumour-related trigeminal pain after radiosurgery (2025 meta-analysis, 454 patients); disabling headache scores fell from 60 to 10 and self-rated health rose from 50 to 80 after transsphenoidal surgery (prospective, 2024). The patient has been carrying this for a year on painkillers, and for two months it has been closing her eye. The thing draining her is the lesion. Delay does not conserve her energy; it spends it.

(d) Steroids and biopsy have to be sequenced. If she is offered dexamethasone for the pain, a reasonable thing to offer, then unless lymphoma has been excluded the biopsy should come first. The European lymphoma guideline (EHA-ESMO 2024): "Corticosteroid therapy before tissue biopsy should be avoided whenever clinically possible" and "in case of clinical deterioration, urgent biopsy should be carried out before the start of corticosteroids." A 2024 meta-analysis of 1,226 patients: steroids beforehand roughly double to triple the chance of a non-diagnostic biopsy. So delaying the decision while she gets steroids for pain can cause the very outcome her partner fears, a procedure that has to be repeated. Progressive cranial nerve palsies are the one situation where a team might reasonably want steroids before tissue; if so, the guideline's answer is not "no steroids" but "biopsy first, this week, then steroids"; the deterioration is a reason to move faster, not to skip the step.

(e) Time is not neutral: if it is chordoma, size decides the outcome. The single biggest determinant of chordoma survival is whether the surgeon can get all of it out. In a 2024 meta-analysis of 3,871 patients, 5-year progression-free survival was 52% overall but 74% with complete resection; an older meta-analysis found incomplete resection made recurrence 3.8 x and death 5.9 x more likely at five years. What stops complete resection? Size and lateral spread: tumour volume over ~20 cm³ and extension away from the midline (Pittsburgh, Beijing and Seoul series; lateral extension carried a 6 x worse odds of complete removal). Proton therapy tells the same story: local control falls as the residual volume rises (the thresholds in the published series range from about 9 mL (Indiana) to 25 mL (Orsay, Paul Scherrer Institute)) or when the residual compresses the brainstem or optic nerves. A pre-existing cranial nerve or visual deficit at the time of surgery independently predicts a worse outcome, and the patient already has two: a drooping right eyelid since about 1 July and double vision since about 22 August. The lesion is taking the nerves in the cavernous sinus one at a time, in a visible sequence (pain, then lid, then eye movement). Cranial nerves recover better the shorter the time they have been compressed. That progression, more than anything else in this document, is why nothing about her pathway should be on a routine timescale.

Corrected 4 September from the discharge summary: the MRI of 2 September describes a 3.8 cm heterogeneous invasive mass centred on the clivus, with partial encasement of the basilar artery and involvement of the third and fourth nerves. If its other dimensions are in the 2.5 to 3 cm range, that is roughly 12 to 18 cm³, so it is already past the lower proton-control thresholds and approaching the 20 cm³ resection threshold; the earlier 2 by 3 cm figure given verbally was an underestimate. How fast it is growing is what the November 2025 scan will settle. Either way the thresholds are here, not distant, and basilar encasement is itself a reason complete removal gets harder with time. And one more point straight from the MD Anderson series: "Biopsy followed by attempted radical resection at a dedicated center does not affect PFS and, therefore, represents a reasonable first step in management." A biopsy does not spend her chances. Waiting does.

5. What the experts actually say (verbatim)

6. Where the argument is weaker: the honest counter-case

A friend asked me to say so if the evidence didn't support biopsy. It does. But these points are real and her partner deserves them straight:

  1. The guidance is expert opinion, not trial data. Chordoma is one-in-a-million; there is no randomised trial of biopsy vs no biopsy and there never will be. Every recommendation above is graded "V, C" (lowest evidence level). What we have instead is consistent expert consensus plus a large literature of things that looked like chordoma and weren't.
  2. The skull-base exception is genuine. If the scan looked classically like chordoma and the plan were to go straight to a full resection at an expert centre, skipping a separate biopsy would be entirely standard; the tissue comes from frozen section at the start of the operation. So the honest framing is not "biopsy at all costs"; it is "tissue before irreversible treatment, obtained the least harmful way, at an expert centre." If the team's proposal is straight-to-resection with frozen section, that is a tissue-first plan and her partner's objection is really to surgery in general, a different conversation. But given the "atypical" label, the 2026 consensus explicitly points to a biopsy before committing to resection.
  3. The procedure has a real complication rate. The small series above each contain one serious event. Anyone who says "it's nothing" is overselling. The right question is the team's own numbers, and the answer to "what if something goes wrong" is that these are managed in the same theatre by the same team (that is why it should be done at the skull-base centre, Southampton or Derriford, not locally).
  4. A biopsy done badly can harm. The Chordoma Foundation is blunt: "If not done properly, these procedures can cause the chordoma to spread." This is an argument for where, not whether.
  5. A separate biopsy adds a step. If it is conventional chordoma, a biopsy-then-resection path takes a few weeks longer than resection-with-frozen-section. Against that: if it is anything else, the biopsy saves her a major operation she didn't need.
  6. Some cavernous-sinus lesions are treated without tissue. If the CT shows no bone destruction and the MRI is characteristic of a meningioma or a trigeminal schwannoma, skull-base teams do sometimes go to radiosurgery or observation on imaging alone. That is a fair exception, but the team has seen the CT and still said chordoma, which almost always means bone involvement. Ask the bone question (sheet, Q2). If the answer is "bone intact, looks like a nerve-sheath tumour", the conversation changes and this document should be re-read in that light.
  7. Steroid response would not settle it. If someone suggests "try steroids and see if it shrinks", note that chordoma, lymphoma, sarcoid and infections can all respond clinically to steroids (a cavernous sinus chordoma at Mayo did), and steroids compromise the later biopsy. It is not a substitute.

7. Bottom line

Declining tissue is not supported by any guideline we found. The literature's only live debate is how and where to get it. For an "atypical", painful lesion in a 34-year-old that has probably been growing for over a year, the most recent global consensus specifically recommends a biopsy before surgery, so that a poorly differentiated chordoma, or a non-chordoma, is not missed. The procedure is a night in hospital with mild pain and a small, real risk that should be quantified by the team doing it. What she is living with now is worse, and it is the only thing on the table that cannot be fixed without a diagnosis.

The kindest thing her partner can do with his fear is aim it: not at the biopsy, but at making sure it is done at the right place, by the right team, before any steroids, with the nasoseptal flap protected, and with the pain properly controlled while she waits.

Sources

Everything above was checked against the original source where it was reachable; where only an abstract or the Chordoma Foundation's authorised plain-language rendering was available, that is noted. Full research notes with every quotation are held separately.

Guidelines and consensus - Strauss SJ et al. ESMO-EURACAN-GENTURIS-ERN PaedCan bone sarcoma guideline. Ann Oncol 2021;32:1520; chordoma biopsy statement (verbatim). https://www.annalsofoncology.org/article/S0923-7534(21)04280-0/fulltext - Radaelli S, Frezza AM, … Stacchiotti S, Gronchi A. Global Consensus on the Management of Primary Localized Chordoma. JAMA Oncol July 2026 (paywalled); plain-language version: https://www.chordomafoundation.org/treatment-guidelines/new-diagnosis/ (quotes verified 3 Sep 2026) - Stacchiotti S, Sommer J et al. Building a global consensus approach to chordoma. Lancet Oncol 2015;16:e71 (abstract/snippets only). https://pubmed.ncbi.nlm.nih.gov/25638683/ - Stacchiotti S, Gronchi A et al. Best practices for local-regional recurrent chordoma. Ann Oncol 2017;28:1230 (full text); "unusually fast growth" means histology. - Chordoma Foundation, Symptoms and diagnosis / Expert Recommendations booklet; INI1 testing under 35 or fast growth. https://www.chordomafoundation.org/diagnosis/ - NHS England Clinical Commissioning Policy B01/P/d, proton beam therapy for adult skull-base tumours (2015; still the current routinely-commissioned policy). https://www.england.nhs.uk/wp-content/uploads/2018/07/Proton-beam-radiotherapy-high-energy-for-skull-base-tumour-treatment--NHS-overseas-programme-Adult.pdf - Potter et al. Skull base chordoma and chondrosarcoma. Radiographics 2024; "Essential requirements for PBT … are tissue diagnosis". https://pubmed.ncbi.nlm.nih.gov/39298353/ - Ferreri AJM et al. EHA-ESMO primary CNS lymphoma guideline. HemaSphere 2024; steroids before biopsy. https://pmc.ncbi.nlm.nih.gov/articles/PMC11148853/ - Fallon M et al. ESMO cancer pain guideline. Ann Oncol 2018. https://pubmed.ncbi.nlm.nih.gov/30052758/ - NICE CG173 https://www.nice.org.uk/guidance/cg173 · CG140 https://www.nice.org.uk/guidance/cg140 · CG138 https://www.nice.org.uk/guidance/cg138 - Royal College of Surgeons (FDS), Guidelines for the management of trigeminal neuralgia 2021. https://www.rcseng.ac.uk/-/media/files/rcs/fds/guidelines/trigemina-neuralgia-guidelines_2021_v4.pdf - Faculty of Pain Medicine, Core Standards for Pain Management Services in the UK, 2021. https://fpm.ac.uk/standards-guidelines/core-standards

Differential diagnosis and imaging accuracy - Rosenberg AE et al. Chondrosarcoma of the base of the skull, 200 cases. Am J Surg Pathol 1999; 37% arrived labelled chordoma. https://pubmed.ncbi.nlm.nih.gov/10555005/ - Almefty K et al. Chordoma and chondrosarcoma: similar, but quite different. Cancer 2007. https://pubmed.ncbi.nlm.nih.gov/17894390/ - Bozer A, Pekçevik Y. Clival lesions review. Diagn Interv Radiol 2025. https://pmc.ncbi.nlm.nih.gov/articles/PMC12591083/ - Yamazawa E et al. MRI radiomics chordoma vs chondrosarcoma. Cancers 2022; human readers 40 to 80%. https://pmc.ncbi.nlm.nih.gov/articles/PMC9265125/ - Nagaraja et al. Isolated cavernous sinus lesions. J Neurosci Rural Pract 2022; 29% not diagnosable on imaging. https://pmc.ncbi.nlm.nih.gov/articles/PMC9357489/ - Zhang AS et al. Endoscopic endonasal biopsy of undifferentiated cavernous sinus lesions. World Neurosurg 2023. https://pubmed.ncbi.nlm.nih.gov/37004883/ - Messerer M et al. Percutaneous biopsy of parasellar lesions. J Neurosurg 2012. https://pubmed.ncbi.nlm.nih.gov/22098200/ - Metastatic disease to the clivus mimicking chordoma. https://pmc.ncbi.nlm.nih.gov/articles/PMC3774822/ - Karras CL et al. Clival prolactinoma presumed chordoma. 2016. https://pubmed.ncbi.nlm.nih.gov/26885420/ - Transnasal endoscopic clival tumour series (42 patients). Sci Rep 2023. https://pmc.ncbi.nlm.nih.gov/articles/PMC9944302/ - Wang et al. Growth rate of untreated skull base chordoma. World Neurosurg 2020. https://pubmed.ncbi.nlm.nih.gov/31926359/ - Sakai et al. Skull base on routine brain MRI. Jpn J Radiol 2023. https://pmc.ncbi.nlm.nih.gov/articles/PMC10066091/ - Tosefsky et al. Steroids and non-diagnostic biopsy in CNS lymphoma, meta-analysis. Neurosurgery 2024. https://pubmed.ncbi.nlm.nih.gov/38865324/ - Scheichel et al. BMC Cancer 2021 (pooled OR 3.3). https://pmc.ncbi.nlm.nih.gov/articles/PMC8243818/ - Ruff et al. Steroid-responsive cavernous sinus chordoma. J Clin Neurosci 2018. https://pubmed.ncbi.nlm.nih.gov/29685413/

Procedure risk, recovery, seeding - Liu HS, Di X. Endoscopic biopsy of cavernous sinus lesions. 2009. https://pubmed.ncbi.nlm.nih.gov/19452412/ - Kelley TF et al. Endoscopic transsphenoidal biopsy. Am J Rhinol 1999. https://pubmed.ncbi.nlm.nih.gov/10088024/ - Kang H et al. 39 endonasal biopsies. Pituitary 2022. https://pubmed.ncbi.nlm.nih.gov/34471994/ - Dellaretti M et al. Percutaneous cavernous sinus biopsy systematic review. World Neurosurg 2017. https://pubmed.ncbi.nlm.nih.gov/28843761/ - Arend et al. Endoscopic endonasal clival chordoma meta-analysis, 1,197 patients. Oper Neurosurg 2026. https://pubmed.ncbi.nlm.nih.gov/42525859/ - Komotar RJ et al. Endoscopic vs open for clival chordoma. World Neurosurg 2011. https://pubmed.ncbi.nlm.nih.gov/21986431/ - Perry A et al. Carotid injury in 11,149 transsphenoidal cases. 2019. https://pubmed.ncbi.nlm.nih.gov/31319187/ - Koutourousiou M et al. Endoscopic cavernous sinus surgery, 234 patients. 2017. https://pubmed.ncbi.nlm.nih.gov/28450229/ - Ambulatory endoscopic pituitary surgery protocol. Sci Rep 2020. https://pmc.ncbi.nlm.nih.gov/articles/PMC7297807/ - Thomas JG et al. Next-day discharge after pituitary surgery. World Neurosurg 2014. https://pubmed.ncbi.nlm.nih.gov/24239739/ - Return to daily activities after transsphenoidal surgery. Sci Rep 2023. https://pmc.ncbi.nlm.nih.gov/articles/PMC10009828/ - Little AS et al. Sinonasal quality of life, multicentre. J Neurosurg 2015. https://pubmed.ncbi.nlm.nih.gov/25884256/ - NAP7, Royal College of Anaesthetists 2023, ch.16; Schiff JH et al. Br J Anaesth 2014 (1.37M patients). https://pubmed.ncbi.nlm.nih.gov/24801456/ - Fernandes Cabral DT, Gardner PA et al. Iatrogenic seeding after endoscopic endonasal chordoma surgery (1.15%). J Neurosurg 2018. https://pubmed.ncbi.nlm.nih.gov/29271711/ - Krengli et al. Surgical pathway seeding review. 2016. https://pmc.ncbi.nlm.nih.gov/articles/PMC4899426/ - Freeman JL et al. MD Anderson chordoma series; biopsy then resection does not affect PFS. Acta Neurochir 2018. https://pubmed.ncbi.nlm.nih.gov/29270681/

Cost of delay and benefit of treatment - Fiore et al. Skull base chordoma outcomes meta-analysis, 3,871 patients. 2024. https://pubmed.ncbi.nlm.nih.gov/38691853/ - Di Maio S et al. Meta-analysis, 807 patients. J Neurosurg 2011. https://pubmed.ncbi.nlm.nih.gov/21819197/ - Zou et al. Prognostic factors meta-analysis, 1,754 patients. 2018. https://pubmed.ncbi.nlm.nih.gov/29045855/ - Koutourousiou M, Gardner PA et al. Endoscopic endonasal resection of cranial base chordomas, 60 patients. Neurosurgery 2012; volume >20 cm³ limits complete resection. https://pubmed.ncbi.nlm.nih.gov/22592328/ - Wang L, Wu Z et al. Skull base chordoma, 238 patients. J Neurosurg 2017; volume 40 cm³ or more. https://pubmed.ncbi.nlm.nih.gov/28059654/ - Kim YH, Jeon C et al. Endoscopic transclival/transpetrosal approach for clival malignancies. J Neurosurg 2018; paramedian extension OR 6.25 against complete resection. https://pubmed.ncbi.nlm.nih.gov/28574308/ - McDonald MW et al. Residual tumour volume and outcomes for clival chordoma. IJROBP 2016 (Indiana; 9.3 cm³ threshold). https://pubmed.ncbi.nlm.nih.gov/26519991/ - Fung V et al. Proton therapy for skull base chordoma, 106 patients. Radiother Oncol 2018 (Orsay; GTV >25 mL HR 2.2). https://pubmed.ncbi.nlm.nih.gov/29310903/ - Ares C et al. Spot-scanning proton therapy for skull base chordoma/chondrosarcoma. IJROBP 2009 (PSI). https://pubmed.ncbi.nlm.nih.gov/19386442/ - Weber DC et al. Long-term outcomes, pencil-beam scanning protons. Radiother Oncol 2016 (PSI). https://pubmed.ncbi.nlm.nih.gov/27247057/ - Fatemi N, Kelly DF et al. Endonasal chordoma removal; cranial neuropathy improvement. 2008. https://pubmed.ncbi.nlm.nih.gov/18234296/ - Hantelius V et al. Headache after transsphenoidal surgery, prospective. Pituitary 2024. https://pubmed.ncbi.nlm.nih.gov/38767698/ - Tumour-related trigeminal neuralgia meta-analysis. BMC Neurol 2025. https://pmc.ncbi.nlm.nih.gov/articles/PMC12048969/

Poorly differentiated chordoma - Shih AR et al. Mod Pathol 2018. https://pubmed.ncbi.nlm.nih.gov/29483606/ - Harada K et al. Adult clival PDC, 32-year-old. Cureus 2024. https://pmc.ncbi.nlm.nih.gov/articles/PMC10764176/ - Gounder M et al. Tazemetostat basket trial. Nat Commun 2026. https://pubmed.ncbi.nlm.nih.gov/41882006/

[Correction 11 Oct 2026: tazemetostat was withdrawn from all markets in March 2026 (Ipsen, after SYMPHONY-1 showed second blood cancers in 5.7% vs 0); trials and compassionate access stopped. PMID 42599162; research 12 A3. Read anything above or below about it as history, not an option.]

Brief for the family · source file em-docs/em-biopsy-case.md · last edited 11 October 2026