HIRA loss is frequent but not universal in human fumarate hydratase-deficient renal cell carcinoma and defines divergent oncogenic programs

Abstract

Fumarate hydratase-deficient renal cell carcinoma (FH-dRCC) is a rare and highly aggressive malignancy driven by fumarate accumulation and widespread epigenetic reprogramming. Although Valcarcel-Jimenez et al. have demonstrated that loss of the histone chaperone Hira is a critical and necessary event for the malignant transformation of Fh1-deficient cells through activation of MYC-dependent transcriptional programs based on murine models, by analyzing HIRA protein expression and transcriptomic profiles in a multicenter cohort of human FH-dRCC tumors harboring either germline or somatic FH mutations, we found that HIRA loss was frequent (73.9%, 17/23) but not universal. Although HIRA status was not associated with clinical outcomes, transcriptomic analyses revealed two divergent oncogenic programs: HIRA-negative tumors are characterized by activation of MYC and mTORC1 signaling, whereas HIRA-positive tumors preferentially engage Wnt and Notch pathways. These findings suggest that HIRA loss is not an obligate event in human FH-dRCC tumorigenesis, but instead may direct toward distinct molecular trajectories.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40364-026-00990-4.

Main text

To the editor,

Fumarate hydratase-deficient renal cell carcinoma (FH-dRCC) is a highly aggressive malignancy characterized by early metastasis and marked molecular heterogeneity [1, 2]. The loss of fumarate hydratase (FH) leads to the accumulation of fumarate, driving widespread metabolic and epigenetic rewiring [3–5]. A recent mechanistic study using Clustered Regularly Interspaced Short Palindromic Repeats screening in murine models identified loss of the histone chaperone Hira as a critical event that enables the malignant transformation of Fh1-deficient lesions from premalignant cysts to carcinomas, through non-canonical activation of Myc-dependent transcriptional programs [6]. These findings suggest an indispensable role of Hira in tumorigenesis and implicated the Hira-Myc axis as a potential oncogenic mechanism in Fh1-deficient tumors. However, it remains largely unknown whether this mechanism also applies to human FH-dRCC.

In parallel, human FH-dRCC exhibits profound clinical and morphological heterogeneity [7, 8]. Our group has recently revealed that FH-dRCC exhibits substantial transcriptional heterogeneity, with tumors clustering into three molecular subtypes characterized by distinct oncogenic pathways [1, 8]. Specifically, the C3 subtype displays a strong enrichment of MYC and E2F target signatures, represents a highly proliferative transcriptional state, and correlates with poor response to the combination of immune checkpoint blockade (ICB) and anti-angiogenic therapy [8]. Consequently, these observations raise a key question of whether HIRA loss universally drives the MYC-dependent phenotype observed in human FH-dRCC, or whether it instead defines a specific molecular subset of tumors. Therefore, we aimed to assess HIRA expression in FH-dRCC tissue microarrays and, through transcriptomic analyses, to characterize the distinct oncogenic mechanisms associated with its loss versus retention.

To address this question, we evaluated HIRA protein expression in a multicenter cohort of 23 primary FH-dRCC tumors harboring either germline or somatic FH mutations. RNA-seq data were available for 21 (91.3%, 21/23) patients. Immunohistochemical analysis revealed a dichotomous pattern of HIRA protein expression: 17 of 23 tumors (73.9%) lacked nuclear HIRA expression (HIRA-negative), whereas 6 tumors (26.1%) retained nuclear HIRA expression (HIRA-positive) (Fig. 1A and Table 1).

Fig. 1.

Clinical, prognostic, and transcriptomic characterization of FH-dRCC stratified by HIRA expression. (A) Representative histological and immunohistochemical images of human FH-dRCC. (B) Heatmap summarizing the clinicopathological features, immunohistochemical biomarker expression, RNA-seq data availability, FH germline and somatic alteration, FH variant classification, NF2 co-mutation status, and molecular subtypes of the included FH-dRCC cases. (C) Bar chart comparing the best overall response to first-line systemic therapy between the HIRA-negative and HIRA-positive groups. (D-F) Kaplan-Meier survival curves show DFS (D), PFS (E), and OS (F) of FH-dRCC patients stratified by HIRA expression status (HIRA-positive versus HIRA-negative). p-values were determined by the two-sided log-rank test. (G) GSEA plots showing the enrichment of hallmark gene sets comparing HIRA-negative with HIRA-positive tumors. Abbreviations: FH-dRCC, fumarate hydratase-deficient renal cell carcinoma; DFS, disease-free survival; PFS, progression-free survival; OS, overall survival; GSEA, gene set enrichment analysis; HR, hazard ratio; CI, confidence interval; NR, not reached

Table 1.

Baseline characteristics of included patients

Characteristics	All (N = 23)	HIRA-negative (N = 17)	HIRA-positive (N = 6)	P value
Sex, no.(%)				0.621
Female	8 (34.8)	5 (29.4)	3 (50.0)
Male	15 (65.2)	12 (70.6)	3 (50.0)
Age at diagnosis, no.(%)				1.000
≤40	13 (56.5)	10 (58.8)	3 (50.0)
>40	10 (43.5)	7 (41.2)	3 (50.0)
Prior surgery, no.(%)				0.692
CN	10 (43.5)	6 (35.3)	4 (66.7)
RN	8 (34.8)	7 (41.2)	1 (16.7)
PN	4 (17.4)	3 (17.6)	1 (16.7)
No	1 (4.3)	1 (5.9)	0 (0.0)
Macroscopic appearance, no.(%)				1.000
Solid	13 (56.5)	10 (58.8)	3 (50.0)
Solid and cystic	10 (43.5)	7 (41.2)	3 (50.0)
Morphological features, no.(%)				0.434
Papillary	9 (39.1)	8 (47.1)	1 (16.7)
Adenoid	7 (30.4)	4 (23.5)	3 (50.0)
Tubular	4 (17.4)	3 (17.6)	1 (16.7)
Adenoid+Solid	1 (4.3)	1 (5.9)	0 (0.0)
Nested	1 (4.3)	0 (0.0)	1 (16.7)
Papillary+Adenoid	1 (4.3)	1 (5.9)	0 (0.0)
WHO/ISUP grade, no.(%)				0.539
G3	19 (82.6)	13 (76.5)	6 (100.0)
G4	4 (17.4)	4 (23.5)	0 (0.0)
T stage at initial diagnosis, no.(%)				1.000
<T3a	12 (52.2)	9 (52.9)	3 (50.0)
≥T3a	11 (47.8)	8 (47.1)	3 (50.0)
N stage at initial diagnosis, no.(%)				1.000
N0	12 (52.2)	9 (52.9)	3 (50.0)
N1	11 (47.8)	8 (47.1)	3 (50.0)
M stage at initial diagnosis, no.(%)				1.000
M0	16 (69.6)	12 (70.6)	4 (66.7)
M1	7 (30.4)	5 (29.4)	2 (33.3)
FH alterations, no.(%)				1.000
Somatic	10 (43.5)	7 (41.2)	3 (50.0)
Germline	13 (56.5)	10 (58.8)	3 (50.0)
Molecular subtype, no.(%)				1.000
C1	9 (39.1)	7 (41.2)	2 (33.3)
C2	9 (39.1)	6 (35.3)	3 (50.0)
C3	3 (13.0)	2 (11.8)	1 (16.7)
Unknown	2 (8.7)	2 (11.8)	0 (0.0)
IMDC risk group, no.(%)				0.723
Favorable	1 (4.4)	1 (5.9)	0 (0.0)
Intermediate	14 (60.9)	9 (52.9)	5 (83.3)
Poor	6 (26.1)	5 (29.4)	1 (16.7)
Unknown	2 (8.7)	2 (11.8)	0 (0.0)
First-line systemic treatment (N = 20), no.(%)				1.000
TKI+ICB	12 (60.0)	8 (57.1)	4 (66.7)
TKI	8 (40.0)	6 (42.9)	2 (33.3)

Abbreviations: CN, cytoreductive nephrectomy; RN, radical nephrectomy; PN, partial nephrectomy; WHO, World Health Organization; ISUP, International Society of Urological Pathology; IMDC, International Metastatic Renal Cell Carcinoma Database Consortium; TKI, tyrosine kinase inhibitor; ICB, immune checkpoint blockade

Clinicopathological analyses revealed no significant association between HIRA status and germline or somatic FH mutations (p = 1.000) (Fig. 1B and Table 1). Among the 23 patients, 20 (87.0%) subsequently developed metastatic disease and received at least one line of systemic treatment. Survival analyses further demonstrated that HIRA expression status did not significantly correlate with disease-free survival (HR: 0.684; 95% CI: 0.083–5.618; p = 0.737) (Fig. 1D). Most patients received the combination of ICB and anti-angiogenic therapy (57.1% in the HIRA-negative group and 66.7% in the HIRA-positive group). Response to first-line systemic therapy showed no significant difference between the two groups (p = 1.000) (Fig. 1C), progression-free survival (HR: 0.922; 95% CI: 0.276–3.085; p = 0.895) (Fig. 1E), or overall survival (HR: 0.637; 95% CI: 0.074–5.454; p = 0.678) (Fig. 1F).

Although HIRA status showed no clear clinical associations, transcriptomic analyses revealed that HIRA expression significantly influences the tumorigenesis of FH-dRCC by directing tumors toward distinct molecular states. Gene set enrichment analysis demonstrated that MYC targets (NES = 1.61; P.adj < 0.001) and mTORC1 (NES = 1.42; P.adj = 0.014) signaling were significantly enriched in HIRA-negative tumors (Fig. 1G). In contrast, Wnt/β-catenin (NES = −1.55; P.adj = 0.050) and Notch (NES = −1.56; P.adj = 0.053) signaling pathways were preferentially enriched in HIRA-positive tumors (Fig. 1G). The above findings were further supported in an independent TCGA-KIRP cohort including 10 cases of confirmed or suspicious FH-dRCC (Supplementary Fig. S1).

Notably, these transcriptional signatures correspond closely with recently described molecular subtypes of FH-dRCC. HIRA-negative tumors resemble the C3 proliferation/stemness subtype, characterized by MYC activation [8]. Conversely, HIRA-positive tumors display features consistent with the C2 subtype, which is enriched for Wnt and Notch signaling [8]. The observation that HIRA-positive tumors appear to retain aggresive clinical behavior despite lacking MYC enrichment indicates that alternative oncogenic mechanisms likely compensate for the absence of HIRA loss. FH-dRCC is characterized by profound genomic instability and recurrent co-mutations in genes regulating chromatin organization and cell signaling [1, 7]. Previous genomic studies have identified NF2 as one of the most frequently co-mutated genes in FH-dRCC, with truncating mutations predicted to activate Hippo-YAP signaling [1, 8, 9]. Mutations in epigenetic regulators such as KMT2D may further promote enhancer reprogramming and transcriptional plasticity [8, 10]. These co-occurring mutations in FH-dRCC may shape tumor behavior through distinct biological mechanisms. Within this framework, HIRA status may function as a molecular switch that determines which transcriptional programs dominate during tumor evolution, leading to either MYC-driven proliferation or alternative signaling pathways such as Wnt and Notch signaling.

In conclusion, our integrated clinical and transcriptomic analyses of human FH-dRCC highlight three key insights. First, HIRA loss is a frequent but not universal event in human FH-dRCC. Second, HIRA loss is not significantly associated with clinical outcomes. Third, HIRA status functions as a pivotal molecular determinant that directs the tumor toward divergent transcriptional states, thereby influencing the tumorigenesis of FH-dRCC. Collectively, these clinical observations contextualize the HIRA-MYC oncogenic axis discovered in murine models as one crucial branch within the multidimensional oncogenic landscape of human FH-dRCC.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization: X.S. and Z.H.T. Methodology: X.S. and Z.H.T. Investigation: X.S., Z.H.T, G.X.S, H.Z. and X.M.Z. Analysis: X.S. and Z.H.T. Visualization: X.S. and Z.H.T. Patient samples: X.S., Z.H.T, G.X.S, H.Z. and X.M.Z. Writing (original draft): X.S., Z.H.T, G.X.S, H.Z. and X.M.Z. Writing (review and editing): H.Z. and X.M.Z. All authors have read and approved the final version of the manuscript and agree to be accountable for all aspects of the work.

Funding

This work was supported by the Natural Science Foundation of China (NSFC) (82103097, 82202901, 82203110, 82273047, 82172785, 82403766, 82472673, 82573800). The funding organization had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

This study was approved by the Institutional Review Board of West China Hospital, Sichuan University in compliance with the Declaration of Helsinki. The tissue samples were obtained with written informed consent was obtained from each patient to allow the use of his/her samples and data in research.

Consent for publication

The requirement for informed consent for publication was waived due to the retrospective nature of the study and the use of de-identified data.

Competing interests

The authors declare no competing interests.