Integrated Molecular Characterization of Fumarate Hydratase-Deficient Renal Cell Carcinoma

Author Manuscript Published OnlineFirst on January 7, 2021; DOI: 10.1158/1078-0432.CCR-20-3788 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Integrated Molecular Characterization of Fumarate Hydratase-deficient Renal Cell Carcinoma

Guangxi Sun 1,#, Xingming Zhang 1,#, Jiayu Liang 1,#, Xiuyi Pan 2,#, Sha Zhu 1,#, Zhenhua Liu 1,#, Cameron M. Armstrong 3, Jianhui Chen 4, Wei Lin 5, Banghua Liao 1, Tianhai Lin 1, Rui Huang 6, Mengni Zhang 2, Linmao Zheng 2, Xiaoxue Yin 2, Ling Nie 2, Pengfei Shen 1, Jinge Zhao 1, Haoran Zhang 1, Jindong Dai 1, Yali Shen 7, Zhiping Li 7, Jiyan Liu 8, Junru Chen 1, Jiandong Liu 1,9, Zhipeng Wang 1, Xudong Zhu 1, Yuchao Ni 1, Dan Qin 10, Ling Yang 11, Yuntian Chen 11, Qiang Wei 1, Xiang Li 1, Qiao Zhou 2, Haojie Huang 12,＋, Jin Yao 11,＋,*, Ni Chen 2,＋,*, Hao Zeng 1,＋,*

1. Department of Urology, Institute of Urology, West China Hospital, Sichuan University, 610041, Chengdu, China 2. Department of Pathology, West China Hospital, Sichuan University, 610041, Chengdu, China 3. Department of Urology and Comprehensive Cancer Center, University of California Davis, Sacramento, 95817, CA, USA 4. Department of Urology, Fujian Medical University Union Hospital, Fuzhou, China 5. Department of Urology, Zigong Fourth People’s Hospital, 643000, Zigong, China 6. Department of Nuclear medicine, West China Hospital, Sichuan University, 610041, Chengdu, China 7. Department of Oncology, West China Hospital, Sichuan University, 610041, Chengdu, China 8. Department of Biotherapy, West China Hospital, Sichuan University, 610041, Chengdu, China

Downloaded from clincancerres.aacrjournals.org on October 4, 2021. © 2021 American Association for Cancer Research. Author Manuscript Published OnlineFirst on January 7, 2021; DOI: 10.1158/1078-0432.CCR-20-3788 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

9. Department of Urology, The First Affiliated Hospital, School of Medicine, Zhejiang University, 310009, Hangzhou, China 10. The Bioinformatics Department, Basebiotech Co., Ltd, 610000, Chengdu, China 11. Department of Radiology, West China Hospital, Sichuan University, 610041, Chengdu, China 12. Departments of Biochemistry and Molecular Biology and Urology, Mayo Clinic College of Medicine and Science, Rochester, 55905, MN, USA

# These authors are co-first authors and contributed equally. ＋ These authors are co-senior authors. * These authors are co-corresponding authors.

Running title Genomic and epigenomic profiling of FH-deficient RCC

Corresponding Author: Address: No.37 Guoxue Alley, Wuhou District, Chengdu City, Sichuan Province, PR China; Phone number: +86 28 85422026; Fax: +86 28 85582944. Email address: [email protected] (Jin. Yao); [email protected] (Ni. Chen); [email protected] (Hao. Zeng)

Conflicts of interest The authors declare no conflicts of interest.

Words count: 3,903 words Total number of Figures and Tables: 4 figures and 1 table in main manuscript, 6 figures and 4 tables in supplementary appendix.

Translational Relevance Fumarate hydratase-deficient renal cell carcinoma (FH-deficient RCC) is a rare but lethal malignancy, and its genetic basis is poorly understood. We comprehensively profiled untreated primary FH-deficient RCC tissues to characterize the molecular landscape. We identified that DNA methylation profiles could be the biological foundation for tumor pathogenesis; hence, compared with other RCC subtypes, methylation sequencing should be recommended for patients with FH-deficient RCC. Moreover, FH-deficient RCC was characterized by an immunogenic phenotype despite its low overall mutation burden. Patients with this tumor type could benefit from immune checkpoint blockade-based strategies. Our results provide insights into the biology of FH-deficient RCC and pave the way for the development of novel promising therapies.

Abstract Purpose: Fumarate hydratase-deficient renal cell carcinoma (FH-deficient RCC) is a rare but lethal subtype of RCC. Little is known about the genomic profile of FH-deficient RCC, and the therapeutic options for advanced disease are limited. To this end, we performed a comprehensive genomics study to characterize the genomic and epigenomic features of FH-deficient RCC. Experimental Design: Integrated genomic, epigenomic, and molecular analyses were performed on 25 untreated primary FH-deficient RCCs. Complete clinicopathological and follow-up data of these patients were recorded. Results: We identified that FH-deficient RCC manifested low somatic mutation burden (median 0.58 mutations per megabase), but with frequent somatic copy number alterations. The majority of FH-deficient RCCs were characterized by a CpG sites island methylator phenotype (CIMP), displaying concerted hypermethylation at numerous CpG sites in genes of transcription factors, tumor-suppressors and tumor hallmark pathways. However, a few cases (20%) with low metastatic potential showed relatively low DNA methylation levels, indicating the heterogeneity of methylation pattern in FH-deficient RCC. Moreover, FH-deficient RCC is potentially highly immunogenic, characterized by increased tumor T cell infiltration but high expression of immune checkpoint molecules in tumors. Clinical data further demonstrated that patients receiving immune checkpoint blockade-based treatment achieved improved progression free-survival over those treated with antiangiogenic monotherapy (median: 13.3 vs. 5.1 months, P = 0.03). Conclusions: These results reveal the genomic features and provide new insight into potential therapeutic strategies for FH-deficient RCC.

1 Introduction

2 Fumarate hydratase-deficient renal cell carcinoma (FH-deficient RCC) is

3 a rare subtype of kidney cancer characterized by pathologic germline/somatic

4 mutation of the fumarate hydratase (FH) gene, with a lack of FH staining on

5 immunohistochemistry (IHC). FH-deficient RCC is difficult to diagnose, since it

6 frequently shows mixed pathologically architectural patterns; therefore, it is

7 highly susceptible to be misdiagnosed as other RCC subtypes (1,2). Clinically,

8 most cases of FH-deficient RCC are very aggressive and frequently present

9 with locally advanced or metastatic disease at initial diagnosis (3).

10 Unfortunately, owing to its rarity, little is known about its genetic basis, and

11 there are currently no standard treatment strategies for patients with advanced

12 disease.

13 To date, our knowledge of the genetic basis of FH-deficient RCC mainly

14 comes from the reports of the Cancer Genome Atlas (TCGA) database. Using

15 comprehensive genomic analyses, TCGA identified a distinct group of papillary

16 RCC characterized by a CpG island methylator phenotype (CIMP), of which

17 five cases were found to harbor germline or somatic FH mutation (4). However,

18 due to the limited cases in the TCGA cohort, specific research aiming to

19 explore the genomic landscape of FH-deficient RCC is still lacking. In this

20 study, we performed a high-throughput genomic and epigenomic analysis of

21 untreated primary FH-deficient RCC tumors from multi-institutional patient

22 cohorts in order to provide molecular insights into FH-deficient RCC and

23 suggest paths for the development of biology-driven precision medicine.

25 Materials and Methods

26 Patient Identification

27 We cooperated with nine medical centers to collect FH-deficient RCC

28 samples. All suspicious RCC tumors were referred for pathologic consultation

29 at our medical center (West China Hospital). Between 2014 and 2019, a total

1 of 429 RCC samples were retrospectively analyzed (Figure S1). All candidate

2 cases were reviewed by two experienced uropathologists (Ni Chen and

3 Mengni Zhang) and subsequently subjected to immunohistochemical staining

4 (IHC) of FH protein. Diagnosis of FH-deficient RCC was confirmed by Sanger

5 sequencing for FH mutation (Figure S2). Clinicopathological data were

6 retrospectively collected. A total of 25 cases with FH-deficient RCC were

7 identified. Untreated primary tumor tissues and matched germline samples

8 were collected. DNA and methylation sequencing of formalin-fixed

9 paraffin-embedded (FFPE) tissues were subsequently performed. Twenty-two

10 FFPE tumor tissues and matched adjacent normal tissues (n = 12) or blood

11 samples (n = 10) were collected for whole-exome sequencing (WES) to

12 assess the genomic mutation characteristics. Targeted sequencing was

13 performed on two biopsy tumor tissues (FHRCC18 and FHRCC24) and

14 matched blood samples due to insufficient tumor DNA for WES. One patient

15 (FHRCC14) had only a blood sample for WES due to insufficient tumor tissue.

16 EPIC array was performed on 20 tumors and 12 adjacent normal tissues. The

17 study was conducted in accordance with the Declaration of Helsinki and all

18 clinical samples were acquired with written informed consents under

19 permission from the Ethics Committee of West China Hospital of Sichuan

20 University. All patients or family members provided written consent for genetic

21 analysis.

23 Clinicopathological characteristics

24 Clinicopathological data were retrospectively collected, including age,

25 gender, family history, metastatic sites, TNM stage, histological type, ISUP

26 grade, surgery type, and systemic treatment type. Synchronous metastasis

27 was defined as the diagnosis of a distant metastasis at the time of the primary

28 renal cell carcinoma diagnosis. Metachronous metastasis was defined as the

29 metastasis occurred after three months postoperatively.

2 Outcomes

3 Patients visited the urologists for regular evaluations (physical

4 examination, radiologic assessment and laboratory tests) every 4-6 weeks.

5 Response was defined by Response Evaluation Criteria in Solid Tumors

6 (RECIST) version 1.1 (5). The end-points were objective response rate (ORR),

7 disease control rate (DCR), disease-free survival (DFS), progression-free

8 survival (PFS), and overall survival (OS). ORR and DCR were assessed

9 according to RECIST criteria 1.1. ORR was defined as the proportion of

10 patients with complete or partial responses as the best radiological response

11 to therapy. DCR encompassed the proportion of patients who achieved an

12 objective response plus stable disease. DFS was defined as the time from

13 diagnosis to local or regional recurrence or distance metastasis or death from

14 any cause. PFS was defined as the time from initial diagnosis to disease

15 progression or death. First-line PFS was defined as the time from the start of

16 first-line systemic treatment to disease progression or death. OS was defined

17 as the time from initial diagnosis to all-cause death. Systemic treatment OS

18 was defined as the time from the start of systemic treatment to all-cause death.

20 WES and methylation sequencing

21 For samples performing WES, high-quality genomic DNA was extracted

22 using the GeneRead DNA FFPE Kit (180134, QIAGEN, Hilden, GER)

23 according to the manufacturer’s specifications. Exome capture was performed

24 using Agilent SureSelect Human All ExonV6 kit (Technologies, CA, USA)

25 followed by paired-end sequencing using the Illumina Hiseq Xten platform

26 (Illumine Inc, CA, USA). For panel sequencing, DNA was extracted and

27 estimated by a targeted sequencing strategy capturing all exons of 642

28 tumor-related genes (Table S1). Human Infinium MethylationEPIC BeadChip

29 (Illumina) was used to analyze DNA methylation. Extracted DNA was treated

1 with RNase A/T1 Mix (Thermo Scientific, Oberhausen, Germany) and

2 subsequently purified using the Genomic DNA Clean & Concentrator™−10 Kit

3 (Zymo Research, CA, USA). Bisulfite pyrosequencing was used to validate the

4 results of the microarray analysis. The WES data generated in this paper have

5 been deposited at the NCBI Sequence Read Archive (SRA) hosted by the NIH

6 (SRA accession: PRJNA533711). The methylation data reported in this paper

7 were deposited in the Gene Expression Omnibus (GEO) database (accession

8 numbers GSE155207). More detailed materials and methods can be found in

9 the supplementary methods.

11 Immunohistochemistry (IHC)

12 IHC for FH, CD8, CD4 and PD-L1 were performed using an automatic

13 staining platform, Ventana NexES (Roche, Basel, CH). Commercially available

14 primary anti-FH (1: 800, sc-100743, Santa Cruz, TX, USA), CD8 (1:100, clone

15 C8/144B, Dako, Copenhagen, DEN), CD4 (ZM-0418, 1:100, Zhongshan,

16 Beijing, CHN) and PD-L1 (740-4859, 1:40, Roche) were used in this study.

17 PD-L1 was determined by quantification of the density of positive cells (defined

18 as the number of positive tumor cells per mm2) and the percentage of tumor

19 cells. PD-L1 positivity was defined as PD-L1 expression ≥ 1%. CD4 and CD8

20 were determined by quantification of the density of positive cells defined as the

21 number of positive tumor infiltrating lymphocytes (TIL) per mm2 and the

22 percentage of TILs.

24 Multiple immunofluorescence

25 Multiplex staining and multispectral imaging were performed to identify the

26 different types of immune cells. Multiplex immunofluorescence staining was

27 obtained using PANO 7-plex kit (0004100100, Panovue, Beijing, CHN).

28 Different primary antibodies, including CD8 (clone C8/144B, Panovue), CD56

29 (clone 123C3, Panovue), HLA-DR (clone EPR3692, Panovue,), CD68

1 (Panovue, clone BP6036) and PanCK (0057000050, Panovue) were

2 sequentially applied, followed by horseradish peroxidase-conjugated

3 secondary antibody incubation and tyramide signal amplification. The slides

4 were microwave heat-treated after each Trichostatin A (TSA) operation. Nuclei

5 were stained with 4′-6′-diamidino-2-phenylindole (D9542, DAPI, Sigma,

6 Darmstadt, GER) after all the human antigens had been labeled. The stained

7 slides were scanned via the Mantra System to obtain multispectral images

8 (PerkinElmer, Massachusetts, USA). The scans were combined to build a

9 single stack image. All slides were scanned at an absolute magnification of ×

10 200 and × 100. Images of unstained and single-stained sections were used to

11 extract the spectrum of autofluorescence of tissues and each fluorescein,

12 respectively. The extracted images were further used to establish a spectral

13 library required for multispectral unmixing by Inform image analysis software

14 (PerkinElmer). The CD8, macrophage and NK cells expression were quantified

15 by the number of positive cells per mm2 and by the number of positive cells per

16 number of nucleated cells.

18 Sanger sequencing

19 The presence of somatic mutations of FH was validated by Sanger

20 sequencing. For this analysis, primers to amplify a 220-bp fragment covering

21 each exon (1-10) of FH gene were designed (Table S2). PCR was performed

22 with the following thermal cycling conditions: a 95°C for 5 min; 35 cycles of

23 95°C for 20 s; 60°C for 30 s and 72°C for 30 s; and a 72°C for 7 min. PCR

24 amplification of 10ng of genomic DNA was performed using the Premix Taq™

25 (RR901A, Takara, Shiga, JPN) on a Veriti96 PCR Cycler (Life Technologies,

26 Massachusetts, USA). PCR fragments were purified with ChargeSwitch™

27 PCR Clean-Up Kit (CS12000, Invitrogen, Oberhausen, GER), and the

28 sequencing reactions were performed on an ABI 3730XL automatic sequencer

29 (Life Technologies) according to the manufacturer’s instructions. All analyses

1 were performed in duplicate. Sequences of the forward and reverse strands

2 were analyzed using Chromas (Technelysium, AUS).

4 Bisulfite pyrosequencing

5 Bisulfite modification of genomic DNA isolated from 30 samples was

6 performed by standard methods. PCR primers were designed with the

7 PyroMark Assay Design V.2.0 software (QIAGEN) (Table S3). PCR products

8 were pyrosequenced with the PyromarkTM Q24 system (QIAGEN), according

9 to the manufacturer’s protocol.

11 Statistics

12 Fisher’s exact test was used to compare the difference between

13 categorical variables. Median survival was estimated using the Kaplan–Meier

14 method and the difference was tested using the log-rank test. All tests were

15 two-sided and performed using R software v.3.2.3, SPSS v.16.0 and

16 GraphPad V.6.0. A P-value < 0.05 was considered statistically significant.

18 Results

19 Patient characteristics

20 From 2014 to 2019, a total of 25 patients with FH-deficient RCC were

21 identified from 429 suspicious cases (Figure S1A). Among them, 16 (64%)

22 patients had germline FH mutations, eight (32%) patients had somatic FH

23 mutations, and one patient (FHRCC20) had a germline mutation (c.1256C>T)

24 and a second hit of an additional frameshift mutation (c.642delA) in the tumor.

25 Baseline characteristics of all patients are summarized in Table 1 and S4. The

26 median age at initial diagnosis was 38 years (range: 13-66 years) and the male:

27 female ratio was 1.8:1. Male patients tended to be older at onset of primary

28 RCC (P = 0.026, Figure S1B). At the end of follow-up (median: 14.3 months,

29 range 3.8-137.1 months), 21 patients (84%, 21/25) presented with metastases,

1 including 14 synchronous metastases and 7 metachronous metastases. The

2 most common metastases were lymph nodes (76%, 16/21) and bones (48%,

3 10/21). Five patients died by the end of follow-up. We found that those

4 diagnosed over the age of 40 years old had a higher incidence of metastasis

5 (P = 0.032, Figure S1C) and worse prognosis compared with the younger

6 patients (Figure S1D and E).

8 The somatic mutational landscape of FH-deficient RCC

9 WES was performed on 22 untreated primary tumors and matched

10 germline samples. A median of 31 somatic mutations per tumor (range, 1-156)

11 was identified, corresponding to an average of 0.58 mutations per megabase

12 (range, 0.02–2.60) (Figure 1A). A total of 27 FH mutations were identified,

13 including 15 missense mutations, 5 frame-shift indels, 5 nonsense mutations,

14 1 splice site change, and 1 large deletion (Figure 1B). Apart from FH, the most

15 frequently mutated genes in FH-deficient RCC were TTN (17%, 4/24) and NF2

16 (13%, 3/24). Assessment of mutations within pathways demonstrated a high

17 number of alterations involving chromatin modification (33%, 8/24),

18 RTK-RAS-PI3K (33%, 8/24) and DNA damage repair pathway (29%, 7/24).

19 Signature analysis identified that signature 6 contributed the most to these

20 base substitutions, followed by signature 22 and signature 7, indicating that the

21 defective DNA mismatch repair (signature 6), ultraviolet light exposure

22 (signature 7) and aristolochic acid exposure (signature 22) might contribute to

23 in the development of FH-deficient RCC (Figure S3A and B).

24 Somatic copy number alterations (SCNAs) analysis revealed that

25 FH-deficient RCC had frequent copy number loss of 1p, 8q, 10p, 15q and gain

26 of 4p, 7q, 11q, and 19p (Figure 1C). Of note, frequent deletion of genes

27 involving chromatin modification (KDM4B and KDM5A), SWI/SNF complex

28 (SMARCA4 and ARID1A), DNA demethylases (TET1) and DNA damage repair

29 pathway (ERCC1, ERCC2, XRCC1 and ATM) and cell cycle regulation

1 (CDKN2C and CDKN2D) were identified in this cohort (Figure 1C).

2 We further investigated the relationship between somatic mutations,

3 SCNAs, and clinical features. Compared with young patients (< 40 years old),

4 tumors of older patients (≥ 40 years old) tended to have a higher somatic

5 mutation burden (median mutations per megabase: 0.75 vs. 0.27, Figure 1A);

6 however, the difference did not reach a statistical significance (P = 0.051).

7 Patients older than 40 years old also had increased chromatin instability than

8 the younger patients (P = 0.047, Figure S3C and D). These results may reveal

9 why older patients have poorer outcomes.

11 DNA methylation alterations associated with the pathogenesis of

12 FH-deficient RCC

13 EPIC array was performed on 20 FH-deficient RCC tumors with adjacent

14 normal tissues. The study of the methylation profiles revealed that FH-deficient

15 RCCs had a sharp global hypermethylation compared to adjacent normal

16 tissues in all epigenomic structures, except for open sea (Figure S4A). In

17 contrast to control samples, we identified a total of 199,372 differentially

18 methylated CpG sites (CpGs) in FH-deficient RCCs, of which 66% (132,220)

19 were hypermethylated CpG sites, whereas only 34% (67,152) were

20 hypomethylated CpGs (Figure 2A, Figure S4B and C).

21 Gene ontology enrichment analysis revealed that hypermethylated CpGs

22 were significantly enriched in genes related to DNA-binding transcription

23 activity and sequence-specific DNA binding (Figure 2B), indicating the

24 transcriptional regulators might participate in the pathogenesis of FH-deficient

25 RCC. Moreover, we found part of hypermethylated CpGs located in multiple

26 well-established tumor suppressor genes (CDKN2A, APC, MGMT and TP53)

27 and small RNAs (MIR200a, MIR200b, MIR200c and MIR429) (Figure S4D

28 and E). Bisulfite pyrosequencing was used to validate DNA methylation levels

29 in four gene regions in three hypermethylated loci (CDKN2A, APC and MGMT)

1 and one hypomethylated locus (IL4R). A high correlation was observed

2 between values obtained from microarray and pyrosequencing (Figure S4F).

3 Furthermore, functional enrichment analysis identified an enrichment of genes

4 of hallmark pathways in cancer for aberrant methylation, including WNT,

5 mTOR, TGF-β, NOTCH and EMT pathways (Figure 2B and Figure S4G).

6 Hypermethylated CpGs also mapped to genes involving the DNA-damage

7 repair pathways (Figure S4H). Gene set enrichment analysis (GSEA) using

8 genome array data further revealed that genes linked to EMT, mTORC1,

9 DNA-damage repair and glycolysis pathways were upregulated in FH-mutated

10 RCCs versus normal tissues (Figure 2C).

12 Identification of the methylation patterns of FH-deficient RCC

13 To illuminate the DNA methylation pattern of FH-deficient RCC, we

14 merged data from our FH-deficient RCC and TCGA-KIRP cohorts. Using

15 unsupervised clustering analysis, we identified three DNA methylation clusters

16 (Figure 2D). The majority of our FH-deficient RCC cases (80%,16/20) and all

17 of the CIMP-RCC cases in KIRP cohort were categorized into the CIMP cluster,

18 exhibiting a global DNA hypermethylation phenotype. This finding supports

19 that most FH-deficient RCCs are largely characterized by a CIMP. However, a

20 small proportion of our FH-deficient RCC cases (20%,4/20) were classified into

21 cluster 1, manifested a relatively low genome-wide DNA methylation, indicating

22 that not all FH-deficient RCCs exhibited CIMP. To further validated these

23 findings, a T-distributed stochastic neighbor embedding analysis was

24 performed to integrated ours and all TCGA RCC subtypes data. The results

25 confirmed that the majority of FH-deficient RCCs were distributed in the

26 CIMP-RCC cluster, while the presence of two different DNA methylation

27 profiles existed in our FH-deficient RCCs (Figure 2E). Notably, we found

28 patients in the non-CIMP group had a lower incidence of metastasis compared

29 to the CIMP group (25% vs. 94%, P = 0.013). Three patients in the non-CIMP

1 group did not develop recurrences or metastases up to the end of follow-up

2 (median: 19.7 months), with one patient (FHRCC03) followed for over six

3 years after nephrectomy. Morphologically, the majority of CIMP tumors

4 (69%,11/16) showed a predominated papillary pattern, but only one non-CIMP

5 tumor (25%, 1/4) showed a dominated papillary pattern. Two non-CIMP tumors

6 (FHRCC03 and FHRCC12) were presented with a dominated low-grade

7 oncocytic pattern, which has been reported with a favorable outcome (2,6,7).

8 Methylation profiling analysis indicated that CIMP tumors had sharp global

9 hypermethylation compared to non-CIMP tumors in all epigenomic structures,

10 except for open seas (Figure S5A and B). Signature analysis further identified

11 a total of 234,563 different signatures between these two methylation groups

12 (Figure S5C).

14 FH-deficient RCC manifests high immune infiltration and imprints of

15 immune evasion

16 As mentioned above, a fraction of CpGs were hypomethylated in

17 FH-deficient RCC. Strikingly, these hypomethylated CpGs were enriched in

18 genes involving the inflammation pathway and cytokine activity (Figure 3A).

19 GSEA analysis of the genome array further validated the increased expression

20 of TNF-α-NF-κB and inflammatory response signaling in FH-mutated RCC,

21 indicating the enhanced anti-tumor inflammatory signature in FH-deficient

22 RCC (Figure 3A). Using immunohistochemistry (IHC) staining, we identified

23 that the majority of tumors (96%, 24/25) showed enriched CD8+ T and CD4+ T

24 cell infiltration. Sixty-four percent (16/25) of tumors had extensive CD8+ T cell

25 infiltration (CD8+ T cells ≥ 100 mm2) (Table S4 and Figure 3B). In situ multiple

26 immunofluorescence staining further revealed the increased immune-infiltrated

27 in FH-deficient RCC, with marked activated NK (CD56+) cells and

28 macrophages (CD68+/HLA-DR+) infiltrations in tumor stroma (Figure 3C).

29 Importantly, GSEA analysis confirmed that IFN-γ response and IFN-γ

1 dependent cancer immunostimulation pathways (IL-6–JAK-STAT3) were

2 significantly enriched in FH-mutated RCC, indicating the presence of activated

3 T cell signature. (Figure S6A).

4 Considering that the production of IFN-γ can induce PD-L1 expression

5 and subsequently leads to immune evasion (8), we next checked the

6 expression of PD-L1 in FH-deficient RCC. Intriguingly, we found that PD-L1

7 associated CpGs were hypomethylated modified in tumors (Figure S6B). In

8 addition, hypomethylated CpGs also mapped to other immune inhibitory

9 checkpoint genes (including PDCD1L2 and CTLA4, Figure 3D and Figure

10 S6B). IHC experiments further demonstrated a high PD-L1 positive rate (80%,

11 20/25) in tumors (Table S4). Of note, nearly half of tumors (48%,12/25)

12 exhibited type II tumor immune microenvironment (TIME) (9), characterized by

13 massive CD8+ T cell infiltration but positive PD-L1 expression in tumors

14 (Figure 3D). Collectively, our data demonstrated that the anti-tumor immunity

15 in FH-deficient RCC might be largely blocked by PD-1/PD-L1 pathway

16 mediated suppression mechanism, suggesting the potential of PD-1/PD-L1

17 inhibitors.

19 Patients with metastatic FH-deficient RCC benefit from immune

20 checkpoint blockade-based immunotherapy

21 At the end of follow-up, 18 patients with metastatic diseases received

22 systemic therapy. First-line therapy was tyrosine kinase inhibitor (TKI)

23 monotherapy (67%, 12/18), anti-PD-1 inhibitor plus TKI (27%, 5/18) and

24 anti-PD-1 inhibitor monotherapy (6%, 1/18). For those receiving first-line

25 therapy, the DCR was higher in patients receiving immune checkpoint

26 blockade (ICB)-based treatment (anti-PD-1 inhibitor + TKIs or anti-PD-1

27 inhibitor monotherapy) versus those on TKIs monotherapy (DCR: 100% vs.

28 58.3%, Figure 4A). Moreover, median first-line PFS was superior in the

29 ICB-based treatment group compared with that in TKIs group (median: 13.3 vs.

1 5.1 months, P = 0.024, Figure 4B). All patients (6/6, 100%) in the ICB-based

2 first-line treatment group achieved a PFS ≥ 6 months, higher than those (5/12,

3 42%) in the TKIs group (Figure 4C and a typical case was illustrated in Figure

4 4D). The median OS of systematic treatment was also longer in the ICB-based

5 group compared to the TKIs group (median: not reached vs. 21.4 months);

6 however, the differences did not reach a statistical significance (P = 0.209,

7 Figure 4B). Two patients received additional anti-PD-1 therapy plus TKIs after

8 disease progression during first-line TKIs monotherapy; remarkably, both of

9 them achieved favorable disease control (PFS: 13.5 and 10.3 months,

10 respectively).

12 Discussion

13 FH-deficient RCC is a rare but lethal malignancy, and its genetic basis is

14 poorly understood. In the present study, we performed comprehensive

15 molecular profiling to characterize the genomic foundation of FH-deficient RCC.

16 We determined that FH-deficient RCC was a genetically simple tumor with a

17 relatively low mutation burden, but its distinct epigenetic features could be the

18 biological foundation for tumor pathogenesis. Moreover, FH-deficient RCC was

19 an intrinsically immunogenic tumor that is likely to benefit from ICB-based

20 treatment strategies. Our study provides novel insights into the molecular

21 landscape of FH-deficient RCC and identifies promising therapeutic targets for

22 potential exploitation.

23 Although with low somatic mutation burden, we identified increased

24 SCNAs in FH-deficient RCC, with significant loss of 1p, 8q, 10p, 15q and gain

25 of 4p, 7q, 11q, and 19p. Previous studies identified that a subgroup of papillary

26 type 2 RCCs were characterized by a high degree of aneuploidy and frequent

27 loss of 9p (4), which was not significantly altered in our FH-deficient RCCs.

28 The different SCNAs patterns between FH-deficient RCC and papillary type 2

29 RCC indicates the distinct genomic alterations during tumorigenesis, although

1 the majority of FH-deficient RCC was predominantly of type 2 histology.

2 Recent studies also showed an increased chromosome 22 loss in CIMP-RCC

3 (10), however, it was not a significant event in our cohort. This discrepancy

4 may reflect the heterogeneity of CIMP-RCC and warrant further investigation.

5 Loss-of-function mutations in FH resulting in the accumulation of fumarate

6 impairs the function of α-KG-dependent dioxygenases, such as histone and

7 DNA demethylases, which disrupts the normal epigenetic genes regulation

8 (11). Consistent with this, we identified FH-deficient RCC-specific DNA

9 methylation changes within a broad panel of transcription factors, as well as

10 genes significantly associated with tumor progression and patient outcomes,

11 including CDKN2A, MGMT, APC and TP53. The loss of CDKN2A by either

12 mutation, deletion or promoter hypermethylation occurred in 15.8% RCCs and

13 correlated with decreased survival in all RCC histological subtypes (10). Of

14 note, we observed that CDKN2A promoter hypermethylation is a common

15 event in FH-deficient RCCs. Loss of CDKN2A can be targeted by inhibitors of

16 cyclin-dependent kinase 4/6 (CDK4/6) (12), suggesting the potential of

17 CDK4/6 inhibitors for FH-deficient RCCs. Also, it should be pointed out that it

18 remains unclear if the loss of CDKN2A could be a predictor for response to

19 CDK4/6 inhibitors, and still warrant further investigation(13). Currently,

20 epigenetic agents have been approved for the treatment of hematological

21 malignancies (14) and are now undergoing testing in HLRCC

22 (ClinicalTrials.gov identifier NCT03165721). Our data imply that these aberrant

23 DNA methylations could be the biological foundation of FH-deficient RCC,

24 highlighting the potential of demethylation agents for treating this disease.

25 Another important finding of our study is that FH-deficient RCC exhibits a

26 distinct immunophenotype. Albeit with low mutation burden, most FH-deficient

27 RCCs exhibited typical types II TIME, characterized by increased levels of

28 tumor infiltrating lymphocytes and activated chemotaxis signaling, but with

29 high expression level of PD-L1. Therefore, it is reasonable to assume that

1 anti-PD-1/PD-L1 therapy can be a potential treatment option for patients with

2 FH-deficient RCC (9,15). Indeed, we observed a favorable PFS for patients

3 receiving ICB-based treatment compared to those receiving antiangiogenic

4 monotherapy. Mechanistically, this immunogenicity may be mediated by

5 epigenomic and metabolic alterations. Using the TCGA data, Singh et al. found

6 that hypomethylation of genes was associated with immune function in

7 papillary RCC (16). Recent studies also show that tricarboxylic acid cycle

8 metabolites, including fumarate, can regulate methylation status in immune

9 cells and facilitate transcription and expression of proinflammatory factors

10 (17-20). Similarly, we observed epigenetically enhanced genes involved in

11 pro-inflammatory and immune suppression pathways in FH-deficient RCC.

12 Taken together, our data suggest that FH-deficient RCC is intrinsically

13 immunogenic and could be targeted by ICB-based immunotherapy. Given the

14 global hypermethylation phenotype, co-administration of epigenomic agents

15 could be a promising strategy to improve the efficacy of PD-1/PD-L1 inhibitors.

16 In addition to the treatments mentioned above, there are other potential

17 therapeutic targets according to our findings. Mutation of the Hippo pathway

18 tumor suppressor, NF2, was detected in several tumors. Thus, targeting NF2

19 (e.g. Dasatinib) (21) may be effective in a subset of patients. Moreover,

20 preclinical studies suggested that elevated fumarate could suppress the

21 homologous recombination DNA repair pathway through inhibition of

22 demethylases (22,23). We observed mutation and frequently aberrant

23 methylation of genes involved in DNA damage repair pathways in FH-deficient

24 RCC, which supports the poly (ADP)-ribose polymerase inhibitors for targeted

25 therapy. Additionally, both ours and previous studies showed the activation of

26 glycolysis (4,24,25) and mTOR signaling in FH-deficient RCCs. Results from a

27 single-center phase II trial (ClinicalTrials.gov Identifier: NCT01130519) and

28 retrospective studies (26,27) reported the combination of bevacizumab plus

29 erlotinib could achieve fair objective response for patients with FH-deficient

1 RCC, indicating the potential therapeutic value of this combination therapy.

2 In summary, the current study expands our knowledge of the FH-deficient

3 RCC somatic alteration landscape, emphasizes the importance of DNA

4 methylation profiles for tumor pathogenesis and molecular classification, and

5 reveals its immunogenic phenotype. Our study represents a step forward in

6 understanding the biology of FH-deficient RCC and provides a genetic basis

7 for biomarker classification and potential therapeutic strategies.

9 Acknowledgements

10 This work was supported by the Natural Science Foundation of China (NSFC

11 81672547, 81872107, 81872108, 81974398, 81972502 and 81902577), China

12 Postdoctoral Science Foundation (2020M673239) and 1.3.5 project for

13 disciplines of excellence, West China Hospital, Sichuan University

14 (No.0040205301E21). The authors thank all patients involved in this study, as

15 well as their families/caregivers. We offer sincere gratitude to Prof. Allen C.

16 Gao from UC Davis Medical Center for the constructive suggestions and

17 careful revision for this manuscript.

19 Author contributions

20 Study concept and design: S.G.X., X.M.Z., J.Y.L., X.Y.P., S.Z., Z.H.L., H.J.H,

21 J.Y., N.C., H.Z.;

22 Methodology, acquisition, analysis, or interpretation of the data: S.G.X., X.M.Z.,

23 J.Y.L., X.Y.P., S.Z., Z.H.L., J.H.C., W.L., B.H.L., T.H.L., R.H., X.X.Y., M.N.Z.,

24 L.N., P.F.S., J.G.Z., H.R.Z., J.D.D., Y.L.S., Z.P.L., J.Y.L., J.R.C., J.D.L., Z.P.W.,

25 X.D.Z., Y.C.N., D.Q., L.Y., Y.T.C., H.J.H, J.Y., N.C., H.Z.;

26 Financial support: S.G.X., R.H., Q.Z., N.C., H.Z.;

27 Drafting of the manuscript: S.G.X., X.M.Z., J.Y.L., X.Y.P., S.Z., Z.H.L.;

28 Critical revision of the manuscript for important intellectual content: S.G.X.,

29 X.M.Z., J.Y.L., X.Y.P., S.Z., Z.H.L., H.J.H, C.M. A., Z.H.L., J.Y., N.C., H.Z.;

1 Statistical analysis: S.G.X., X.M.Z., J.Y.L., X.Y.P., S.Z., Z.H.L., J.G.Z.

2 Study supervision: Q.W, X.L., Q.Z., H.J.H, J.Y., N.C., H.Z.

Reference: 1. Trpkov K, Hes O, Agaimy A, Bonert M, Martinek P, Magi-Galluzzi C, et al. Fumarate Hydratase-deficient Renal Cell Carcinoma Is Strongly Correlated With Fumarate Hydratase Mutation and Hereditary Leiomyomatosis and Renal Cell Carcinoma Syndrome. Am J Surg Pathol 2016;40(7):865-75 doi 10.1097/pas.0000000000000617. 2. Lau HD, Chan E, Fan AC, Kunder CA, Williamson SR, Zhou M, et al. A Clinicopathologic and Molecular Analysis of Fumarate Hydratase-deficient Renal Cell Carcinoma in 32 Patients. Am J Surg Pathol 2020;44(1):98-110 doi 10.1097/PAS.0000000000001372. 3. Grubb RL, 3rd, Franks ME, Toro J, Middelton L, Choyke L, Fowler S, et al. Hereditary leiomyomatosis and renal cell cancer: a syndrome associated with an aggressive form of inherited renal cancer. J Urol 2007;177(6):2074-9; discussion 9-80 doi 10.1016/j.juro.2007.01.155. 4. Cancer Genome Atlas Research N, Linehan WM, Spellman PT, Ricketts CJ, Creighton CJ, Fei SS, et al. Comprehensive Molecular Characterization of Papillary Renal-Cell Carcinoma. N Engl J Med 2016;374(2):135-45 doi 10.1056/NEJMoa1505917. 5. Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009;45(2):228-47 doi 10.1016/j.ejca.2008.10.026. 6. Smith SC, Sirohi D, Ohe C, McHugh JB, Hornick JL, Kalariya J, et al. A distinctive, low-grade oncocytic fumarate hydratase-deficient renal cell carcinoma, morphologically reminiscent of succinate dehydrogenase-deficient renal cell carcinoma. Histopathology 2017;71(1):42-52 doi 10.1111/his.13183. 7. Li Y, Reuter VE, Matoso A, Netto GJ, Epstein JI, Argani P. Re-evaluation of 33 'unclassified' eosinophilic renal cell carcinomas in young patients.

Histopathology 2018;72(4):588-600 doi 10.1111/his.13395. 8. Keenan TE, Burke KP, Van Allen EM. Genomic correlates of response to immune checkpoint blockade. Nat Med 2019;25(3):389-402 doi 10.1038/s41591-019-0382-x. 9. Sanmamed MF, Chen L. A Paradigm Shift in Cancer Immunotherapy: From Enhancement to Normalization. Cell 2018;175(2):313-26 doi 10.1016/j.cell.2018.09.035. 10. Ricketts CJ, De Cubas AA, Fan H, Smith CC, Lang M, Reznik E, et al. The Cancer Genome Atlas Comprehensive Molecular Characterization of Renal Cell Carcinoma. Cell Rep 2018;23(1):313-26 e5 doi 10.1016/j.celrep.2018.03.075. 11. Xiao M, Yang H, Xu W, Ma S, Lin H, Zhu H, et al. Inhibition of alpha-KG-dependent histone and DNA demethylases by fumarate and succinate that are accumulated in mutations of FH and SDH tumor suppressors. Genes Dev 2012;26(12):1326-38 doi 10.1101/gad.191056.112. 12. Rugo HS, Rumble RB, Macrae E, Barton DL, Connolly HK, Dickler MN, et al. Endocrine Therapy for Hormone Receptor-Positive Metastatic Breast Cancer: American Society of Clinical Oncology Guideline. J Clin Oncol 2016;34(25):3069-103 doi 10.1200/JCO.2016.67.1487. 13. Baghdadi TA, Halabi S, Garrett-Mayer E, Mangat PK, Ahn ER, Sahai V, et al. Palbociclib in Patients With Pancreatic and Biliary Cancer With CDKN2A Alterations: Results From the Targeted Agent and Profiling Utilization Registry Study. JCO Precision Oncology 2019(3):1-8 doi 10.1200/po.19.00124. 14. Fenaux P, Mufti GJ, Hellström-Lindberg E, Santini V, Gattermann N, Germing U, et al. Azacitidine Prolongs Overall Survival Compared With Conventional Care Regimens in Elderly Patients With Low Bone Marrow Blast Count Acute Myeloid Leukemia. Journal of Clinical

Oncology 2010;28(4):562-9 doi 10.1200/jco.2009.23.8329. 15. Doroshow DB, Sanmamed MF, Hastings K, Politi K, Rimm DL, Chen L, et al. Immunotherapy in Non-Small Cell Lung Cancer: Facts and Hopes. Clin Cancer Res 2019;25(15):4592-602 doi 10.1158/1078-0432.CCR-18-1538. 16. Singh NP, Vinod PK. Integrative analysis of DNA methylation and gene expression in papillary renal cell carcinoma. Mol Genet Genomics 2020;295(3):807-24 doi 10.1007/s00438-020-01664-y. 17. Arts RJW, Novakovic B, ter Horst R, Carvalho A, Bekkering S, Lachmandas E, et al. Glutaminolysis and Fumarate Accumulation Integrate Immunometabolic and Epigenetic Programs in Trained Immunity. Cell Metabolism 2016;24(6):807-19 doi 10.1016/j.cmet.2016.10.008. 18. Cordes T, Wallace M, Michelucci A, Divakaruni AS, Sapcariu SC, Sousa C, et al. Immunoresponsive Gene 1 and Itaconate Inhibit Succinate Dehydrogenase to Modulate Intracellular Succinate Levels. J Biol Chem 2016;291(27):14274-84 doi 10.1074/jbc.M115.685792. 19. Lampropoulou V, Sergushichev A, Bambouskova M, Nair S, Vincent Emma E, Loginicheva E, et al. Itaconate Links Inhibition of Succinate Dehydrogenase with Macrophage Metabolic Remodeling and Regulation of Inflammation. Cell Metabolism 2016;24(1):158-66 doi 10.1016/j.cmet.2016.06.004. 20. Netea MG, Domínguez-Andrés J, Barreiro LB, Chavakis T, Divangahi M, Fuchs E, et al. Defining trained immunity and its role in health and disease. Nature Reviews Immunology 2020 doi 10.1038/s41577-020-0285-6. 21. R J, G H. The two faces of Hippo: targeting the Hippo pathway for regenerative medicine and cancer treatment. Nature reviews Drug discovery 2014;13(1):63-79 doi 10.1038/nrd4161.

22. Sulkowski PL, Sundaram RK, Oeck S, Corso CD, Liu Y, Noorbakhsh S, et al. Krebs-cycle-deficient hereditary cancer syndromes are defined by defects in homologous-recombination DNA repair. Nat Genet 2018;50(8):1086-92 doi 10.1038/s41588-018-0170-4. 23. Sulkowski PL, Oeck S, Dow J, Economos NG, Mirfakhraie L, Liu Y, et al. Oncometabolites suppress DNA repair by disrupting local chromatin signalling. Nature 2020;582(7813):586-91 doi 10.1038/s41586-020-2363-0. 24. S S, C S, HS K, K B, VA VR, Y Y, et al. Fumarate hydratase deficiency in renal cancer induces glycolytic addiction and hypoxia-inducible transcription factor 1alpha stabilization by glucose-dependent generation of reactive oxygen species. Molecular and cellular biology 2009;29(15):4080-90 doi 10.1128/mcb.00483-09. 25. WH T, C S, G K, SY J, M V, M G, et al. The glycolytic shift in fumarate-hydratase-deficient kidney cancer lowers AMPK levels, increases anabolic propensities and lowers cellular iron levels. Cancer cell 2011;20(3):315-27 doi 10.1016/j.ccr.2011.07.018. 26. Choi Y, Keam B, Kim M, Yoon S, Kim D, Choi JG, et al. Bevacizumab Plus Erlotinib Combination Therapy for Advanced Hereditary Leiomyomatosis and Renal Cell Carcinoma-Associated Renal Cell Carcinoma: A Multicenter Retrospective Analysis in Korean Patients. Cancer Research and Treatment 2019;51(4):1549-56 doi 10.4143/crt.2019.086. 27. Park I, Shim YS, Go H, Hong BS, Lee JL. Long-term response of metastatic hereditary leiomyomatosis and renal cell carcinoma syndrome associated renal cell carcinoma to bevacizumab plus erlotinib after temsirolimus and axitinib treatment failures. BMC urology 2019;19(1):51 doi 10.1186/s12894-019-0484-2.

Figure legends Figure 1. Somatic alterations in FH-deficient RCC. (A) Oncoplot describes somatic genomic alterations of the FH-deficient RCC (n = 24). Recurrently mutated genes, which are grouped by molecular pathways, are shown on the left. Their mutation frequencies in patients with onset of RCC < 40 years and ≥ 40 years old are shown on the right. The top histogram shows the number of somatic mutations in each individual sample. The middle panel shows the clinicopathological features. The bottom histogram shows the mutation spectrum. Red arrow indicates two samples (FHRCC18 and FHRCC24) performing panel test. See also Table S1 and S4. (B) The types and relative positions of somatic mutations in the predicted FH protein. (C) Copy-number profile shows amplifications (red) and deletions (blue) in all

chromosomes for FH-deficient RCC. Cancer-related genes located within corresponding genomic regions are indicated. For genes located in 19 chromosome, the detailed locations are: AKT2 (19q13.2), ARID3A (19p13.3), CDKN2D (19p13.2), ERCC1 (19q13.32), ERCC2 (19q13.32), KDM4B (19p13.3), SMARCA4 (19p13.2), XRCC1 (19q13.31). FH = fumarate hydratase, RCC = renal cell carcinoma, IHC = immunohistochemistry.

Figure 2. FH-deficient RCCs manifest a distinct methylation phenotype. (A) A volcano plot comparing DNA methylation in tumors versus paired adjacent normal tissues. Significantly hypermethylated CpG sites are colored in red, whereas significantly hypomethylated CpG sites are colored in blue. q

values (-log10 p) were calculated using paired two-sided moderated Student’s t-test. (B) Gene ontology enrichment analysis of differentially methylated CpG sites.

Enrichment q values (-log10 p) are calculated by hypergeometric test. See also

Figure S4. (C) GSEA enrichment scores of glycolysis, EMT, DNA-damage repair and mTORC1 pathways in HLRCC versus normal tissues. (D) One-dimensional hierarchical clustering of the 2,000 most variant DNA methylation probes reveal three clusters of tumors from TCGA KIRP cohort (n = 275) and our FH-deficient RCC cohort (n = 20). In contrast to cluster 1 and 2, cluster 3 shows widespread DNA hypermethylation patterns characteristic of CIMP-associated tumors. Adjacent normal tissues (n = 12) are included in the heatmap but did not contribute to the unsupervised clustering. Each row represents a probe; each column represents a sample. (E) T-distributed stochastic neighbor embedding analysis of 842 TCGA RCC samples and 20 FH-deficient RCC samples performed on 2,000 reference probes illustrates three distinct clusters. FH = fumarate hydratase, RCC = renal cell carcinoma, GSEA = gene set enrichment analyses, EMT = epithelial-mesenchymal transition, TCGA = The Cancer Genome Atlas, HLRCC = hereditary leiomyomatosis and renal cell carcinoma, KICH = chromophobe renal cell carcinoma, KIRC = renal clear cell carcinoma, KIRP = renal papillary cell carcinoma, CIMP = CpG island methylator phenotype.

Figure 3. FH-deficient RCCs exhibit intrinsic immunogenic phenotype (A) Bubble chart (upper) shows the significantly hypomethylated CpGs in

FH-deficient RCCs. Enrichment q values (-log10 p) are calculated by hypergeometric test. GSEA enrichment scores (lower) of TNF-α-NF-κB and inflammation response signaling in HLRCC versus normal tissues. See also Figure S6. (B) Representative IHC demonstrating the typical TIME type II in CIMP (FHRCC13) and non-CIMP tumors (FHRCC12) in our cohort, is characterized by PD-L1 positive expression and massive CD8+ and CD4+T cell infiltrations.

TIME type II is defined as PD-L1 positive in tumors and massive CD8+ T cell infiltration (>100 cells/mm2). Magnification X200. Scale bar = 100μm. (C) Representative immunofluorescence demonstrating the presence of overall CD8+ T cell, tumor-associated macrophage (CD68 and HLA-DR) and NK cell (CD56) infiltration in two selected samples (FHRCC16 and FHRCC19) in our cohort. Magnification X400. Scale bar = 100μm. (D) Heatmap shows the immune features among FH-deficient RCCs (n = 25). The top panel shows the number of CD8+ and CD4+ T cell infiltrations, PD-L1 expression and TIME type. The middle panel shows the methylation status of a CpG site in the promoter region of immune modulation associated genes. The bottom histogram shows the FH mutation type, FH protein expression and CIMP status. FH = fumarate hydratase, RCC = renal cell carcinoma, HLRCC = hereditary leiomyomatosis and renal cell carcinoma, GSEA = gene set enrichment analyses, TIME = tumor immune microenvironment, CIMP = CpG island methylator phenotype.

Figure 4. Responses to systemic treatment in patients with metastatic FH-deficient RCC. (A) Histogram shows the best radiographic response in patients receiving first-line TKIs and ICB-based treatment groups. The best response is assessed via Response Evaluation Criteria in Solid Tumors 1.1. ICB-based treatments include patients receiving anti-PD-1 inhibitor plus TKIs or anti-PD-1 inhibitor monotherapy. (B) Kaplan-Meier curves show the first-line PFS (left) and OS (right) between patients receiving first-line TKIs (red) and ICB-based (green) treatments. (C) Swimmer plot depicts the PFS of individual patients receiving first-line systematic treatments. Each patient’s tumor response at the end of follow-up (PR, SD or PD) is noted. Vertical line indicates PFS at 6 months.

(D) Baseline imaging in a patient (FHRCC16) before initiation of systematic treatment, with multiple para-aorta lymphatic metastases. The patient subsequently received the combination of Sintilimab plus Axitinib treatment. Imaging performed at 3 months after the combination therapy demonstrates continued objective response (33.9% by RECIST v1.1). FH = fumarate hydratase, RCC = renal cell carcinoma, PR = partial response, SD = stable disease, PD = progression disease, ORR = objective response rate, PFS = progression-free survival, OS = overall survival, TKIs = tyrosine kinase inhibitors, ICB = immune checkpoint blockade.

Table 1. Baseline characteristics of patients with FH-deficient RCC

Patient Characteristics N(%)

Total number of patients 25(100) Age at diagnosis, years (Median, range) 38(13-66) Male 16(64) Female 9(36) Patients with skin leiomyoma 2(8) Female patients with uterine leiomyoma 5(55.6) Primary tumor size, cm (Median, range) 8.8(3-16.9) T stage <3a 16(64) ≥3a 9(36) N positive 12(48) M stage M0 4(16) Synchronous 7(28) Metachronous 14(56) Histopathological patterns Papillary 14(56) Tubular 4(16) Adenoid 3(12) Solid 2(8) Oncocytic 2(8) ISUP grade 2 1(4) 3 19(76) 4 3(12) Unknown 1(4) FH staining Imaging features(N=19) Solid 6(32) Single or less cavity 7(37) Multilocular 5(26) Unknown 6(32) Nephrectomy 20(80) Metastatic sites (N=21) Lymph node 16(76) Bone 10(48) Lung 4(19) Other 6(29) FH-deficient RCC = Fumarate Hydratase-deficient Renal Cell Carcinoma; ISUP = International Society of Urological Pathology.

Integrated Molecular Characterization of Fumarate Hydratase-deficient Renal Cell Carcinoma

Guangxi Sun, Xingming Zhang, Jiayu Liang, et al.

Clin Cancer Res Published OnlineFirst January 7, 2021.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-20-3788

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2021/01/07/1078-0432.CCR-20-3788.DC1

Author Author manuscripts have been peer reviewed and accepted for publication but have not yet Manuscript been edited.

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/early/2021/01/07/1078-0432.CCR-20-3788. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.