Novel MYBL1 gene rearrangements with recurrent MYBL1-NFIB fusions in salivary adenoid cystic carcinomas lacking t(6;9) translocations

Authors

Yoshitsugu Mitani

Bin Liu

Pulivarthi H. Rao

Vishnupriya J. Borra

Mark Zafereo

Randal S. Weber

Merrill Kies

Guillermina Lozano

P. Andrew Futreal

Carlos Caulin

Adel K. El-Naggar

Doi

10.1158/1078-0432.CCR-15-2867-T

PMID: 26631609 · DOI: 10.1158/1078-0432.CCR-15-2867-T · Journal: Clinical Cancer Research (2016)

TL;DR

Mitani et al. performed whole-genome sequencing on 21 salivary adenoid cystic carcinomas (ACCs) and validated findings in 81 additional tumors (total cohort n=102) to characterize the genetic alterations in tumors lacking the canonical t(6;9) translocation and MYB-NFIB fusion. They discovered a novel MYBL1-NFIB gene fusion arising from a t(8;9) translocation, along with other MYBL1 rearrangements (e.g., MYBL1-YTHDF3 fusions and MYBL1 truncations), in 35% of t(6;9)-negative ACCs. All MYBL1 alterations deleted the C-terminal negative regulatory domain and were associated with high MYBL1 expression, while MYB and MYBL1 expression were mutually exclusive. Additionally, 5’-NFIB fusions to non-MYB/MYBL1 partners (XRCC4, NKAIN2, PTPRD, AIG1) were found in a subset of t(6;9)-positive/MYB-NFIB-negative tumors. The study defines new molecular subclasses of ACC.

Cohort & data

Key findings

  • Novel MYBL1-NFIB fusion in t(6;9)-negative ACCs. WGS identified MYBL1 rearrangements in 5 of 12 t(6;9)-negative test-set tumors: 4 with MYBL1-NFIB fusions and 1 with a MYBL1-YTHDF3 fusion via intra-chromosomal rearrangement PMID:26631609.
  • Validation prevalence. FISH detected t(8;9) translocations in 10/36 (27.8%) of t(6;9)-negative validation tumors; 8 of these had confirmed MYBL1-NFIB fusions. Overall, MYBL1 alterations were found in 17% of the 102-tumor cohort (12% with MYBL1-NFIB fusions; 5% with MYBL1-YTHDF3 fusions or MYBL1 truncations). Among all t(6;9)-negative cases, MYBL1-NFIB fusions represented 25% (12/48).
  • Fusion architecture. Breakpoints in MYBL1 localized to introns 8 or 14 (joining at exons 8, 9, 14, or 15); breakpoints in NFIB clustered in intron 10 (joining at exons 11 or 12). All MYBL1 fusion variants retained intact DNA-binding and transactivation domains but disrupted the C-terminal negative regulatory domain.
  • Mutually exclusive MYB/MYBL1 expression. Quantitative RT-PCR on all 102 ACCs showed high MYBL1 mRNA in all 12 MYBL1-NFIB-fusion tumors and both MYBL1-YTHDF3-fusion tumors. t(6;9)-positive tumors expressed high MYB and low MYBL1; MYBL1-rearranged tumors showed the reciprocal pattern.
  • MYBL1 truncations without fusion. Three tumors with high MYBL1 expression but no t(8;9) translocation carried MYBL1 truncations at exon 9 or 10 (AC79: exon 10 → intergenic chr8; AC78: exon 9 → RAD51B intron 5; AC77: cryptic splice into intron 9 generating a PTC).
  • 5’-NFIB fusions to non-MYB/MYBL1 partners. 3 t(6;9)-positive/MYB-NFIB-fusion-negative tumors had 5’ NFIB (exons 1–2) fused to XRCC4 (AC09, t(5;9)), PTPRD/NKAIN2 (AC08), or AIG1 (AC07). All three carried additional translocations placing genomic segments near (~0.1–10 Mb upstream of) the MYB locus and showed high MYB expression.
  • Mutational landscape. Low somatic mutation rates were observed (consistent with prior reports). MUC12 and TFAP4 mutations occurred in MYB-NFIB fusion tumors; MAP10 and OTOF in the MYBL1 fusion group; NOTCH1, SPEN, and MUC16 mutations were enriched in tumors lacking both MYB and MYBL1 fusions.
  • Expression-based subclasses. Unsupervised clustering of 13 fusion-positive tumors separated “long fusions” (breakpoint after exon 11 of MYB or MYBL1) from “short fusions” (exons 8 or 9). GSEA identified 19 gene sets enriched in long fusions (predominantly RNA processing/translation regulation) and 5 in short fusions (tissue development).
  • Clinical correlation. MYB alterations were significantly associated with recurrence and metastasis (P = 0.042) and shorter survival vs MYBL1-altered tumors (P = 0.010, log-rank; Supplementary Fig. S8).

Genes & alterations

  • MYBL1 — recurrent intra- and inter-chromosomal rearrangements in 35% of t(6;9)-negative ACCs, including fusion to NFIB via t(8;9), fusion to YTHDF3, and truncations. All variants delete the C-terminal negative regulatory domain and drive MYBL1 overexpression.
  • MYB — t(6;9) MYB-NFIB fusions present in ~53% of the 102-tumor cohort; MYB alterations correlate with worse outcome than MYBL1 alterations.
  • NFIB — 5’ end (exons 1–2) fuses with diverse partners (MYB, MYBL1, XRCC4, PTPRD, NKAIN2, AIG1); breakpoints in MYBL1-NFIB cluster in intron 10.
  • YTHDF3 — intra-chromosomal fusion partner with MYBL1 in 2/102 tumors.
  • XRCC4, NKAIN2, PTPRD, AIG1 — novel 5’-NFIB fusion partners in t(6;9)-positive/MYB-NFIB-negative tumors.
  • RAD51B — intron-5 partner in a MYBL1 truncation event (AC78).
  • NOTCH1, SPEN, MUC16 — somatic mutations enriched in tumors lacking both MYB and MYBL1 fusions, suggesting NOTCH-pathway preferential involvement in this subset.
  • MUC12, TFAP4 — exclusive mutations in the MYB-NFIB fusion group.
  • MAP10 — exclusive mutation in the MYBL1 fusion group (alongside OTOF).

Clinical implications

  • ACCs comprise three molecular subclasses defined by mutually exclusive translocations: t(6;9)/MYB-NFIB (~53%), t(8;9)/MYBL1-NFIB (~14%), and tumors lacking both translocations (~33%) — with the latter enriched for NOTCH1/SPEN mutations.
  • MYB alterations are associated with recurrence/metastasis (P = 0.042) and shorter survival vs MYBL1-altered tumors (P = 0.010), suggesting MYB status may carry prognostic information.
  • The enrichment of NOTCH1 mutations in fusion-negative ACCs raises the possibility of substratifying these patients for NOTCH-targeted therapy, though no therapeutic intervention was tested in this study.
  • The C-terminal negative regulatory domain truncation is a common mechanism of MYB-family activation, supporting therapeutic strategies that target MYB-family transcription factor activity.

Limitations & open questions

  • The authors note that the oncogenic role and clinical significance of the recurrent mutations identified (MAP10, OTOF, MUC12, TFAP4) require validation in larger cohorts.
  • The mechanism underlying the observed mutually exclusive reciprocal MYB and MYBL1 expression is unknown.
  • Two tumors with t(8;9) translocation but no MYBL1-NFIB fusion expressed low MYBL1 and high MYB, an exception the authors flag but cannot fully explain.
  • The MYBL1 antibody used (HPA008791) recognized only C-terminally intact MYBL1 (exon 10), so MYBL1 protein expression in tumors with exon-8 fusions could not be directly assessed.
  • Differential gene expression analysis between fusion-positive and fusion-negative tumors found no significant changes via supervised analysis; downstream functional consequences of long vs short fusions remain to be experimentally validated.
  • The finding that MYBL1 rearrangements in ACC are translocation/truncation-driven contrasts with pediatric low-grade gliomas where MYBL1 duplication/amplification predominates — the authors propose tissue specificity but do not investigate the mechanism.
  • The clinical association of MYB alterations with worse outcome despite the proposed coordinate functional relationship between MYB and MYBL1 is unresolved; the authors speculate it reflects additional features (genomic instability, epigenetic alterations) of MYB-associated tumors.

Citations from this paper used in the wiki

  • “We identified novel inter- and intra-chromosomal rearrangements of the MYBL1 gene in 35% of t(6;9)-negative ACCs.” (Discussion, p. 8)
  • “FISH analysis identified the t(8;9) translocation in 10 (27.8%) of the 36 t(6;9)-negative tumors; eight of these ACCs also had MYBL1-NFIB fusions” (Results, p. 5)
  • “we identified 12 tumors with MYBL1-NFIB fusions in a total of 102 ACCs (12%), representing 25% (12/48) of all t(6;9)-negative cases. MYBL1 truncations were found in three cases and MYBL1-YTHDF3 fusion was only detected in two cases in the entire cohort.” (Results, p. 7)
  • “Clinicopathologic correlations with the molecular findings of the entire cohort showed only significant association between MYB alterations and recurrence and metastasis (P = 0.042) … Kaplan-Meier analysis also showed a shorter survival for patients with MYB alterations compared to that of patients with MYBL1 alterations (P = 0.010, log-rank test)” (Results, p. 7)
  • “tumors lacking MYB/MYBL1 alterations had mutations in NOTCH1, SPEN, and MUC16 … alterations of the NOTCH pathway would characterize preferentially non-MYB/MYBL1 tumors” (Discussion, p. 9)
  • “all the fusions involving the MYBL1 gene had an intact DNA binding and transactivation domains and a disrupted C-terminal negative regulatory domain” (Results, p. 5)

This page was processed by crosslinker on 2026-05-14.