Summary Accurate pathological diagnosis is crucial for optimal management of cancer patients. For the ~100 known central nervous system (CNS) tumour entities, standardization of the diagnostic process has been shown to be particularly challenging - with substantial inter-observer variability in the histopathological diagnosis of many tumour types. We herein present the development of a comprehensive approach for DNA methylation-based CNS tumour classification across all entities and age groups, and demonstrate its application in a routine diagnostic setting. We show that availability of this method may have substantial impact on diagnostic precision compared with standard methods, resulting in a change of diagnosis in up to 12% of prospective cases. For broader accessibility we have designed a free online classifier tool (www.molecularneuropathology.org) requiring no additional onsite data processing. Our results provide a blueprint for the generation of machine learning-based tumour classifiers across other cancer entities, with the potential to fundamentally transform tumour pathology.

Background High-grade osteosarcoma is a primary malignant bone tumour mainly affecting children and young adults. The European and American Osteosarcoma Study (EURAMOS)-1 is a collaboration of four study groups aiming to improve outcomes of this rare disease by facilitating randomised controlled trials. Methods Patients eligible for EURAMOS-1 were aged ≤40 years with M0 or M1 skeletal high-grade osteosarcoma in which case complete surgical resection at all sites was deemed to be possible. A three-drug combination with methotrexate, doxorubicin and cisplatin was defined as standard chemotherapy, and between April 2005 and June 2011, 2260 patients were registered. We report survival outcomes and prognostic factors in the full cohort of registered patients. Results For all registered patients at a median follow-up of 54 months (interquartile range: 38–73) from biopsy, 3-year and 5-year event-free survival were 59% (95% confidence interval [CI]: 57–61%) and 54% (95% CI: 52–56%), respectively. Multivariate analyses showed that the most adverse factors at diagnosis were pulmonary metastases (hazard ratio [HR] = 2.34, 95% CI: 1.95–2.81), non-pulmonary metastases (HR = 1.94, 95% CI: 1.38–2.73) or an axial skeleton tumour site (HR = 1.53, 95% CI: 1.10–2.13). The histological subtypes telangiectatic (HR = 0.52, 95% CI: 0.33–0.80) and unspecified conventional (HR = 0.67, 95% CI: 0.52–0.88) were associated with a favourable prognosis compared with chondroblastic subtype. The 3-year and 5-year overall survival from biopsy were 79% (95% CI: 77–81%) and 71% (95% CI: 68–73%), respectively. For patients with localised disease at presentation and in complete remission after surgery, having a poor histological response was associated with worse outcome after surgery (HR = 2.13, 95% CI: 1.76–2.58). In radically operated patients, there was no good evidence that axial tumour site was associated with worse outcome. Conclusions In conclusion, data from >2000 patients registered to EURAMOS-1 demonstrated survival rates in concordance with institution- or group-level osteosarcoma trials. Further efforts are required to drive improvements for patients who can be identified to be at higher risk of adverse outcome. This trial reaffirms known prognostic factors, and owing to the large numbers of patients registered, it sheds light on some additional factors to consider.

SummaryBackgroundWe designed the EURAMOS-1 trial to investigate whether intensified postoperative chemotherapy for patients whose tumour showed a poor response to preoperative chemotherapy (≥10% viable tumour) improved event-free survival in patients with high-grade osteosarcoma.MethodsEURAMOS-1 was an open-label, international, phase 3 randomised, controlled trial. Consenting patients with newly diagnosed, resectable, high-grade osteosarcoma aged 40 years or younger were eligible for randomisation. Patients were randomly assigned (1:1) to receive either postoperative cisplatin, doxorubicin, and methotrexate (MAP) or MAP plus ifosfamide and etoposide (MAPIE) using concealed permuted blocks with three stratification factors: trial group; location of tumour (proximal femur or proximal humerus vs other limb vs axial skeleton); and presence of metastases (no vs yes or possible). The MAP regimen consisted of cisplatin 120 mg/m2, doxorubicin 37·5 mg/m2 per day on days 1 and 2 (on weeks 1 and 6) followed 3 weeks later by high-dose methotrexate 12 g/m2 over 4 h. The MAPIE regimen consisted of MAP as a base regimen, with the addition of high-dose ifosfamide (14 g/m2) at 2·8 g/m2 per day with equidose mesna uroprotection, followed by etoposide 100 mg/m2 per day over 1 h on days 1–5. The primary outcome measure was event-free survival measured in the intention-to-treat population. This trial is registered with ClinicalTrials.gov, number NCT00134030.FindingsBetween April 14, 2005, and June 30, 2011, 2260 patients were registered from 325 sites in 17 countries. 618 patients with poor response were randomly assigned; 310 to receive MAP and 308 to receive MAPIE. Median follow-up was 62·1 months (IQR 46·6–76·6); 62·3 months (IQR 46·9–77·1) for the MAP group and 61·1 months (IQR 46·5–75·3) for the MAPIE group. 307 event-free survival events were reported (153 in the MAP group vs 154 in the MAPIE group). 193 deaths were reported (101 in the MAP group vs 92 in the MAPIE group). Event-free survival did not differ between treatment groups (hazard ratio [HR] 0·98 [95% CI 0·78–1·23]); hazards were non-proportional (p=0·0003). The most common grade 3–4 adverse events were neutropenia (268 [89%] patients in MAP vs 268 [90%] in MAPIE), thrombocytopenia (231 [78% in MAP vs 248 [83%] in MAPIE), and febrile neutropenia without documented infection (149 [50%] in MAP vs 217 [73%] in MAPIE). MAPIE was associated with more frequent grade 4 non-haematological toxicity than MAP (35 [12%] of 301 in the MAP group vs 71 [24%] of 298 in the MAPIE group). Two patients died during postoperative therapy, one from infection (although their absolute neutrophil count was normal), which was definitely related to their MAP treatment (specifically doxorubicin and cisplatin), and one from left ventricular systolic dysfunction, which was probably related to MAPIE treatment (specifically doxorubicin). One suspected unexpected serious adverse reaction was reported in the MAP group: bone marrow infarction due to methotrexate.InterpretationEURAMOS-1 results do not s...

PurposeEURAMOS-1, an international randomized controlled trial, investigated maintenance therapy with pegylated interferon alfa-2b (IFN-α-2b) in patients whose osteosarcoma showed good histologic response (good response) to induction chemotherapy.Patients and MethodsAt diagnosis, patients age ≤ 40 years with resectable high-grade osteosarcoma were registered. Eligibility after surgery for good response random assignment included ≥ two cycles of preoperative MAP (methotrexate, doxorubicin, and cisplatin), macroscopically complete surgery of primary tumor, < 10% viable tumor, and no disease progression. These patients were randomly assigned to four additional cycles MAP with or without IFN-α-2b (0.5 to 1.0 μg/kg per week subcutaneously, after chemotherapy until 2 years postregistration). Outcome measures were event-free survival (EFS; primary) and overall survival and toxicity (secondary).ResultsGood response was reported in 1,041 of 2,260 registered patients; 716 consented to random assignment (MAP, n = 359; MAP plus IFN-α-2b, n = 357), with baseline characteristics balanced by arm. A total of 271 of 357 started IFN-α-2b; 105 stopped early, and 38 continued to receive treatment at data freeze. Refusal and toxicity were the main reasons for never starting IFN-α-2b and for stopping prematurely, respectively. Median IFN-α-2b duration, if started, was 67 weeks. A total of 133 of 268 patients who started IFN-α-2b and provided toxicity information reported grade ≥ 3 toxicity during IFN-α-2b treatment. With median follow-up of 44 months, 3-year EFS for all 716 randomly assigned patients was 76% (95% CI, 72% to 79%); 174 EFS events were reported (MAP, n = 93; MAP plus IFN-α-2b, n = 81). Hazard ratio was 0.83 (95% CI, 0.61 to 1.12; P = .214) from an adjusted Cox model.ConclusionAt the preplanned analysis time, MAP plus IFN-α-2b was not statistically different from MAP alone. A considerable proportion of patients never started IFN-α-2b or stopped prematurely. Long-term follow-up for events and survival continues.