Gene fusions, while rare, have proven to be significant drivers of tumor formation and proliferation. Despite their oncogenicity, gene fusions provide ideal therapeutic targets in specific cancers, such as in leukemia (BCR-ABL) and non–small cell lung cancer (EML4-ALK).1,2 FGFR-TACC fusions are some of the most common chromosomal translocations across tumor types, are found in approximately 3% of human glioblastomas (GBMs) and other low-grade gliomas, and also occur in cancers elsewhere in the body.3,4
In GBM, FGFR-TACC fusion is a tumor-initiating event, resulting in homogenous expression throughout the tumor that is consistent with clonal development.3,5,6 The signaling consequences of FGFR3-TACC3 fusion include activation of oxidative phosphorylation and mitochondrial biogenesis that influence the basal metabolic rate and maximal oxygen consumption of tumors.7 FGFR3-TACC3 GBMs have lower mutational burdens and copy number alterations, are almost mutually exclusive with other oncogenic receptor tyrosine kinase pathways, are less likely to harbor TP53 mutations, and have mesenchymal methylation patterns.8 Although much has been studied in this fusion, targeted therapy of this fusion in patients with gliomas has yet to yield any significant clinical improvement. However, FGFR kinase inhibitors targeting FGFR3-TACC3 fusions in other solid cancers have shown measured success.9,10 Additionally, lower graded tumors have been less well characterized than GBM for FGFR3-TACC3 fusion.
Here, we examine 6 cases of FGFR3-TACC3 fusion in gliomas of various histopathologies ranging from high to low grades. We also compare their respective clinical outcomes and discuss the current literature on FGFR3-TACC3 fusion. Routine molecular testing and research examining this fusion is needed to increase our understanding of its impact on survival in patients with gliomas.
Summary of Cases
Case 1: IDH1/2–Wild-Type Infiltrating Astrocytoma
A 66-year-old woman with a past medical history of hypertension and hyperlipidemia presented with daily episodes of confusion and disorientation. Brain MRI demonstrated abnormal enhancement in the bilateral frontal lobes extending through the corpus callosum (Fig. 1A and B). The lesion was biopsied and initially diagnosed as a low-grade infiltrating astrocytoma (molecular GBM). Molecular analysis demonstrated IDH1/2 wild type, 19q13 deletion, TERT promoter mutation (c.-124C>T), CDKN2A loss, CDKN2B loss, and FGFR3-TACC3 fusion. The histopathological characteristics and molecular profile together were more consistent with molecular GBM.11 No other surgical interventions were attempted. The patient received concurrent chemoradiation with temozolomide, followed by 3 cycles of adjuvant temozolomide with additional temozolomide withheld due to thrombocytopenia. She had progression and died 1 year after diagnosis.
Preoperative MR images of patients with FGFR3-TACC3 gliomas. T1-weighted postcontrast and T2-weighted FLAIR images of case 1 (A–B), case 2 (C–D), and case 3 (E–F); T1-weighted postcontrast and T2-weighted FLAIR images of case 4 (G–H); T1-weighted postcontrast and T2-weighted images of case 5 (I–J); and T1-weighted postcontrast and T2-weighted FLAIR images of case 6 (K–L).
Case 2: IDH1/2–Wild-Type GBM
A 58-year-old man with a history of hyperlipidemia presented with mild headaches, aphasia, and feeling of déjà vu. MRI showed an enhancing hemorrhagic mass in the left temporal lobe (Fig. 1C and D). Within a few days after presentation, the patient underwent left-sided temporal craniotomy and was given the histopathological diagnosis of a IDH1/2–wild-type GBM. Further molecular analysis identified mutations in TERT promoter (c.-146C>T), PTEN (p.C71Y), and RB1 (p.S829*), as well as FGFR3-TACC3 fusion. He received chemoradiation with temozolomide. After 1 cycle of adjuvant temozolomide, he developed thrombocytopenia and required a transfusion. The patient did not have any tumor recurrence and was alive 90 months after initial diagnosis.
Case 3: IDH1/2–Wild-Type GBM
A 43-year-old woman with a history of hepatitis B presented with nausea, vomiting, and left-sided weakness. MRI demonstrated a deep enhancing right parietal white matter mass involving the splenium of the corpus callosum (Fig. 1E and F). The patient underwent a right-sided parietal craniotomy and was given the histopathological diagnosis of a IDH1/2–wild-type GBM. Subsequent molecular analysis of the tumor showed FGFR3-TACC3 fusion. She was enrolled in a clinical trial and treated with chemoradiation with temozolomide, followed by 6 cycles of adjuvant temozolomide plus veliparib or placebo. Four years 7 months later, the patient reported occasional headache and weakness in her right arm. A new area of enhancement was discovered on MRI in the left frontal lobe that was concerning for progression. Left frontal craniotomy discovered a GBM with TERT promoter mutation (c.-146C>T). After additional chemotherapy and another resection 6 months later, the patient died 60 months after initial diagnosis.
Case 4: IDH1/2–Wild-Type GBM
A 67-year-old man with a past medical history of hypertension and hyperlipidemia presented with balance and speech issues. Brain MRI discovered a right-sided pontine stroke and also incidentally found a ring-enhancing lesion in the left mesial temporal lobe (Fig. 1G and H). The patient underwent gross-total resection and was given the histopathological diagnosis of IDH1/2–wild-type GBM. Molecular analysis showed BCOR, DNMT3A, PTEN, and TP53 loss-of-function mutations with TERT promoter mutation (c.124C>T) and FGFR3-TACC3 fusion. The patient was alive 2 months after diagnosis.
Case 5: Polymorphous Low-Grade Neuroepithelial Tumor of the Young
A 28-year-old woman presented with seizures. MRI demonstrated a left temporal lesion that was thought to be of infectious etiology. She was treated with antibiotics for encephalitis and started on antiseizure medication. Six years later, she had headaches, trouble sleeping, and difficulty with speech and memory. Repeat MRI demonstrated a T2-hyperintense lesion in the left temporal lobe involving the amygdala and hippocampus that was larger than that observed on the initial scan (Fig. 1I and J). The patient underwent a temporal craniotomy for tumor resection and was initially diagnosed with oligodendroglioma on the basis of histopathology. Molecular analysis revealed FGFR3-TACC3 fusion (IDH1/2 and BRAF wild type). After consideration of the molecular and histopathological data, the diagnosis was changed to polymorphous low-grade neuroepithelial tumor of the young (PLNTY) (WHO grade 1). The patient underwent serial imaging to monitor the disease. No recurrence was noted, and the patient was alive 14 months after initial diagnosis.
Case 6: Papillary Glioneuronal Tumor
A 68-year-old woman with a history of seizures that were well controlled with levetiracetam, as well as diabetes and hypertension, presented with sudden-onset left arm weakness and numbness. MRI revealed a hemorrhagic mass in the right parietal lobe (Fig. 1K and L). The patient underwent a right-sided parietal craniotomy with subtotal tumor resection and was given the histopathological diagnosis of papillary glioneuronal tumor. After the first resection, the postoperative plan was serial imaging. In the months after surgery, the patient had repeated seizures that were not controlled with antiepileptic medications. Repeat resection of the tumor was performed and adjuvant radiation therapy was administered. Molecular characterization demonstrated an IDH1/2–wild-type tumor with FGFR3-TACC3 fusion, TERT promoter mutation (c.-124C>T), CDKN2A loss, and CDKN2B loss. The molecular profile was more consistent with a GBM.11 Eleven months after the second resection, the patient underwent a third resection, followed by 12 cycles of adjuvant chemoradiation with temozolomide. Finally, 15 months after the third resection, the tumor progressed and another resection was attempted. Postoperatively, the patient was treated with the FGFR inhibitor erdafitinib without any response, and the patient died 36 months after initial diagnosis.
Variable Histomorphological Characteristics of FGFR3-TACC3 Glioma
The initial histomorphological characterizations of FGFR3-TACC3 fusion glioma describe monomorphous ovoid nuclei, nuclear palisading with thin parallel cytoplasmic processes, an endocrine network of thin capillaries, frequent microcalcifications, and desmoplasia.12,13 Looking for these characteristics and considering fusion, our neuropathologists reviewed the histopathological and molecular data of each case. We observed the typical histomorphology in only case 1 (Fig. 2A–C). This was a case of infiltrating astrocytoma with moderate cellularity, no discernible mitotic activity, mild pleomorphism, and cortical infiltration. The MIB-1 index varied but was as high as 3%–4%.
Variable histomorphological features of patients with FGFR3-TACC3 glioma. Characteristic histomorphology with abundant psammomatous calcification (A), delicate branching vasculature (B), and palisading nuclei (C). GBM with focal giant cell features (D–F). PLNTY (G–I). Neurofilament shows an infiltrative growth pattern (G). CD34 staining shows strong reactivity in tumor (H) and highlights ramified cells in the adjacent cortex (I). Papillary glioneuronal tumor (J–L) with tumor cells highlighted with glial fibrillary acidic protein (L).
Cases 2–4 all had characteristic GBM features such as high cellularity and proliferation with numerous mitoses, frequent necrosis, and microvascular proliferation (Fig. 2D–F). Case 5 had characteristics of PLNTY with focal, strong, and diffuse CD34 reactivity (Fig. 2G–I). Case 6 featured a papillary glioneuronal tumor characterized by a papillary neoplasm with a single layer of plumb neoplastic glial cells surrounding the vascular core (Fig. 2J–L).
Discussion
The histomorphological and histopathological characteristics of each tumor varied dramatically (Fig. 2). Although microcalcifications and spindled tumor cells with a myxoid background have been associated with FGFR3-TACC3 fusion in gliomas, they were only identified in 1 case in our study.12,13 The remaining GBMs showed histomorphological characteristics that were indistinguishable from those of GBMs without fusion. Recent studies have found similar results, in that histopathological characteristics were not always present in these tumors.14 As previously reported by Goethe et al., the tumor sample identified as a papillary glioneuronal tumor had FGFR3-TACC3 fusion (case 6).15 Typically, papillary glioneuronal tumors have a characteristic papillary architecture with tumor cells arranged around monolayer or multilayer vessels, which were visualized in our case.15 The infiltrating astrocytoma in our case series would now be considered a molecular GBM on the basis of its IDH1/2–wild-type status with TERT promoter mutation.11 FGFR3-TACC3 fusion has been previously reported in patients with PLNTY (case 5).16,17
Despite the uniqueness of FGFR3-TACC3 gene fusion, it is not currently listed as an entity-defining mutation in the 2021 WHO classification of CNS tumors.11 Here, we have described 6 patients with FGFR3-TACC3 fusion and varying histopathological and clinical courses (Table 1). The tumors in our study varied from PLNTY to papillary glioneuronal tumor, IDH1/2–wild-type infiltrating astrocytoma, and IDH1/2–wild-type GBM. Interestingly, the clinical courses of the tumors varied. Two of the patients with GBM survived more than 5 years, much longer than the typical median overall survival of around 15 months.18 On the other hand, a patient with a morphologically papillary glioneuronal tumor and multiple recurrences died 3 years after diagnosis despite radiation, chemotherapy, and FGFR inhibitor treatment. It is unusual for a papillary glioneuronal tumor to recur and lead to death.15,19,20 These tumors are rare and have favorable outcomes, with a 5-year progression-free survival rate of 85.9%.19 However, the patient was older than the typical patient with a papillary glioneuronal tumor, and the molecular profile suggested that the tumor may have been a IDH1/2–wild-type GBM with an unusual histological appearance.
Clinical characteristics of patients with FGFR3-TACC3 fusion tumors
Age (yrs)/ Sex | Tumor Diagnosis | Molecular Characteristics | Tumor Location | Treatment | Clinical Outcome |
---|---|---|---|---|---|
66/F | IDH1/2-wt infiltrating astrocytoma (molecular GBM) | IDH1/2 wt; 1p/19q intact, 19q13 deletion; TERT 124C>T; CDKN2A loss; CDKN2B loss; FGFR3-TACC3 fusion | Lt frontal & anterior corpus callosum | Surgical biopsy, chemoradiation w/ temozolomide, & 3 cycles of adjuvant temozolomide | Died 1 yr after diagnosis |
58/M | IDH1/2-wt GBM | IDH1/2 wt; TERT 146C>T; PTEN p.C71Y; RB1 p.S829*; FGFR3-TACC3 fusion | Lt temporal | Resection, chemoradiation w/ temozolomide, & 1 cycle of adjuvant temozolomide | Alive 7 yrs after diagnosis |
43/F | IDH1/2-wt GBM | FGFR3-TACC3 fusion (1st tumor); no fusion & TERT 146C>T (2nd tumor) | Rt parietal (1st tumor); lt frontal (2nd tumor) | 3 resections, chemoradiation w/ temozolomide, adjuvant temozolomide for 6 cycles plus veliparib or placebo, & lomustine | Died 5 yrs after diagnosis |
67/M | IDH1/2-wt GBM | IDH1/2 wt; BCOR LOF; DNMT3A LOF; PTEN LOF; TP53 LOF; TERT 124C>T; FGFR3-TACC3 fusion | Lt mesial temporal | Resection | Alive 2 mos after diagnosis |
28/F | PLNTY | BRAFv600E negative; IDH1/2 wt; FGFR3-TACC3 fusion | Lt temporal | Resection | Alive 14 mos after diagnosis |
68/F | Papillary glioneuronal tumor progressing to atypical anaplastic papillary glioneuronal tumor | IDH1/2 wt; TERT 124C>T; CDKN2A loss; CDKN2B loss; FGFR3-TACC3 fusion | Rt parietal | 4 resections, chemoradiation (temozolomide), & erdafitinib | Died 3 yrs after diagnosis |
LOF = loss of function; wt = wild type.
Despite the various histopathological features and grades of these tumors, the unifying feature was FGFR3-TACC3 fusion. FGFR3-TACC3 fusion occurs more often in GBM than other tumors.4,5 Among patients with IDH1/2–wild-type GBM, FGFR3-TACC3 fusion has been associated with better survival when other factors have been controlled for in multivariate analysis.5,8 In GBM patients receiving standard chemoradiation, those with FGFR3-TACC3 fusion had better median overall survival (29.1 months) than patients without fusion (20.5 months).5 This is consistent with the outcomes of our patients, 1 of whom was still alive 7 years after diagnosis with tumor progression and 2 others who have survived 5 and 1 years after diagnosis, respectively. Although there are reports of lower grade glioma with FGFR3-TACC3 fusion, the outcomes have been less well characterized.4,21 All our patients had IDH1/2–wild-type tumors, which is unsurprising because FGFR3-TACC3 fusion is mutually exclusive with IDH1/2 mutation.5 Given the varied histomorphological characteristics of fusion tumors, expansion of molecular testing to regularly include this fusion is advisable.
Research on gene fusion has yielded great opportunities for personalized medicine and specific chemotherapies that directly impact the survival of patient with other cancers.1,2 Some case reports have indicated that specific targeting of FGFR3-TACC3 fusion may be worthwhile. In 1 study, 2 patients with recurrent GBM received erdafitinib, a pan FGFR tyrosine kinase inhibitor. Both patients who received this therapy were still alive 134 and 287 days after starting therapy; however, both had progression.5 Another case report described the use of anlotinib to treat a patient with recurrent GBM with FGFR3-TACC3 fusion. This patient had a partial response that was maintained for > 17 months of follow-up.22 Unfortunately, when evaluated with clinical trials, these results have not yet demonstrated benefit. A phase II trial of infigratinib in patients with recurrent GBM and FGFR-TACC fusion or FGFR amplification reported partial response or stable disease in only one-third of patients.23 In our study, 1 patient (case 6) received the FGFR inhibitor erdafitinib but no response was seen, although it was administered late in the clinical course. Despite these setbacks, currently several experimental therapeutics for FGFR-TACC fusion in GBM are under investigation by registered clinical trials.24–27
Conclusions
Our study examined 6 patients with FGFR3-TACC3 fusion gliomas. The varied histomorphological characteristics mean that morphological features alone cannot always predict presence of fusion. Therefore, gliomas with varying histological characteristics and grades should undergo molecular testing. Consistent with previously reported results, some of our GBM patients had improved survival compared with historical data. Routine testing for FGFR3-TACC3 fusion and subsequent stratification may be warranted in clinical trials of IDH1/2–wild-type GBM.
Acknowledgments
Dr. McDonald thanks BRASS (Baylor Research Advocates for Student Scientists) and the McNair Medical Institute for their financial support.
Disclosures
Dr. Rao is a consultant for Monteris Medical.
Author Contributions
Conception and design: Mandel, McDonald. Acquisition of data: Mandel, McDonald, Athukuri, Anand, Goodman, Lu. Analysis and interpretation of data: Mandel, McDonald, Athukuri, Anand, Gopakumar, Goodman, Lu. Drafting the article: Mandel, McDonald, Lu. Critically revising the article: Mandel, McDonald, Athukuri, Anand, Gopakumar, Patel, Rao, Lu. Reviewed submitted version of manuscript: Mandel, McDonald, Athukuri, Anand, Gopakumar, Jalali, Patel, Rao, Lu. Approved the final version of the manuscript on behalf of all authors: Mandel. Statistical analysis: Mandel, McDonald. Administrative/technical/material support: Mandel. Study supervision: Mandel, McDonald.
References
- 1↑
Ali MAM. Chronic myeloid leukemia in the era of tyrosine kinase inhibitors: an evolving paradigm of molecularly targeted therapy. Mol Diagn Ther. 2016;20(4):315–333.
- 2↑
Gerber DE, Minna JD. ALK inhibition for non-small cell lung cancer: from discovery to therapy in record time. Cancer Cell. 2010;18(6):548–551.
- 3↑
Singh D, Chan JM, Zoppoli P, et al. Transforming fusions of FGFR and TACC genes in human glioblastoma. Science. 2012;337(6099):1231–1235.
- 4↑
Lasorella A, Sanson M, Iavarone A. FGFR-TACC gene fusions in human glioma. Neuro Oncol. 2017;19(4):475–483.
- 5↑
Di Stefano AL, Fucci A, Frattini V, et al. Detection, characterization, and inhibition of FGFR-TACC fusions in IDH wild-type glioma. Clin Cancer Res. 2015;21(14):3307–3317.
- 6↑
Wang J, Cazzato E, Ladewig E, et al. Clonal evolution of glioblastoma under therapy. Nat Genet. 2016;48(7):768–776.
- 7↑
Frattini V, Pagnotta SM, et al. A metabolic function of FGFR3-TACC3 gene fusions in cancer. Nature. 2018;553(7687):222–227.
- 8↑
Mata DA, Benhamida JK, Lin AL, et al. Genetic and epigenetic landscape of IDH-wildtype glioblastomas with FGFR3-TACC3 fusions. Acta Neuropathol Commun. 2020;8(1):186.
- 9↑
Loriot Y, Necchi A, Park SH, et al. Erdafitinib in locally advanced or metastatic urothelial carcinoma. N Engl J Med. 2019;381(4):338–348.
- 10↑
Necchi A, Castellano DE, Mellado B, et al. Fierce-21: Phase II study of vofatmab (B-701), a selective inhibitor of FGFR3, as salvage therapy in metastatic urothelial carcinoma (mUC). J Clin Oncol. 2019;37(7 suppl):409.
- 11↑
Louis DN, Perry A, Wesseling P, et al. The 2021 WHO Classification of Tumors of the Central Nervous System: a summary. Neuro Oncol. 2021;23(8):1231–1251.
- 12↑
Broggi G, Piombino E, Altieri R, et al. Glioblastoma, IDH-wild type with FGFR3-TACC3 fusion: when morphology may reliably predict the molecular profile of a tumor. A case report and literature review. Front Neurol. 2022;13:823015.
- 13↑
Bielle F, Di Stefano AL, Meyronet D, et al. Diffuse gliomas with FGFR3-TACC3 fusion have characteristic histopathological and molecular features. Brain Pathol. 2018;28(5):674–683.
- 14↑
Gilani A, Davies KD, Kleinschmidt-DeMasters BK. Can adult IDH-wildtype glioblastomas with FGFR3:TACC3 fusions be reliably predicted by histological features? Clin Neuropathol. 2021;40(3):165–167.
- 15↑
Goethe EA, Youssef M, Patel AJ, Jalali A, Goodman JC, Mandel JJ. Recurrent papillary glioneuronal tumor. World Neurosurg. 2019;128:127–130.
- 16↑
Giulioni M, Marucci G, Cossu M, et al. CD34 Expression in low-grade epilepsy-associated tumors: relationships with clinicopathologic features. World Neurosurg. 2019;121:e761–e768.
- 17↑
Huse JT, Snuderl M, Jones DTW, et al. Polymorphous low-grade neuroepithelial tumor of the young (PLNTY): an epileptogenic neoplasm with oligodendroglioma-like components, aberrant CD34 expression, and genetic alterations involving the MAP kinase pathway. Acta Neuropathol. 2017;133(3):417–429.
- 18↑
Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med. 2005;352(10):987–996.
- 19↑
Ahmed AK, Dawood HY, Gerard J, Smith TR. Surgical resection and cellular proliferation index predict prognosis for patients with papillary glioneuronal tumor: systematic review and pooled analysis. World Neurosurg. 2017;107:534–541.
- 20↑
Carangelo B, Arrigucci U, Mariottini A, et al. Papillary glioneuronal tumor: case report and review of literature. G Chir. 2015;36(2):63–69.
- 21↑
Johnson A, Severson E, Gay L, et al. Comprehensive genomic profiling of 282 pediatric low- and high-grade gliomas reveals genomic drivers, tumor mutational burden, and hypermutation signatures. Oncologist. 2017;22(12):1478–1490.
- 22↑
Wang Y, Liang D, Chen J, et al. Targeted therapy with anlotinib for a patient with an oncogenic FGFR3-TACC3 fusion and recurrent glioblastoma. Oncologist. 2021;26(3):173–177.
- 23↑
Lassman AB, Sepúlveda-Sánchez JM, Cloughesy T, et al. OS10.6 Infigratinib (BGJ398) in patients with recurrent gliomas with fibroblast growth factor receptor (FGFR) alterations: a multicenter phase II study. Neuro Oncol. 2019;21(suppl 3):iii21-iii22.
- 24↑
Parker Kerrigan BC, Ledbetter D, Kronowitz M, et al. RNAi technology targeting the FGFR3-TACC3 fusion breakpoint: an opportunity for precision medicine. Neurooncol Adv. 2020;2(1):a132.
- 25
Oxidative Phosphorylation Targeting IN Malignant Glioma Using Metformin Plus Radiotherapy Temozolomide. NCT04945148. ClinicalTrials.gov. Accessed October 20, 2022. https://clinicaltrials.gov/ct2/show/NCT04945148
- 26
A Phase II Study of Anlotinib Combined With Dose-Dense Temozolomide for the First Recurrent or Progressive Glioblastoma After STUPP Regimen. NCT04547855. ClinicalTrials.gov. Accessed October 20, 2022. https://clinicaltrials.gov/ct2/show/NCT04547855
- 27↑
Phase II Clinical Trials on Anlotinib for the Treatment of Recurrent Glioblastoma. NCT04004975. ClinicalTrials.gov. Accessed October 20, 2022. https://clinicaltrials.gov/ct2/show/NCT04004975