Multi-omics analysis identifies osteosarcoma subtypes with distinct prognosis indicating stratified treatment

Yafei Jiang, Jinzeng Wang, Mengxiong Sun, Dongqing Zuo, Hongsheng Wang, Jiakang Shen, Wenyan Jiang, Haoran Mu, Xiaojun Ma, Fei Yin, Jun Lin, Chongren Wang, Shuting Yu, Lu Jiang, Gang Lv, Feng Liu, Linghang Xue, Kai Tian, Gangyang Wang, Zifei Zhou, Yu Lv, Zhuoying Wang, Tao Zhang, Jing Xu, Yang Liu, Kewen Zhao, Wei Sun, Yujie Tang, Zhengdong Cai, Shengyue Wang, Yingqi HuaNature Communications 2022 · DOI

TL;DR

Researchers have identified four distinct molecular subtypes of osteosarcoma (OS) by integrating genomic, epigenomic, and transcriptomic data from a 121-patient cohort. This classification moves beyond a "one-size-fits-all" treatment model, identifying specific vulnerabilities—such as PARP inhibitor sensitivity in "HRD-dominant" tumors and metabolic chemoresistance in "MYC-driven" tumors—that could pave the way for precision medicine in a cancer where survival rates have stalled for 40 years.

What Problem Does This Paper Address?

Osteosarcoma is the most common primary malignant bone tumor in children and young adults. While the introduction of neoadjuvant chemotherapy in the 1970s boosted survival rates to 70%, progress has since plateaued. The primary challenge is the disease's extreme "genomic chaos"—it is characterized by massive structural rearrangements rather than simple, recurring point mutations.

Because OS is so heterogeneous, patients respond very differently to standard treatments. This paper seeks to map that complexity by identifying stable molecular "subtypes" that correlate with clinical outcomes, providing a roadmap for stratified (tailored) treatment strategies.

Methods

The authors performed a comprehensive "multi-omics" analysis on the SGH-OS cohort (121 patients). This involved:

  • Whole-Exome Sequencing (WES): To identify somatic mutations and mutational signatures.
  • RNA-seq: To profile gene expression (the transcriptome).
  • DNA Methylation Arrays: To look at epigenetic regulation.
  • Copy Number Alteration (CNA) Analysis: To track large-scale gains or losses of chromosomal segments.

The core innovation was the use of iCluster, a joint multivariate regression algorithm that integrates these disparate data types into a single model. This allowed the researchers to find patterns that wouldn't be visible looking at DNA or RNA alone. They also validated their findings using Patient-Derived Xenograft (PDX) models—growing human tumors in mice to test drug combinations.

Datasets

  • SGH-OS Cohort: A new dataset of 121 primary OS patients from Shanghai General Hospital.
  • TCGA Pan-Cancer Data: Used to compare the "HRD score" of osteosarcoma against 34 other cancer types.
  • COSMIC: Used for identifying known mutational signatures.

Key Insights and Ideas

  • The "TP53" Exception: While most cancers have several frequently mutated genes, OS is unique. Only TP53 was significantly mutated across the cohort. The rest of the "drivers" are mostly large-scale chromosomal gains or losses (CNAs).
  • Four Distinct Flavors: The integration revealed four subtypes:
  • S-IA (Immune Activated): High immune cell infiltration; best prognosis.
  • S-IS (Immune Suppressed): High tumor purity but "exhausted" immune environment.
  • S-HRD (HRD Dominant): Characterized by "Homologous Recombination Deficiency"—a breakdown in DNA repair.
  • S-MD (MYC Driven): Driven by MYC amplification; worst prognosis.
  • Metabolic Resistance: In the S-MD subtype, MYC doesn't just drive growth; it ramps up oxidative phosphorylation (OXPHOS). This metabolic shift makes these tumors highly resistant to standard chemotherapy.
  • The "BRCA-ness" of OS: The researchers found that OS has the second-highest HRD score across all cancers, trailing only ovarian cancer. This suggests that drugs used for BRCA-mutant breast/ovarian cancers (like PARP inhibitors) could be highly effective in a subset of OS patients.

Key Results

  • Survival Stratification: The subtypes showed significantly different 5-year survival rates. S-IA patients fared the best, while S-MD patients had a survival rate of less than 40%.
  • Synergistic Therapy: In S-HRD models, the combination of Cisplatin (standard chemo) and Olaparib (a PARP inhibitor) showed a synergistic effect, significantly shrinking tumors compared to either drug alone.
  • Target Identification: The study identified specific "actionable" targets for each group:
  • S-IA/S-IS: Immune checkpoint inhibitors + anti-VEGF therapy.
  • S-HRD: Platinum-based chemo + PARP inhibitors.
  • S-MD: Anti-MYC agents or OXPHOS inhibitors.

Limitations and Future Work

The authors acknowledge that their cohort, while large for a rare disease like OS, is from a single center (Shanghai), which may introduce regional genetic biases. Furthermore, while the PDX results are promising, the proposed stratified treatments (like the Cisplatin/Olaparib combo) require validation in prospective clinical trials before they can become the standard of care. Future work will likely focus on refining the "HRD score" as a clinical biomarker to identify which patients should skip standard chemo in favor of targeted DNA-repair inhibitors.

Summary updated

Abstract

Osteosarcoma (OS) is a primary malignant bone tumor that most commonly affects children, adolescents, and young adults. Here, we comprehensively analyze genomic, epigenomic and transcriptomic data from 121 OS patients. Somatic mutations are diverse within the cohort, and only TP53 is significantly mutated. Through unsupervised integrative clustering of the multi-omics data, we classify OS into four subtypes with distinct molecular features and clinical prognosis: (1) Immune activated (S-IA), (2) Immune suppressed (S-IS), (3) Homologous recombination deficiency dominant (S-HRD), and (4) MYC driven (S-MD). MYC amplification with HR proficiency tumors is identified with a high oxidative phosphorylation signature resulting in resistance to neoadjuvant chemotherapy. Potential therapeutic targets are identified for each subtype, including platinum-based chemotherapy, immune checkpoint inhibitors, anti-VEGFR, anti-MYC and PARPi-based synthetic lethal strategies. Our comprehensive integrated characterization provides a valuable resource that deepens our understanding of the disease, and may guide future clinical strategies for the precision treatment of OS.
BibTeX
@inproceedings{jiang2022multi,
  author = {Yafei Jiang and Jinzeng Wang and Mengxiong Sun and Dongqing Zuo and Hongsheng Wang and Jiakang Shen and Wenyan Jiang and Haoran Mu and Xiaojun Ma and Fei Yin and Jun Lin and Chongren Wang and Shuting Yu and Lu Jiang and Gang Lv and Feng Liu and Linghang Xue and Kai Tian and Gangyang Wang and Zifei Zhou and Yu Lv and Zhuoying Wang and Tao Zhang and Jing Xu and Yang Liu and Kewen Zhao and Wei Sun and Yujie Tang and Zhengdong Cai and Shengyue Wang and Yingqi Hua},
  title = {{Multi-omics analysis identifies osteosarcoma subtypes with distinct prognosis indicating stratified treatment}},
  booktitle = {Nature Communications},
  year = {2022},
  doi = {10.1038/s41467-022-34689-5},
}
Figure