Rapid Classification of Sarcomas Using Methylation Fingerprint: A Pilot Study

Abstract

Sarcoma classification is challenging and can lead to treatment delays. Previous studies used DNA aberrations and machine-learning classifiers based on methylation profiles for diagnosis. We aimed to classify sarcomas by analyzing methylation signatures obtained from low-coverage whole-genome sequencing, which also identifies copy-number alterations. DNA was extracted from 23 suspected sarcoma samples and sequenced on an Oxford Nanopore sequencer. The methylation-based classifier, applied in the nanoDx pipeline, was customized using a reference set based on processed Illumina-based methylation data. Classification analysis utilized the Random Forest algorithm and t-distributed stochastic neighbor embedding, while copy-number alterations were detected using a designated R package. Out of the 23 samples encompassing a restricted range of sarcoma types, 20 were successfully sequenced, but two did not contain tumor tissue, according to the pathologist. Among the 18 tumor samples, 14 were classified as reported in the pathology results. Four classifications were discordant with the pathological report, with one compatible and three showing discrepancies. Improving tissue handling, DNA extraction methods, and detecting point mutations and translocations could enhance accuracy. We envision that rapid, accurate, point-of-care sarcoma classification using nanopore sequencing could be achieved through additional validation in a diverse tumor cohort and the integration of methylation-based classification and other DNA aberrations.

Keywords: classification; copy-number; machine learning; methylation; nanopore; sarcoma.

Figure 1

Extracts from an example of the final nanoDx analysis report for sample SARC-09 of a female patient with retroperitoneal MDM2 amplified well-differentiated liposarcoma. (A): Copy-number profile. The x-axis is the chromosomal location, and the y-axis is the read counts (log2 transformation). The MDM2 amplification is shown in chromosome 12. (B): Bar plot of the Random Forest classification voting results. The category (y) axis shows this sample’s 10 most frequent methylation classification votes and the percentage voting rate in the x-axis. (C): Bar plot of the confidence score of the voted Random Forest classification (in B). The most confident classification of this sample is “well/dedifferentiated liposarcoma” (WDLS/DDLS), with a confidence score of 0.45. This confidence score validates the top classification in the voting results (B) as a correct one. (D): t-SNE plot shows the clustering of the methylation pattern of the specific sample (circled cross sign) among the other methylation patterns of the sarcomas in the reference set. It shows that the sample clusters very close to the WDLS/DDLS group (dark yellow), as classified by the Random Forest classifier. Abbreviation: AFH, angiomatoid fibrous histiocytoma; AFX/PDS, atypical fibroxanthoma/pleomorphic dermal sarcoma; ALMO/MPC, angioleiomyoma/myopericytoma; AS, angiosarcoma; ASPS, alveolar soft part sarcoma; CB, chondroblastoma; CCS, clear cell sarcoma of soft parts; CCSK, clear cell sarcoma of the kidney; CHORD, chordoma; CHORD (DD), chordoma (dedifferentiated); CSA (A), chondrosarcoma (group A); CSA (B), chondrosarcoma (group B); CSA (CC), chondrosarcoma (clear cell); CSA (IDH A), chondrosarcoma (IDH group A); CSA (IDH B), chondrosarcoma (IDH group B); CSA [25], chondrosarcoma (mesenchymal); CTRL (BLOOD), control (blood); CTRL [26], control (muscle tissue); CTRL (REA), control (reactive tissue); DFSP, dermatofibrosarcoma protuberans; DSRCT, desmoplastic small round cell tumor; DTFM, desmoid-type fibromatosis; EHE, epithelioid hemangioendothelioma; EMCS, extraskeletal myxoid chondrosarcoma; ES, epithelioid sarcoma; ESS (HG), endometrial stromal sarcoma (high grade); ESS [27], endometrial stromal sarcoma (low grade); EWS, Ewing’s sarcoma; FDY, fibrous dysplasia; GCTB, giant cell tumor of bone; GIST, gastrointestinal stromal tumor; IFS, infantile fibrosarcoma; IMT, inflammatory myofibroblastic tumor; Kaposi, Kaposi sarcoma; LCH, Langerhans cell histiocytosis; LGFMS, low-grade fibromyxoid sarcoma; LIPO, lipoma; LMO, leiomyoma; LMS, leiomyosarcoma; MEL (CUT), melanoma (cutaneous); MLS, myxoid liposarcoma; MO, myositis ossificans; MP, myositis proliferans; MPNST, malignant peripheral nerve sheath tumor; MRT, malignant rhabdoid tumor; NFA, nodular fasciitis; NFB, neurofibroma; NFB (PLEX), neurofibroma (plexiform); OB, osteoblastoma; OFMT, ossifying fibromyxoid tumor; OS (HG), osteosarcoma (high grade); RMS [4], rhabdomyosarcoma (alveolar); RMS [28], rhabdomyosarcoma (embryonal); RMS (MYOD1), rhabdomyosarcoma (MYOD1); SARC (MPNST-like), sarcoma (MPNST-like); SARC (RMS-like), sarcoma (RMS-like); SBRCT (BCOR), small blue round cell tumor with BCOR alteration; SBRCT [29], small blue round cell tumor with CIC alteration; SCC (CUT), squamous cell carcinoma (cutaneous); SEF, sclerosing epithelioid fibrosarcoma; SFT, solitary fibrous tumor; SWN, schwannoma; SYSA, synovial sarcoma; USARC, undifferentiated sarcoma; WDLS/DDLS, well/dedifferentiated liposarcoma; t-SNE, t-distributed stochastic neighbor embedding.