Metastatic clear-cell renal cell carcinoma: a frequent NOTCH1 mutation predictive of response to anti-NOTCH1 treatment

doi:10.21203/rs.3.rs-1627883/v1

Download PDF

Research Article

Metastatic clear-cell renal cell carcinoma: a frequent NOTCH1 mutation predictive of response to anti-NOTCH1 treatment

https://doi.org/10.21203/rs.3.rs-1627883/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Despite some improvement in the prognosis of metastatic clear-cell renal cell carcinoma (ccRCC), the identification of new therapeutic targets is essential. Up to now, only limited genomic data has been obtained from metastatic samples.

Methods

We aimed to characterize metastatic ccRCC by way of whole-genome analyses of metastatic formalin-fixed samples, using OncoScan® technology. We identified a frequent, unexpected pL1575P NOTCH1 mutation which we set out to characterize for translational purposes. We thus implemented patient-derived xenografts from metastatic samples of human ccRCC to explore its clinical significance .

Results

We showed that pL1575P NOTCH1 mutation was an activating mutation, leading to the expression of NOTCH1-intracellular domain-active fragments in both cancer cells and tumor endothelial cells, suggesting a trans-differentiation of cancer cells into tumor micro-vessels. We demonstrated that this mutation could be used as a predictive biomarker of response to CB-103, a specific NOTCH1-ICD inhibitor. One striking result was the considerable anti-angiogenic effect, coherent with the presence of NOTCH1 mutation in tumor micro-vessels.

Conclusions

We identified a frequent, unexpected pL1575P_c4724T_C NOTCH1 mutation as a new biomarker for ccRCC metastases, predictive of response to the CB103 NOTCH1-intracellular domain inhibitor. This opens the way for further clinical trials among patients with metastatic RCCs.

NOTCH1 mutation

metastatic clear-cell renal cell carcinoma

anti-NOTCH1 treatment

CB-103

biomarker

Clear-cell renal cell carcinomas (ccRCCs) are malignant tumors with high metastatic potential since more than 50% of patients develop metastases, either at initial diagnosis, or at the time of disease relapse [1]. Anti-angiogenic drugs, and more recently immunotherapies, have improved the prognosis of metastatic RCCs [2, 3]. However, almost all patients develop resistance to these treatments, with a median survival of up to 30 months [2, 3]. The resistance mechanisms remain largely to be deciphered, but intra-tumor heterogeneity could explain some of the clinical resistance observed. Intra-tumor heterogeneity of RCCs has been characterized at molecular level in primary tumors, with evidence of sub-clonal spatial heterogeneity [4, 5]. In addition, metastases derive from selected aggressive clones that have acquired resistance to first-line treatment with some gene abnormalities that expand from primary tumors to metastases [6]. These metastatic clones, which may be a minority in the primary tumor [7], are precisely those on which genomic analyses need to be carried out, to optimally guide targeted therapies for metastatic RCCs. To date, only limited genomic data has been obtained from metastatic samples. In a recent meta-analysis on RCC gene abnormalities, we reported data from 14,179 primary tumor samples, including 433 metastatic samples, with some striking differences between primary tumors and metastases [6].

Whole genome analyses are usually performed on formalin-fixed samples, which provide a considerable resource worldwide, with over one billion samples available [8]. In our meta-analysis, we showed that the sample conservation method (formalin-fixed or frozen samples) did not alter mutation prevalence for the most frequently mutated genes, including VHL, BAP1, PBRM1, and SETD2. However, formalin-fixation is associated with conformational DNA modifications due to a disulfide bond, DNA-protein crosslinks and DNA fragmentation of less than 100bp fragments [9], thus limiting detection of certain mutations depending on the technology used. In 2003, Oncoscan® FFPE Assay Kit (OncoScan assay) was developed to detect copy number variations, loss of heterozygosity (LOH), and cancer-related somatic mutations, using DNA extracted from formalin-fixed tissues. This assay uses probes with a genomic 40bp footprint [10–14]. Compared to PCR-based next-generation sequencing (NGS), the Affymetrix OncoScan® Array platform had greater sensitivity and specificity exceeding 98%, even using formalin-fixed samples of poor quality, and a lower allele variant frequency (5–10%) [12].

In this study, we aimed to characterize metastatic RCCs by way of whole-genome analyses of metastatic formalin-fixed samples. Using OncoScan® technology, we identified several unexpected alterations including a pL1575P NOTCH1 mutation and we set out to characterize this mutation more accurately for translational purposes.

Patient metastatic samples processed for OncoScan® analysis

Four patients were included in the first analysis. They all had a metastatic clear-cell renal cell carcinoma (ccRCC) with a biopsy of a metastatic site performed before any medical treatment. All four metastatic samples processed for genomic analyses were formalin-fixed paraffin-embedded biopsy samples.

In compliance with the French bioethics legislation (2004 − 800; June 8, 2004), all patients had been informed of the research use of part of their samples remaining after diagnosis, and none opposed it. Informed consent was obtained from each patient.

DNA purification and whole-genome analysis on human samples

Whole genome analysis was performed on laser-micro-dissected tumor cells from the four metastatic samples using OncoScan-Express® (Affymetrix, USA). OncoScan-Express® uses a molecular inversion probe (MIP)-based genotyping system dedicated to formalin-fixed paraffin-embedded tissues, which determines the genotype of 330,000 SNPs, copy number alterations, loss of heterozygosity (LOH) and somatic mutations (Supplementary Fig. 1). MIP probes are circularizable oligonucleotides, the two ends of which carry two sequences complementary to two sequences on the genome, separated by one nucleotide (where the variant to be genotyped is located). After hybridization of the genomic DNA, the product of the reaction is divided between two tubes, and two nucleotides are added to each tube (A/T and C/G). In the tube including the nucleotide complementary to the allele on the genome, the MIP probe is ligated and becomes circular. This structure is selected using exonucleases, and linearized. The products are amplified and hybridized onto an Affymetrix microarray for product identification. At the time we performed the analysis, OncoScan® Assay searched for 541 somatic mutations specific to cancer, with a coverage of over 200 tumor suppressors and oncogenes, and a median spacing of 1 probe per 0.5 kb for the top 10 “actionable” tumor suppressor genes, a median spacing of 1 probe per 2 kb for the top 190-plus actionable oncogenes, and a median backbone spacing of 1 probe per 9 kb.

Data analyses were performed on Nexus-7 software (BioDiscovery,USA).

ddPCR and validation of the NOTCH1 mutation in metastatic samples

A Droplet Digital Polymerase Chain Reaction (ddPCR) was performed using the QX100 ddPCR workflow system (Biorad, Hercules, CA, USA). The mix contained 20 ng of genomic DNA, 10 µL of ddPCR Supermix for Probes (No dUTP) (Bio Rad), 1µL of NOTCH1 probes either wild-type or specific to the pL1575P NOTCH1 mutation (12772, Qiagen, Germany) (Supplementary Word 1) and 1 µL RNaseP probe (Taqman® copy number Reference Assay, 4403326, Life Technologies) per well, and the final volume for the reaction was 20 µL. Droplets were generated by a QX200 Droplet Generator (Biorad). PCR was carried out on the CFX96 Real Time System (Bio Rad). It consisted in an initial denaturing step at 95°C for 10 min., followed by 40 denaturing cycles (94°C for 30 s), and annealing (60°C for 1 min.). A post-amplification melting curve program was initiated by heating to 98°C for 10 min. and then cooling to 12°C. Each PCR run included a no-template control. The results for ddPCR were generated using QX100 Droplet Reader (Biorad) and analysed using QuantaSoft software (Biorad). The ratio of NOTCH1-positive droplets to RnaseP-positive droplets was calculated.

We calculated the mutant allele frequency using the mathematical formula Mmu/ MDNAconc, as previously reported, Mmu being the number of mutant copies per droplet and MDNAconc being the DNA concentration in the reaction. Mmu and MDNAconc are calculated as follows:

Mmu = -ln (1-(nmu/n)), where nmu = number of droplets positive for mutant NOTCH1 probe and n = total number of droplets. MDNAconc = -ln (1-(nDNAcon/n)), where nDNAconc = number of droplets positive for wild-type NOTCH1 probe and/or mutant NOTCH1 probe and n = total number of droplets [15].

A DNA obtained from a patient with lymphoblastic acute leukaemia T and the presence of pL1575P NOTCH1 mutation was provided by P.V and used as the positive control.

Detection of NOTCH1-ICD-expressing cancer cells in metastases and tumor xenografts

Immunohistochemistry staining

An indirect immunoperoxidase method was performed on 5µm-thick tissue sections using anti-NOTCH1-ICD (ab8925,1:200, Abcam) as primary antibody.

The systematic controls used were the absence of primary antibody and the use of an irrelevant primary antibody of the same isotype.

Immunofluorescence staining

To detect NOTCH1-ICD in the endothelial cells, a double indirect immunofluorescence method was performed using NOTCH1-ICD and CD31 as primary antibodies. Staining was performed with Tyramide detection kit 488CF (biotium, 99824), and 543CF (biotium, 99825) respectively, and a fluorescent mounting medium with DAPI was used for nucleus detection (E19-18, GBI labs).

Western-blot analysis

Protein extraction was performed from cryopreserved tissue with RIPA buffer (Thermo) supplemented with 1 EDTA-free protease inhibitor cocktail tablet (Roche Diagnostics) and phosphatase inhibitors (Sigma-Aldrich).

Western blot was performed on 10% Mini-PROTEAN TGX precast gels (Biorad), then transferred onto a 0.2 µm Nitrocellulose membrane (Biorad) using the Trans-Blot Turbo Transfer System. Immunostaining was performed using anti-NOTCH1-ICD (ab8925, 1:500, Abcam) and anti-GAPDH (ab9485, 1:2500, Abcam) as primary antibodies. An anti-rabbit HRP (ab32568, 1:1000, Abcam) was used as a second antibody. Western blot revelation was performed using Clarity Western ECL Substrate (Biorad) and detected on ChemiDoc XRS + detection system (Biorad). Analyses were performed with Image Lab Software 6.1 (Biorad).

Laser micro-dissection of NOTCH1-ICD expressing cells

To confirm that NOCTH1-ICD expression in cancer cells or tumor endothelial cells was associated with NOTCH1 mutation, we combined laser micro-dissection of NOTCH1-ICD expressing cells with ddPCR for NOTCH1 mutation.

For each sample analysed, 7 µm-thick tissue sections were laser-micro-dissected to select a minimum of 300 cancer cells or 100 tumor endothelial cells expressing NOCTH1-ICD, using a PALM-Microbeam/Zeiss-system (Carl Zeiss, Germany). Total DNA was extracted from the micro-dissected cells using DNeasy-Micro-Kit (Qiagen, Courtaboeuf, France), and concentrated in a final volume of 10 µL.

For NOTCH1 gene copy number analyses, total DNA extracted from micro-dissected tumor cells was processed using ddPCR as described above.

Patient-derived clear-cell renal cell carcinoma xenografts and treatment

Five patient-derived clear-cell RCC xenograft models obtained from biopsies of metastases were implemented in our research unit [16] (Supplementary Table 1). The National Ethics Committee for experimental animal studies approved this study (APAFIS#17190-2018101814245111). The ethical guidelines that were followed met the standards required by the US guidelines [17]. Animals were maintained in pathogen-free housing at Sorbonne Paris Nord University (agreement number: C9300801).

After successful engraftment, tumor growth was measured in two perpendicular diameters with a caliper. Tumor volumes were calculated as follows: V = L × l2 ÷ 2, L being the larger diameter (length), l the smaller (width). When tumors reached a volume of 400 mm3 (n = 6 mice per treatment group), the mice were separated into four groups: i) treatment by gavage with sunitinib at 40 mg/kg/day (group 1), ii) treatment by gavage with NOTCH1 inhibitor (CB-103, HY-135145, MedChem Express at 25 mg/kg/twice daily or (LY411575, S2714, Selleckchem) at 10 mg/kg/day (group 2)); iii) combined treatment by gavage with sunitinib, at 40 mg/kg/day, and with NOTCH1 inhibitor (either CB-103 or LY411575) (group 3); iv) treatment with 100 µL of 0.9% NaCl as a control (group 4). All treatments were administered for 30 days. A daily clinical score was recorded and tumor growth was measured weekly until tumor weight reached the ethically recommended limit of under 10% of mouse body weight (Directive 2010/63/EU of European Parliament and Council of 22 September 2010 on the protection of animals used for scientific purposes; Official Journal of European Union L 276/33). An inhibition growth coefficient was calculated for each treatment arm (Supplementary Fig. 2), using a ratio of the slopes (a and a’) of the straight lines before and after treatment. In all xenograft models, the coefficient of inhibition for a drug is was calculated as (a’-a)/a - a, being the slope of the curve before the start of treatment (Day 0) - and a’ the slope of the curve between Day 0 and Day 30 of treatment. For a given drug combination, if the inhibition coefficient was negative, the tumor was considered sensitive to the drug. If, in contrast, it was positive, the tumor was considered resistant to this drug.

In situ assessment of necrosis, cell proliferation, angiogenesis and apoptosis

When present, necrosis was delineated on virtual slides created on a Nanozoomer2.0H scanner (Hamamatsu/ Japan), and quantified using DotSlide2 software. Results were expressed as the sum of necrotic areas for each section, and the mean ± SEM. For micro-vessel density, proliferation and apoptotic counts, an indirect immunoperoxidase method was performed on 5 µm-thick tissue sections, using monoclonal mouse anti-human Ki67 antibody (M724001-2, 1:100, Agilent) as primary antibody for proliferation, polyclonal rabbit anti-human cleaved-caspase-3 antibody (Asp175, 1:200, Cell signalling Technology) as primary antibody for apoptosis, and rabbit polyclonal anti-mouse CD31 antibody (SZ31, 1:30, Dianova) as primary antibody for micro-vessel density. For CD31 staining, the secondary antibody was a rabbit anti-rat IgG H&L (ab6703,1:200, Abcam), for Ki67 staining, a rabbit anti-mouse IgG1 H&L (clone M1gG51-4, 1:200, Abcam) was used. The secondary antibody was coupled with an anti-rabbit OmniMap detection kit (Roche diagnostic, Meylan, France).

For each tumor section analyzed, proliferation and apoptotic cell counts were performed on five different fields at x400 magnification, using a ProvisAX70 microscope (Olympus, Tokyo) with a wide-field eyepiece number 26.5 providing a field size of 0.344mm2 at X400 magnification. Microscope images were captured using a ColorView-III digital camera, and analyzed using Olympus-SIS Cell F software. The percentage of positive cells in 100 cancer cells was determined, and results were expressed as means ± SEM. For micro-vessel density, CD31-positive micro-vessels were counted on ten different fields, at X400 magnification. The number of positive micro-vessels was related to the total number of micro-vessels in a given surface area studied. Results were expressed as means ± standard deviation.

Statistical analysis

Statistical analyses were performed using R Studio 1.3.1073 statistical software.

For analysis of the copy number variation derived from the result from Oncoscan® technology and our meta-analysis data, we used the copynumber package in R.

For counts of NOTCH1-expressing tumor cells, the mean ± SEM was calculated for each tumor sample (primary RCC, metastasis or tumor-xenograft), and represented in bar graphs.

Quantitative values were compared using Student's t-test (two-tailed), while proportions were compared using the Z-test. P values under 0.05 were considered significant.

Whole-genome signature of the four RCC metastases using OncoScan® technology

Four patients were included in the first analysis (Supplementary Table 2).

Using OncoScan® technology, we identified a panel of copy number alterations and loss of heterozygosity (LOH) (Fig. 1A). We compared this to data obtained from the recent meta-analysis we performed on RCC genomic data (Fig. 1B) [6] and found common events known to occur early in the carcinogenesis of RCCs, such as whole 3p loss and whole 5q amplification. In contrast, we identified some abnormalities not previously described, including 9q11.2 and 15q11.1-11.2 amplification. In particular, the 9q11.2 amplification was identified in the four samples with LOH in the same locus, suggesting its potential role in the RCC metastatic process. Within these loci, very few genes have been studied for their potential role in RCC carcinogenesis (Supplementary Table 3). Among them, MiR1299, located on 9q11.2, encodes the acting circRNA and circ-EGLN3, which promotes renal cell carcinoma proliferation by regulating of IRF7 expression [18].

We also identified potential hotspot mutations in the four metastases analyzed (Supplementary File Excel 1). As recommended, using a threshold of 9 for probability score to filter the data, we retrieved 48 mutations (Table 1). Typically, we found VHL gene mutations for 3 of the 4 patients. We also identified frequent mutations activating the mTOR pathway, in PI3KCA and PTEN genes. The three other genes that are frequently mutated in RCCs, PBRM1, SETD2, and BAP1, were not included in this OncoScan® panel.

Table 1

Point mutation identified on RCC metastases before treatment
	Patient 1	Patient 2	Patient 3	Patient 4
ABL1_pF359V_c1075T_G	12	12	16	12
NPM1_pW288fs12_c863_864insCATG_allele1	13	9	9	11
NOTCH1_pL1575P_c4724T_C	11	21	0	9
MEN1_p_c654_plus_3A_G	3	20	10	11
VHL_pE160K_c478G_A	11	1	9	10
RET_pM918T_c2753T_C	28	0	1	1
PTEN_pI101T_c302T_C	17	5	14	11
PTEN_pE235X_c703G_T	14	12	6	5
PTEN_pR173C_c517C_T	19	-1	7	3
PIK3CA_pC420R_c1258T_C	16	-1	-1	-1
PIK3CA_pC901F_c2702G_T	15	6	4	6
PIK3CA_pH701P_c2102A_C	14	-1	1	-1
PIK3CA_pY1021C_c3062A_G	10	5	8	8
PDGFRA_pD1071N_c3211G_A	11	1	12	8
NF1_pR816X_c2446C_T	19	10	10	5
NF1_pK1444E_c4330A_G	19	-1	7	5
NF1_pR1276X_c3826C_T	16	1	8	4
NF1_pR461X_c1381C_T	12	5	12	5
NF2_pQ362X_c1084C_T	16	1	10	4
RB1_p_c2107_minus_2A_G	11	-1	13	9
RB1_pQ702X_c2104C_T	13	6	12	11
RB1_pR787X_c2359C_T	14	13	8	-1
RB1_p_c1961_minus_1G_A	8	9	12	8
RB1_pR251X_c751C_T	7	6	12	0
CDKN2A_p_c151_min_1G_A	7	11	7	5
BRCA1_pG1077W_c3229G_T	13	-1	1	14
BRCA2_pR2678S_c8034G_T	10	15	11	-1
BRCA2_pS1682S_c5046T_C	10	3	13	11
APC_pQ1367X_c4099C_T	15	0	14	10
APC_pR213X_c637C_T	15	-1	0	7
APC_pQ789X_c2365C_T	13	-1	3	-1
APC_pE853X_c2557G_T	4	5	12	4
NPM1_pW288fs12_c863_864insTCTG_allele1	19	4	6	4
NPM1_pW288fs12_c863_864insCCTG_allele1	11	4	7	6
ERBB2_pG776S_c2326G_A	12	6	5	10
TP53_pE336X_c1006G_T	10	11	8	6
TP53_pC135F_c404G_T	-1	11	0	-1
KRAS_pQ61K_c181C_A	9	14	0	8
NRAS_pQ61H_c183A_C	12	6	3	-1
PAK7_pT397K_c1190C_A	19	12	4	6
SMAD4_pD537Y_c1609G_T	16	11	6	4
RUNX1_pR166X_c496C_T	15	8	4	8
MLH1_pC233R_c697T_C	11	-1	-1	4
IRAK1_pS690G_c2068A_G	14	-1	-1	11
TSHR_pM453T_c1358T_C	13	8	6	1
CBL_pR420Q_c1259G_A	11	7	11	5
MSH2_pR711X_c2131C_T	3	11	9	-1
INPP4A_pE940D_c2820A_C	11	-1	2	8

In contrast, several mutations had never been described in RCC. We chose to focus on the pL1575P_c4724T_C NOTCH1 mutation, located on chromosome 9q34.3, because it was potentially present in at least 3 of the 4 patient metastases with a high probability score of 21 for Patient 2, and because the NOTCH pathway had been reported as a potential therapeutic target in clear-cell renal cancer [19–23]. In addition, the 9q arm is lost in 75% of RCCs as reported in our meta-analysis (Fig. 1B), and we found an allelic imbalance 9q34.3 cytoband for Patient 1 and Patient 2 (Supplementary File Excel 1).

pL1575P_c4724T_C NOTCH1 mutation is a frequent molecular event in RCC metastases

Using Sanger sequencing, we were not able to detect the pL1575P_c4724T_C NOTCH1 mutation in any of the 3 metastatic samples, neither in lymphoblastic acute leukemia T (LAL-T) positive control. In contrast, using digital droplet PCR and specific probes for the pL1575P_c4724T_C NOTCH1 mutation, we confirmed that it was present in the LAL-T positive control samples and in the 3 metastatic samples from Patients 1, 2 and 4 (Fig. 2A), but not in the sample from Patient 3. We showed that the absolute numbers of mutated NOTCH1 copies were 1516, 1221 and 1346 for Patients 1, 2 and 4, respectively. We also showed tumor heterogeneity with a mutant allele frequency of 61%, 65% and 59% respectively.

When we tested 9 metastatic biopsy samples from 7 additional patients with metastatic RCCs, we identified pL1575P NOTCH1 mutations in all metastatic samples, with a mean mutant allele frequency of 53.5%, and ranging from 43–80.5% (Fig. 2B).

pL1575P NOTCH1 mutation is an activating mutation in RCC

NOTCH1 is a trans-membrane receptor that belongs to the highly conserved NOTCH signalling pathway. Most NOTCH1 mutations occur in the hetero-dimerization (HD) and/or PEST domains [24] (Fig. 3A). The pL1575P NOTCH1 mutation is located in the N-terminal part of the HD domain (HD-N) and is responsible for NOTCH1 constitutive activation in lymphoblastic acute leukemia T [25]. Indeed, when this point mutation is present in lymphoblastic acute leukaemia T cells, it opens the enzymatic cleavage domain in S2, enabling NOTCH1 receptor cleavage by disintergrin and metalloprotease (ADAM) enzyme. This enzymatic cleavage leads to the dissociation of the extracellular domain and to the generation of a truncated form of NOTCH1, comprising the transmembrane and the intracellular domains. The third cleavage (S3) mediated by a γ-secretase protease complex releases the active NOTCH1 intracellular domain (NOTCH1- ICD) into the cytoplasm. After translocation into the nucleus, NOTCH1-ICD binds to ubiquitous transcription factors including mastermind-like protein (MAML, a transcriptional co-activator), and RNA-binding protein (RBP J) implicated in differentiation, proliferation, and cell survival (Fig. 3B).

To identify the intra-cellular domain of NOTCH1 (NOTCH1-ICD), we performed immunohistochemistry staining using an anti-NOTCH1 activated antibody (aa 1755-1767-intracellular) which specifically recognizes an ICD epitope that is only exposed after cleavage by gamma secretase. We found predominant nuclear staining with some cytoplasmic staining in the 3 metastatic samples from Patients 1, 2 and 4 (Fig. 4A). We also found that not all cancer cells were stained. When we separated cancer cells expressing NOTCH1-ICD from those that did not using laser-microdissection, the pL1575P NOTCH1 mutation was mainly present in cancer cells expressing NOTCH1-ICD (57% vs. 17%, P < 0.01, Fig. 4B).

Surprisingly, using peroxydase immunostaining, we found that tumor vessels and thus tumor endothelial cells also expressed NOTCH1-ICD (Fig. 4C). To confirm this observation, we performed double immunofluorescence staining using anti-CD31 and anti-NOTCH1-ICD antibodies. We showed that some CD31-expressing cells, but not all, co-expressed NOTCH1-ICD (Fig. 4C). Using laser-micro-dissection, we selected tumor endothelial cells expressing NOTCH1-ICD and identified the pL1575P NOTCH1 mutation with an allelic frequency of 48%, comparable to that found in cancer cells (Fig. 4C).

pL1575P NOTCH1 is frequently retained in RCC xenografts obtained from metastatic samples

Five patient-derived xenograft models obtained from RCC metastases were developed in our research unit (XRCC1 to XRCC5). Using allelic discrimination for the pL1575P_c4724T_C NOTCH1 mutation and ddPCR, we identified the mutation in all five RCC xenograft models at Passage 1 (P1), with allelic mutation frequencies ranging from 10.1 to 48.8%. Except for the XRCC5 model, the allelic frequency of the pL1575P_c4724T_C NOTCH1 mutation significantly increased between P1 and P5 in the other four models, suggesting progressive tumor enrichment with this mutation (Supplementary Fig. 3).

Targeting NOCH1-ICD inhibits tumor growth in vivo

For in vivo experiments using NOTCH1 inhibitors, we chose models XRCC4 and XRCC5 because of the marked enrichment for NOTCH1 mutation in the XRCC4 xenograft (48% allelic mutation frequency at passage 5) and a much lower allele mutation frequency for the XRCC5 xenograft (19% at passage 5).

These two RCC xenograft models were obtained from metastatic samples from two patients responding to sunitinib treatment in first-line setting (Supplementary Fig. 4), predicting response to sunitinib in the two models [26].

We treated these two models with two different NOTCH1 inhibitors, CB-103 and LY411575. They were chosen from a panel of 10 inhibitors commercially available because of their two different biological mechanisms: LY411575 is a γ-secretase inhibitor (GSI) with an IC50 of 0.39nM while CB-103 is a NOTCH1-ICD transcription complex inhibitor (Supplementary Table 4 and Fig. 3B).

Using LY411575 administered daily by gavage, we did not observe any anti-tumor effect (data not shown).

In contrast, using CB-103 mono-therapy, there was a significant anti-tumor effect for both xenograft models, more marked with XRCC4 than with XRCC5 (Fig. 5A and Table 2, Supplementary Fig. 5). Using sunitinib alone, we obtained complete tumor growth inhibition, associated with a significant increase in necrotic areas after histological analysis. Unexpectedly, there was also a strong induction of necrosis with CB-103 mono-therapy, with an additive effect when CB-103 and sunitinib were combined (Fig. 5B and Supplementary 4B). There was also a significant gradual decrease in micro-vessel density in treated mice. For model XRCC4, the number of CD31-expressing micro-vessels decreased from 28 for untreated mice to 7.4 for mice treated with a combination of CB-103 and sunitinib (P < 0.01, Fig. 5C). Surprisingly, there was a limited direct cytotoxic effect on cancer cells, with no difference between untreated and treated mice for apoptotic counts using cleaved-caspase 3 staining (data not shown). We showed a significant inhibition of proliferation using sunitinib combined with CB-103 compared to the untreated group (Fig. 5D). It is important to note that CB-103 tumor growth inhibition was stronger with XRCC4 than with XRCC5 (growth inhibition coefficient of -0.21 vs. 0.57 respectively), coherent with a higher pL1575P_c4724T_C NOTCH1 allelic mutation frequency in model XRCC4. Finally, when we assessed NOTCH1-ICD expression and NOCTH1 allelic mutation frequency in tumors after treatment, we found a significant decrease for both markers. In particular, for model XRCC4 treated with CB-103 monotherapy, NOTCH1 allelic frequency decreased from 64–1%, and NOTCH1-ICD was no longer seen to be expressed on Western blot (Fig. 5E, 5F).

Table 2

Growth inhibition coefficient for drugs tested in XRCC model
XRCC Model	Drug	Growth inhibition coefficient
	Untreated	xx
XRCC4	Sunitinib	-0.63
	CB103	-0.21
	Sunitinib + CB103	-0.65
XRCC5	Untreated	xx
	Sunitinib	-0.60
	CB103	0.57
	Sunitinib + CB103	-0.62

In this study, using two different methods for genomic analysis, we identified the pL1575P_c4724T_C NOTCH1 activating mutation in metastatic RCCs, a new biomarker predictive of response to NOTCH1-ICD inhibition.

The NOTCH1 pathway is frequently activated in metastatic cancers including renal cell carcinomas [27], and high NOTCH1 expression in primary RCCs is associated with an increased risk of metastatic relapse and decreased survival [23, 28]. In our study, we showed an increased allelic frequency of the pL1575P_c4724T_C NOTCH1 mutation after successive passages through RCC xenografts, suggesting enrichment by these aggressive clones, as observed in the metastatic process in patients.

Surprisingly, this pL1575P_c4724T_C NOTCH1 mutation, present in 50–70% of LAL-T [29, 30], had not hitherto been identified in RCCs. There could be several reasons for this : i) the sample type (metastasis vs. primary tumor); ii) the conservation method (frozen vs. formalin-fixed); or iii) the type of genomic analysis. We only analyzed metastatic samples, and this is a strength of our study, since metastatic clones can derive from minority clones in the primary tumor, and can thus be missed in primary tumor analyses. In our series of 11 metastases, this pL1575P_c4724T_C NOTCH1 mutation was systematically identified, using either Oncoscan® technology on FFPE samples or allelic discrimination on frozen samples. In a recent meta-analysis on genomic data from 14,696 RCCs, we reported a prevalence of 2% for NOTCH1 mutations, but in a single study and on a limited number of 60 metastatic samples [31], suggesting that NOTCH1 mutation prevalence could be largerly under-estimated. For the conservation method, formalin fixation induces alterations that could limit certain genomic analyses [32]. Oncoscan® technology is restricted to formalin-fixed samples because of its original circulation probes recognizing small DNA sequences of less than 100bp [9]. In one comparative study, Oncoscan® technology had a higher sensitivity than the PCR- based next-generation sequencing (NGS) using formalin-fixed samples [12]. In two clear-cell papillary RCCs, Oncoscan® enabled the identification of new copy number alterations. In our study, we also identified new alterations, typically the amplification of 15q11.1-11q.2, which comprises several genes (including UBE3A, SNRPN, SNHG14, PWRN1) and potential new therapeutic targets for metastatic RCCs [33–35]. UBE3A, also known as E6AP ubiquitin-protein ligase (E6AP), is an enzyme synthesized by the UBE3A gene, linked to the degradation of the p53 tumor suppressor [33]. In preclinical models of RCCs, Liguisticum wallichii, a Chinese medicinal herb, led to tumor cell proliferation inhibition via UBE3A expression down regulation [36].

As with LAL-T [37], we showed that the pL1575P_c4724T_C NOTCH1 mutation is an activating mutation in metastatic RCCs, leading to NOTCH1 cleavage and then NOTCH1-ICD nuclear overexpression (Fig. 3, [38]). The pL1575P_c4724T_C NOTCH1 mutation was mainly restricted to cancer cells expressing NOTCH1-ICD and selected through laser micro-dissection. We also demonstrated that the pL1575P_c4724T_C NOTCH1 mutation was present in tumor endothelial cells expressing NOTCH1-ICD, and this was unexpected. Indeed, in patient-derived xenografts, micro-vessels are usually of murine origin [39]. However, some tumor micro-vessels can derive from a trans-differentiation of cancer cells, a phenomenon called vascular mimicry, which we have previously described in RCCs [40, 41].

To determine the benefit of targeting the NOTCH1 pathway in metastatic clear-cell RCC, we treated RCC patient-derived xenografts obtained from metastases with two different NOTCH1 inhibitors. We have previously shown that patient-derived xenografts are relevant preclinical models to predict treatment response in patients [16, 42, 43]. NOTCH1 inhibitors have shown their benefit in preclinical models of NOTCH-driven tumors, and clinical trials are on-going in LAL-T, metastatic breast and pancreatic cancer [44–49]. Using LY411575, a γ secretase inhibitor (GSI), we did not observe any anti-tumor effect. This inhibitor may not be sufficient to prevent NOTCH1 receptor cleavage at the S3 site. Indeed, the pL1575P_c4724T_C NOTCH1 mutation leads to a conformational change in the receptor and its permanent cleavage in an active NOTCH1-ICD. In contrast to LY411575, using CB-103 designed to block GSI-insensitive dominant active forms of NOTCH1-ICD, we obtained a strong anti-tumor effect, particularly when this drug was combined with sunitinib. In vitro, CB-103 was able to inhibit the growth of LAL-T expressing NOTCH1-ICD, and of triple negative breast cancer cell lines, while GSIs were inactive [50]. CB-103 is currently being evaluated in a phase-I/II clinical trial for advanced solid tumors and haematological malignancies (https://clinicaltrials.gov/ct2/show/NCT03422679), but our study opens the way to further development in metastatic RCC.

In our study, using two opposed RCC patient-derived xenografts, we showed that the pL1575P_c4724T_C NOTCH1 mutation could be used as a predictive biomarker for response to CB-103 inhibitor: the higher the NOTCH1 allelic frequency, the greater the anti-tumor effect. And this was associated with a quasi-disappearance of mutated clones after treatment. One striking result was the considerable anti-angiogenic effect at tissue and cell level, coherent with the unexpected presence of NOTCH1 mutation in tumor micro-vessels, and this had not hitherto been reported in preclinical studies [50].

In our study, we identified the pL1575P_c4724T_C NOTCH1 mutation as a new biomarker for RCC metastases, predictive of response to the CB103 NOTCH1-intracellular domain inhibitor. It opens the way for further clinical trials among patients with metastatic RCCs.

ccRCC

clear-cell rencal cell carcinoma

LAL-T

lymphoblastic acute leukemia T

XRCC

xenograft renal cell carcinoma

FFPE

Formalin-Fixed Paraffin-Embedded

Availability of data and materials

All data in this study are included either in this article or in the supplementary information files.

Acknowledgments

We would like to thank Angela Swaine and Sarah Leyshon for the revision of the English language.

Funding

Bourse d’Excellence de l’Ambassade de France au Viet Nam (Grant number: 971522C). INSERM.

Authors' Contributions

Conception and design: A. Janin, G. Bousquet

Development of the methodology: T.O. Bui, M. El Bouchtaoui, G. Bousquet

Data acquisition (provided animals, acquired and managed patients, provided facilities, etc.): T.O. Bui, G. Gapihan, J. Paris, M. Soussan, M. Ziol, V. Glabeke, M. El Bouchtaoui, A. Janin, G. Bousquet

Data analysis and interpretation (e.g. statistical analysis, biostatistics, computational analysis): T.O. Bui, M. El Bouchtaoui, C. Leboeuf, J.-P. Feugeas, A. Janin, G. Bousquet

Writing, review: T.O. Bui, G. Bousquet.

Revision of the manuscript: All authors

Administrative, technical, or material support (i.e. reporting or organizing data, constructing databases): C. Leboeuf, A. Janin, G. Bousquet

Study supervision: G. Bousquet

Ethics declarations

Ethics approval and consent to participate

Five patient-derived clear-cell RCC xenograft models obtained from biopsies of metastases were implemented in our research unit. The National Ethics Committee for experimental animal studies approved this study (APAFIS#17190-2018101814245111). The ethical guidelines that were followed met the standards required by the US guideline. Animals were maintained in pathogen-free housing at Sorbonne Paris Nord University (agreement number: C9300801).

Consent for publication

All subjects have written informed consent.

Competing interests

The authors declare no conflict of interest.

Ballarin R, Spaggiari M, Cautero N, De Ruvo N, Montalti R, Longo C, et al. Pancreatic metastases from renal cell carcinoma: the state of the art. World J Gastroenterol. 2011;17:4747–56.
Motzer RJ, Tannir NM, McDermott DF, Arén Frontera O, Melichar B, Choueiri TK, et al. Nivolumab plus Ipilimumab versus Sunitinib in Advanced Renal-Cell Carcinoma. N Engl J Med. 2018;378:1277–90.
Rini BI, Plimack ER, Stus V, Gafanov R, Hawkins R, Nosov D, et al. Pembrolizumab plus Axitinib versus Sunitinib for Advanced Renal-Cell Carcinoma. N Engl J Med. 2019;380:1116–27.
Gerlinger M, Horswell S, Larkin J, Rowan AJ, Salm MP, Varela I, et al. Genomic architecture and evolution of clear cell renal cell carcinomas defined by multiregion sequencing. Nat Genet. 2014;46:225–33.
Gerlinger M, Rowan AJ, Horswell S, Math M, Larkin J, Endesfelder D, et al. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med. 2012;366:883–92.
Bui TO, Dao VT, Nguyen VT, Feugeas J-P, Pamoukdjian F, Bousquet G. Genomics of Clear-cell Renal Cell Carcinoma: A Systematic Review and Meta-analysis. European Urology. 2022;S0302283821022193.
Bousquet G, Bouchtaoui ME, Leboeuf C, Battistella M, Varna M, Ferreira I, et al. Tracking sub-clonal TP53 mutated tumor cells in human metastatic renal cell carcinoma. Oncotarget. 2015;6:19279–89.
Blow N. Tissue issues. Nature. 2007;448:959–60.
Hykin SM, Bi K, McGuire JA. Fixing Formalin: A Method to Recover Genomic-Scale DNA Sequence Data from Formalin-Fixed Museum Specimens Using High-Throughput Sequencing. Orlando L, editor. PLoS ONE. 2015;10:e0141579.
Hardenbol P, Banér J, Jain M, Nilsson M, Namsaraev EA, Karlin-Neumann GA, et al. Multiplexed genotyping with sequence-tagged molecular inversion probes. Nat Biotechnol. 2003;21:673–8.
Foster JM, Oumie A, Togneri FS, Vasques FR, Hau D, Taylor M, et al. Cross-laboratory validation of the OncoScan® FFPE Assay, a multiplex tool for whole genome tumour profiling. BMC Med Genomics. 2015;8:5.
Wood HM, Foster JM, Taylor M, Tinkler-Hundal E, Togneri FS, Wojtowicz P, et al. Comparing mutation calls in fixed tumour samples between the affymetrix OncoScan® array and PCR based next-generation sequencing. BMC Med Genomics. 2017;10:17.
Yu M-C, Lee C-W, Lee Y-S, Lian J-H, Tsai C-L, Liu Y-P, et al. Prediction of early-stage hepatocellular carcinoma using OncoScan chromosomal copy number aberration data. WJG. 2017;23:7818–29.
Jung H-S, Lefferts JA, Tsongalis GJ. Utilization of the oncoscan microarray assay in cancer diagnostics. Appl Cancer Res. 2017;37:1.
Hrebien S, O’Leary B, Beaney M, Schiavon G, Fribbens C, Bhambra A, et al. Reproducibility of Digital PCR Assays for Circulating Tumor DNA Analysis in Advanced Breast Cancer. Mazoyer S, editor. PLoS ONE. 2016;11:e0165023.
Bousquet G, El Bouchtaoui M, Sophie T, Leboeuf C, de Bazelaire C, Ratajczak P, et al. Targeting autophagic cancer stem-cells to reverse chemoresistance in human triple negative breast cancer. Oncotarget. 2017;8:35205–21.
U.S. Department of Health and Human Services National Institutes of Health Office of Laboratory Animal Welfare. Public Health Service Policy on Humane Care and Use of Laboratory Animals. NIH Publication revised; 2015.
Lin L, Cai J. Circular RNA circ-EGLN3 promotes renal cell carcinoma proliferation and aggressiveness via miR‐1299‐mediated IRF7 activation. J Cell Biochem. 2020;121:4377–85.
Zhuang Z, Lin J, Huang Y, Lin T, Zheng Z, Ma X. Notch 1 is a valuable therapeutic target against cell survival and proliferation in clear cell renal cell carcinoma. Oncology Letters. 2017;14:3437–44.
Bhagat TD, Zou Y, Huang S, Park J, Palmer MB, Hu C, et al. Notch Pathway Is Activated via Genetic and Epigenetic Alterations and Is a Therapeutic Target in Clear Cell Renal Cancer. J Biol Chem. 2017;292:837–46.
Wu K, Hu L, Hou J. Selective suppression of Notch1 inhibits proliferation of renal cell carcinoma cells through JNK/p38 pathway. Oncology Reports. 2016;35:2795–800.
Jędroszka D, Orzechowska M, Bednarek AK. Predictive values of Notch signalling in renal carcinoma. aoms. 2017;6:1249–54.
Ai Q, Ma X, Huang Q, Liu S, Shi T, Zhang C, et al. High-Level Expression of Notch1 Increased the Risk of Metastasis in T1 Stage Clear Cell Renal Cell Carcinoma. Guan X, editor. PLoS ONE. 2012;7:e35022.
Shah PA, Huang C, Li Q, Kazi SA, Byers LA, Wang J, et al. NOTCH1 Signaling in Head and Neck Squamous Cell Carcinoma. Cells. 2020;9:2677.
Malecki MJ, Sanchez-Irizarry C, Mitchell JL, Histen G, Xu ML, Aster JC, et al. Leukemia-Associated Mutations within the NOTCH1 Heterodimerization Domain Fall into at Least Two Distinct Mechanistic Classes. Mol Cell Biol. 2006;26:4642–51.
Varna M, Gapihan G, Feugeas J-P, Ratajczak P, Tan S, Ferreira I, et al. Stem Cells Increase in Numbers in Perinecrotic Areas in Human Renal Cancer. Clin Cancer Res. 2015;21:916–24.
Zhu H, Zhou X, Redfield S, He Z, Lewin J, Miele L. Elevated Expression of Notch1 Is Associated With Metastasis of Human Malignancies. Int J Surg Pathol. 2013;21:449–54.
Liu S, Ma X, Ai Q, Huang Q, Shi T, Zhu M, et al. NOTCH1 functions as an oncogene by regulating the PTEN/PI3K/AKT pathway in clear cell renal cell carcinoma. Urologic Oncology: Seminars and Original Investigations. 2013;31:938–48.
Weng AP, Ferrando AA, Lee W, Morris JP, Silverman LB, Sanchez-Irizarry C, et al. Activating mutations of NOTCH1 in human T cell acute lymphoblastic leukemia. Science. 2004;306:269–71.
Chiang MY, Xu L, Shestova O, Histen G, L’heureux S, Romany C, et al. Leukemia-associated NOTCH1 alleles are weak tumor initiators but accelerate K-ras-initiated leukemia. J Clin Invest. 2008;118:3181–94.
Becerra MF, Reznik E, Redzematovic A, Tennenbaum DM, Kashan M, Ghanaat M, et al. Comparative Genomic Profiling of Matched Primary and Metastatic Tumors in Renal Cell Carcinoma. European Urology Focus. 2018;4:986–94.
Srinivasan M, Sedmak D, Jewell S. Effect of fixatives and tissue processing on the content and integrity of nucleic acids. Am J Pathol. 2002;161:1961–71.
Beaudenon S, Huibregtse JM. HPV E6, E6AP and cervical cancer. BMC Biochem. 2008;9 Suppl 1:S4.
Luo Q, Cui M, Deng Q, Liu J. Comprehensive analysis of differentially expressed profiles and reconstruction of a competing endogenous RNA network in papillary renal cell carcinoma. Mol Med Rep. 2019;19:4685–96.
Liu G, Ye Z, Zhao X, Ji Z. SP1-induced up-regulation of lncRNA SNHG14 as a ceRNA promotes migration and invasion of clear cell renal cell carcinoma by regulating N-WASP. Am J Cancer Res. 2017;7:2515–25.
Wu JS, Zhang PM, Sun LJ, Liu XZ. Liguisticum wallichii inhibits renal carcinoma progression by downregulating UBE3A and through suppression of NF-κB signaling. Genet Mol Res [Internet]. 2016 [cited 2021 Oct 22];15. Available from: http://www.funpecrp.com.br/gmr/year2016/vol15-4/pdf/gmr-15-gmr15049023.pdf
Lin C, Zheng H, Wang C, Yang L, Chen S, Li B, et al. Mutations increased overexpression of Notch1 in T-cell acute lymphoblastic leukemia. Cancer Cell Int. 2012;12:13.
Sanchez-Irizarry C, Carpenter AC, Weng AP, Pear WS, Aster JC, Blacklow SC. Notch subunit heterodimerization and prevention of ligand-independent proteolytic activation depend, respectively, on a novel domain and the LNR repeats. Mol Cell Biol. 2004;24:9265–73.
Hylander BL, Punt N, Tang H, Hillman J, Vaughan M, Bshara W, et al. Origin of the vasculature supporting growth of primary patient tumor xenografts. J Transl Med. 2013;11:110.
Serova M, Tijeras-Raballand A, Dos Santos C, Martinet M, Neuzillet C, Lopez A, et al. Everolimus affects vasculogenic mimicry in renal carcinoma resistant to sunitinib. Oncotarget. 2016;7:38467–86.
Angara K, Borin TF, Arbab AS. Vascular Mimicry: A Novel Neovascularization Mechanism Driving Anti-Angiogenic Therapy (AAT) Resistance in Glioblastoma. Transl Oncol. 2017;10:650–60.
Bousquet G, Feugeas J-P, Ferreira I, Vercellino L, Jourdan N, Bertheau P, et al. Individual xenograft as a personalized therapeutic resort for women with metastatic triple-negative breast carcinoma. Breast Cancer Res. 2014;16:401.
Bousquet G, Janin A. Patient-Derived Xenograft: An Adjuvant Technology for the Treatment of Metastatic Disease. Pathobiology. 2016;83:170–6.
Hu ZI, Bendell JC, Bullock A, LoConte NK, Hatoum H, Ritch P, et al. A randomized phase II trial of nab-paclitaxel and gemcitabine with tarextumab or placebo in patients with untreated metastatic pancreatic cancer. Cancer Med. 2019;8:5148–57.
Minuzzo S, Agnusdei V, Pusceddu I, Pinazza M, Moserle L, Masiero M, et al. DLL4 regulates NOTCH signaling and growth of T acute lymphoblastic leukemia cells in NOD/SCID mice. Carcinogenesis. 2015;36:115–21.
Yen W-C, Fischer MM, Axelrod F, Bond C, Cain J, Cancilla B, et al. Targeting Notch signaling with a Notch2/Notch3 antagonist (tarextumab) inhibits tumor growth and decreases tumor-initiating cell frequency. Clin Cancer Res. 2015;21:2084–95.
Han J, Shen Q. Targeting γ-secretase in breast cancer. Breast Cancer (Dove Med Press). 2012;4:83–90.
Massard C, Azaro A, Soria J-C, Lassen U, Le Tourneau C, Sarker D, et al. First-in-human study of LY3039478, an oral Notch signaling inhibitor in advanced or metastatic cancer. Ann Oncol. 2018;29:1911–7.
Zheng R, Li M, Wang S, Liu Y. Advances of target therapy on NOTCH1 signaling pathway in T-cell acute lymphoblastic leukemia. Exp Hematol Oncol. 2020;9:31.
Lehal R, Zaric J, Vigolo M, Urech C, Frismantas V, Zangger N, et al. Pharmacological disruption of the Notch transcription factor complex. Proc Natl Acad Sci USA. 2020;117:16292–301.

No competing interests reported.

SupplementaryFigure1.png
Oncoscan® technology and the Molecular Inversion Probe: Target Generation and Hybridization Procedures. a) Annealing: Probe and gDNA hybridization; b) Gap filling with A/T or G/C nucleotides; c) Exonuclease selection for gap filled probes; d) Cleavage at site 1 for probe opening and inversion; e) Probe amplification and biotinylation; f) Cleavage at site 2 to release the tag sequence; f) Array hybridization followed by staining with phycoerythrin through the biotin-streptavidin interaction; g) Array scanning. Blue and gray colors indicate presence and absence of the phycoerythrin fluorescence signal respectively.
SupplementaryFigure2.png
Tumor growth inhibition coefficient in XRCC models. For a drug or a drug combination, the coefficient of inhibition is calculated as (a’-a)/a, a being the slope of the curve before the start of treatment (Day 0), and a’ the slope of the curve between Day 0 and Day 30 of treatment. If this growth inhibition coefficient is found to be less than 0, the tumor is considered sensitive to the drug administered; if it is above 0, the tumor is considered resistant to the drug.
SupplementaryFigure3.png
ddPCR allelic discrimination for the pL1575P_c4724T_C NOTCH1 mutation in 5 XRCC tumor xenografts at the first (in gray) and the fifth (in black) passages.
SupplementaryFigure4.png
Response to sunitinib treatment in first-line setting for two patients who provided metastatic samples for xenograft in murine models. (A) Patient corresponding to the XRCC4 model. (B) Patient corresponding to the XRCC5 model.
SupplementaryFigure5.png
In vivo anti-tumor effect of CB-103 in the XRCC5 xenograft model. (A) CB-103 monotherapy, sunitinib monotherapy, or the combination of CB-103 and sunitinib significantly inhibit tumor growth after 30 days of treatment (n = 6 per treatment group). This is associated with a significant gradual increase in the percentage of necrotic areas (B), and significant decrease in microvessel density (C). For cell proliferation, the decrease is only significant with the combination of CB-103 and sunitinib compared to untreated mice (C). After 30 days of treatment with CB-103 monotherapy, the pL1575P_c4724T_C NOTCH1 mutation (D) has virtually disappeared. (*P < 0.05, ***P < 0.001).
SupplementaryFileExcel1.xlsx
SupplementaryTables.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Metastatic clear-cell renal cell carcinoma: a frequent NOTCH1 mutation predictive of response to anti-NOTCH1 treatment

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Background

Materials And Methods

Patient metastatic samples processed for OncoScan® analysis

DNA purification and whole-genome analysis on human samples

Detection of NOTCH1-ICD-expressing cancer cells in metastases and tumor xenografts

Immunohistochemistry staining

Immunofluorescence staining

Western-blot analysis

Laser micro-dissection of NOTCH1-ICD expressing cells

Patient-derived clear-cell renal cell carcinoma xenografts and treatment

Statistical analysis

Results

Whole-genome signature of the four RCC metastases using OncoScan® technology

Discussion

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1