Bone Sarcoma

Landmark trial proving adjuvant chemotherapy is mandatory in osteosarcoma. Ended the 'historical controls are good enough' era and made surgery-alone unethical. Everything downstream assumes this baseline.

Population: Non-metastatic high-grade osteosarcoma of extremity, age 5-40, after definitive surgery

Adding ifosfamide to the MAP backbone in high-risk patients did NOT meaningfully improve long-term EFS/OS versus historical COSS-82. Added toxicity without clear survival gain — earliest European signal that intensification has hit a wall.

Population: High-risk non-metastatic osteosarcoma of extremity (large tumor, axial, or young age), neoadjuvant then histology-adapted adjuvant

Neoadjuvant timing did NOT change survival — but it enabled limb-sparing surgery and histologic response assessment, which then became a prognostic and stratification tool. The 'neoadjuvant chemo is standard' dogma is based on surgical feasibility, not a survival RCT.

Population: Non-metastatic high-grade osteosarcoma of extremity, age <=30

MTP-PE added to MAP improved OS by ~8 percentage points; ifosfamide addition failed. Interpretation controversial due to factorial interaction. FDA rejected mifamurtide; EMA approved it. Ifosfamide failure here was the first warning for EURAMOS-1.

Population: Newly diagnosed resectable non-metastatic high-grade osteosarcoma, age <30

First positive TKI signal in osteosarcoma. Short duration of benefit — most patients progressed within 6 months — but reproducible activity established angiogenesis inhibition as a viable axis. Opened the door to regorafenib/cabozantinib trials.

Population: Unresectable relapsed osteosarcoma after standard chemotherapy

Adding pegylated interferon maintenance to MAP in good responders did NOT significantly improve EFS or OS. IFN toxicity drove low completion (~40% completed planned therapy). Another negative EURAMOS arm.

Population: Resectable high-grade osteosarcoma, age <=40, GOOD histologic response (<10% viable tumor) to neoadjuvant MAP

Adding mTOR inhibition to sorafenib did not clear the pre-defined futility bar, though numerically similar to sorafenib alone. Combination toxicity was substantial. Another example of 'add a drug, lose on tolerability'.

Population: Relapsed/refractory osteosarcoma progressing on standard chemo, age >=18

Adding ifosfamide/etoposide to MAP for poor responders did NOT improve EFS/OS. Significant grade 3-4 toxicity and secondary AML excess. MAP backbone confirmed as unbeaten for 40+ years.

Population: Resectable high-grade osteosarcoma, age <=40, POOR histologic response (>=10% viable tumor) to neoadjuvant MAP

Independent European replication of SARC024. Regorafenib delayed progression but did not produce RECIST responses. Combined with SARC024 the result is as convincing as 80-patient data gets in this rare disease.

Population: Metastatic osteosarcoma progressing after >=1 prior chemo line, age >=10

Regorafenib significantly delayed progression in relapsed osteosarcoma adults. No RECIST responses — benefit is disease stabilization, not tumor shrinkage. Toxicity typical of multi-TKI (HFSR, hypertension, fatigue).

Population: Metastatic osteosarcoma progressing on >=1 prior therapy, age >=18

Cabozantinib active in both osteo and Ewing relapse, with actual RECIST responses (unlike regorafenib). Ewing ORR of 26% is one of the highest ever reported in relapsed Ewing. Tolerability issues typical of cabo.

Population: Advanced/metastatic osteosarcoma (n=45) or Ewing sarcoma (n=45) after >=1 prior therapy

Small underpowered study. Pazopanib landed in the same narrow 4-month PFS ~40-50% band as sorafenib/regorafenib/cabozantinib, consistent with a class effect of VEGFR-directed TKIs. Publication also a methodological lesson on RECIST in pulmonary-only osteo.

Population: Metastatic osteosarcoma with lung metastases after standard chemotherapy

Denosumab monotherapy failed as a single-agent anti-tumor drug in osteosarcoma despite strong preclinical rationale in the bone microenvironment. Retains role only as skeletal-related-event prophylaxis. PMID 41159913 indexed 2026 (Janeway KA, Clin Cancer Res); confirm exact year if citing.

Population: Pediatric/AYA recurrent or refractory osteosarcoma, two cohorts (measurable disease and surgically rendered NED)

Adding ifosfamide/etoposide to VDC improved EFS and OS by ~10-15 percentage points in localized Ewing. Metastatic patients did not benefit. Established VDC/IE as the North American standard.

Population: Newly diagnosed localized or metastatic Ewing sarcoma family of tumors, age <30

Original pediatric signal that irinotecan + temozolomide is active in relapsed Ewing. The vincristine was added empirically by later groups, producing the VIT triplet. This small phase 2 is the historical seed of the entire VIT narrative.

Population: Children/AYA with advanced Ewing sarcoma family of tumors after >=1 prior regimen

Simply giving the same chemotherapy more frequently improved EFS by 8 percentage points without increased toxicity. Established dose-dense VDC/IE q2w as the North American standard for localized Ewing.

Population: Newly diagnosed localized Ewing sarcoma family of tumors, age <50

Consistent ~60-70% ORR in relapsed Ewing with outpatient-delivered triplet. One of several VIT cohorts (MSK, CHOP, Warsaw, Virginia) that collectively established VIT without ever running an RCT. rEECur 2024 is now superseding it with high-dose ifosfamide.

Population: Relapsed or refractory Ewing sarcoma across pediatric centers in Poland

In standard-risk patients, cyclophosphamide consolidation is non-inferior to ifosfamide consolidation with less renal toxicity. Allows de-escalation in good-prognosis Ewing.

Population: Standard-risk localized Ewing sarcoma after VIDE induction with good response / small tumor / complete resection

BuMel + auto-HSCT improved EFS by ~14 percentage points in high-risk localized Ewing. One of few modern positive bone-sarcoma intensification RCTs. European standard for this subset.

Population: High-risk localized Ewing (poor histologic response post-induction, or large axial tumor)

First ever randomized relapsed-Ewing trial. High-dose ifosfamide emerged as the new reference standard. Full primary publication not yet indexed in PubMed at time of writing — PMID to verify once the Lancet Oncology / NEJM paper is released. VIT regimen NOT directly in this trial but IT arm informative.

Population: First recurrence / primary refractory Ewing sarcoma, age 4-50

Near-universal histologic response validating RANKL as the central driver in GCTB. Redefined the disease as biologically targetable rather than purely surgical.

Population: Recurrent or unresectable giant cell tumour of bone (GCTB), RANK-ligand–driven osteoclast-rich stromal tumor

Established imatinib as the first systemic therapy with demonstrable activity in chordoma — low RECIST response but meaningful disease stabilization in a disease with no cytotoxic option. Mechanism: PDGFRB pathway inhibition rather than oncogenic addiction.

Population: Advanced PDGFRB-positive unresectable/metastatic chordoma progressing after local therapy; all histologic subtypes allowed

Confirmed in ~280 patients that denosumab provides durable disease control and enables surgical de-escalation in the majority of GCTB patients. Became the definitive registration dataset.

Population: Three parallel cohorts: surgically unsalvageable GCTB; GCTB with surgery associated with severe morbidity; previously enrolled from phase 2. Adults and skeletally mature adolescents.

Combination added modest incremental activity over imatinib alone, primarily in patients who had already progressed on single-agent imatinib — extended PFS by a few months. Supported mTOR co-inhibition hypothesis.

Population: Advanced chordoma progressing on imatinib monotherapy or with PDGFR pathway activation plus downstream mTOR pathway rationale

Durable, years-long disease control confirmed in the largest GCTB denosumab dataset. Safety signal (ONJ, atypical fractures) clarified for long-term dosing, informing current practice of treatment interruptions and holiday schedules.

Population: Combined pooled GCTB cohort (surgically unsalvageable + salvageable with severe morbidity) with extended follow-up through ~5 years of denosumab

First prospective evidence that a targeted therapy produces clinically meaningful disease stabilization in IDH1-mutant chondrosarcoma — a subtype historically refractory to cytotoxic chemo. Responses were not RECIST PRs but PFS and biomarker data supported a real on-target effect.

Population: Advanced unresectable/metastatic conventional chondrosarcoma with centrally confirmed IDH1 R132 mutation, progressed on prior therapy or no standard option

Modest but real disease-control signal in conventional chondrosarcoma with pazopanib — enough to justify the drug as a non-IDH option in a disease with no chemotherapy. Dedifferentiated chondrosarcoma fared worse.

Population: Surgically unresectable or metastatic chondrosarcoma (conventional, dedifferentiated, mesenchymal subtypes allowed), progressive disease required at entry

Numerical response-rate advantage with brachyury vaccine plus radiotherapy but underpowered; toxicity mild. Demonstrated feasibility and immunogenicity of targeting the chordoma master transcription factor.

Population: Locally advanced unresectable chordoma scheduled for standard definitive radiotherapy; brachyury is the near-universal transcription-factor driver in chordoma

PARP inhibition in IDH-mutant chondrosarcoma produced durable disease stabilization in the majority of patients, establishing a second biomarker-driven strategy after IDH1 inhibition. Small N but mechanistically coherent.

Population: Advanced IDH1- or IDH2-mutant mesenchymal sarcomas (chondrosarcoma dominant, plus cholangiocarcinoma excluded per design), refractory to standard therapy

Regorafenib doubled short-term non-progression rate vs placebo and delayed PFS in advanced chondrosarcoma — again without objective responses, consistent with a disease-stabilization paradigm. Randomized placebo-controlled evidence is rare in this tumor.

Population: Metastatic or locally advanced progressive chondrosarcoma (conventional and dedifferentiated subtypes allowed), progressed on or intolerant to at least one prior systemic line

Long-term follow-up confirms durable disease stabilization in a meaningful minority of IDH1-mutant chondrosarcoma patients. Established ivosidenib as a reasonable biomarker-selected option in a setting with no standard therapy.

Population: Same cohort as NCT02073994 chondrosarcoma expansion: advanced IDH1-mutant conventional chondrosarcoma, long-term follow-up update

First report of any clinical response to immunotherapy in chordoma. Generated the hypothesis that chordoma — despite being a low-TMB tumor — has enough PD-L1 and infiltrating lymphocytes to sometimes respond to PD-1 blockade.

Population: Three relapsed advanced chordoma patients after failure of surgery, radiotherapy, and prior systemic therapy (including imatinib)

Pembrolizumab monotherapy essentially inactive in bone sarcoma overall; only dedifferentiated chondrosarcoma showed any meaningful signal. The headline 'IO-in-bone-fails' trial that reset enthusiasm for checkpoint inhibitors in osteosarcoma and Ewing.

Population: Advanced unresectable or metastatic bone sarcoma (osteosarcoma, Ewing sarcoma, chondrosarcoma, dedifferentiated chondrosarcoma; n=40) and soft-tissue sarcoma (n=40) after prior therapy

Combination nivo + ipi tripled response rate over nivo monotherapy but responses were soft-tissue-dominant. Bone sarcoma subtypes again contributed few responders — consistent with SARC028.

Population: Metastatic or unresectable sarcoma (bone and soft-tissue; included chondrosarcoma, dedifferentiated chondrosarcoma, and bone sarcoma subtypes alongside predominant STS histologies) after prior systemic therapy

Expansion confirmed combination IO outperforms monotherapy in selected soft-tissue histologies but did not rescue bone-sarcoma activity. Dedifferentiated chondrosarcoma responses support 'dedifferentiated' as a candidate IO-sensitive bucket.

Population: Metastatic sarcoma expansion cohorts enriched for histologies signalling in original A091401: undifferentiated pleomorphic sarcoma, dedifferentiated liposarcoma, and dedifferentiated/other chondrosarcoma