Investigation on the Genomic Characterization of Uterine Sarcoma for rAd- p53 Combined with Chemotherapy Treatment

Abstract

The aim is to investigate the genomic characterization of uterine sarcoma for rAd-p53 (Gendicine^®) combined with chemotherapy treatment. We recently published an article on 12 cases of uterine sarcomas, which were treated with rAd-p53 combined with chemotherapy. We found that rAd-p53 combined with chemotherapy is effective for various uterine sarcomas. Pretreatment pathological specimens of four uterine sarcoma patients were collected from the above recent clinical research and numbered 1–4A/B. Tumor samples were subjected to targeted sequencing by using a 416 genes panel. We profiled the mutation spectrum and tumor mutation burden in the tumors, identified mutated genes, and explored their gene function. We also verified the p53 protein expression using immunohistochemistry. We identified a total of 30 mutated genes that were found from the next-generation sequencing test results. The average number of mutated genes was up to seven in the five samples. TP53 gene was mutated in two of the four patients, No. 1 and No. 4B. They are c.C833G (p.P278R) missense mutation and a point mutation (C141*) that result in a premature stop codon. We did not find a mutated TP53 gene in the other two cases, but we identified mutated genes, including CREBBP, LYN, CDKN2A, and JAK2, which were located upstream of the TP53 gene; they may have an impact on TP53. We also identified 11 additional genes which are involved in p53-related signaling pathways or have interaction with p53. Compared to solid tumor mutational burden (TMB) distribution, none of their TMB was ranking in the top 25%. Mutant p53 protein expression was positive in two specimens. Our results demonstrated that the TP53 signaling pathway plays an important role in uterine sarcoma tumorigenesis. TP53 and the upstream genes such as CREBBP, LYN, CDKN2A, and JAK2 may be involved in the genomic characterization for rAd-p53 (Gendicine) combined with chemotherapy in uterine sarcoma. Besides, the average amount of mutated genes from every patient is large.

Introduction

The essence of cancer is the process of accumulation of a series of genetic mutations. During this process, the random combination of mutated genes results in a serious heterogeneity of the tumor. Heterogeneity is not only observed between people but is also temporal and spatial. Even if from the same person, the composition of tumor cells may still be different at different times or in different parts. This is the root cause of the difficulty of tumor treatment, and another reason is that the overall efficiency of tumor-targeted therapy without markers in the past was ∼30%. Iyer et al.¹ found that the selection of sensitive patients based on genomic characteristics for targeted anticancer therapy can improve the treatment efficacy of lung adenocarcinoma, colorectal cancer, breast cancer, and melanoma. Kris et al. ² found that in patients with lung cancer, the treatment efficacy of patients who have clear genetic mutations and were treated with available targeted drugs is significantly better than that of patients without mutations or targeted drug therapy. Therefore, it is the trend of an accurate tumor therapy to select the target gene by genomic sequencing and corresponding targeted drug therapy. Large-scale sequencing studies such as the International Cancer Genome Consortium (ICGC) and The Cancer Genome Atlas (TCGA) have begun to catalog the spectra of somatic mutations present in different solid tumor types.³ However, large-scale sequencing data are not enough as longitudinal data on therapy and outcome. Such data are essential for better target conventional cytotoxics, as well as emerging drugs such as immunotherapeutics.

Using molecular markers could improve the benefit for cancer patients undergoing some cancer treatments. Clinical trial results^4,5 showed that rAd-p53 is effective against a variety of malignancies, including colon, glioma, lung, ovarian, and head and neck tumors.^6
–8 We published an article on the clinical treatment of rAd-p53,⁹ evaluated the efficacy of rAd-p53 (Gendicine^®) followed by chemotherapy in the treatment of uterine sarcoma. But not all tumor patients could get a good effect from rAd-p53. So to bring precise treatment for more patients, a predictive biomarker is needed to indicate which patient is suitable for rAd-p53 treatment. So this article intended to explore the genomic characterization of uterine sarcoma for rAd-p53 (Gendicine) combined with chemotherapy treatment, next-generation gene sequencing, and immunohistochemical analysis of the tissue samples available before treatment, and the results are analyzed as provided in the Results section.

Materials and Methods

Patients and tumor samples

Previous study

In 2018, we published an article on 12 cases of uterine sarcomas, 11 of which had recurred cancer and 1 had primary cancer. They were all treated with rAd-p53 combined with chemotherapy. The ages of the 12 patients ranged from 43 to 62 years, with an average of 52.25 ± 2.01 years. The three pathological types of LMS (leiomyosarcoma), CS (carcinosarcoma), and ESS (endometrial stromal sarcoma) had four cases each. There was one patient at stage I, five at stage III, one at stage IV, and five at an unknown stage. The range of progression-free survival (PFS) time between the initial surgery and diagnosis of recurrence was defined as PFS Time 1 (PFS1) was from 1 to 18 months, with a median of 3 months. They received rAd-p53 (Gendicine) combined with chemotherapy treatment. Regimen: (1) tumor-localized therapy: ultrasound-mediated rAd-p53 followed by the multipoint injection into the tumor. Intraperitoneal perfusion was performed instead if the tumor is not puncturable. For liver lesion, hepatic artery perfusion of rAd-p53 was performed. (2) Upon 72 h at the peak of rAd-p53 transfection, chemotherapy was followed by intravenous cisplatin, ifosfamide, and epirubicin regimen: Cisplatin (P) (60–70 mg/m², day 1), Ifosfamide (I) (1.5 g/m², day 1–3), and Epirubicin (E) (60 mg/m², day 1). At the same time, 15–30 mg bleomycin was injected locally. (3) A course of treatment was 3 weeks, and a total of two to five courses were applied.

Results show that the treatment resulted in one complete remission (CR), seven partial remission (PR), three stable disease (SD), and one progressive disease (PD), respectively. The remission (CR + PR) rate was 66.7%, and the responsive (CR + PR + SD) rate was 91.7%. The PFS time between the baseline period and the next recurrence was defined as PFS2, which ranged from 2 to 62 months, with a median of 13 months, which is 10 months longer than that of PFS1 and statistically significant (p = 0.0038, <0.05). The overall survival time OS ranged from 6 to 62 months, with a median of 24 months. After the combined treatment, four of the patients got the operation opportunity to debulk again. Of the two patients with liver metastases, one had complete remission of liver foci and one had partial remission. In conclusion, rAd-p53 combined with chemotherapy is effective for various uterine sarcomas.

Tumor samples

Pretreatment pathological specimens from 4 patients collected from the above 12 patients were numbered 1–4 (all pathological paraffin were from the Department of Pathology of Shengjing Hospital of China Medical University). Patient No. 4 had two surgeries before any rAd-p53 treatment. The initial surgery pathological sample was labeled as No. 4-A, and the second pathological sample was labeled as No. 4-B. The combined treatment for the four patients all resulted in CR. The PFS1 of the four patients between initial surgery or last surgery before rAd-p53 treatment and recurrence (baseline) was 2, 2, 3, and 6 months, respectively. PFS2 after rAd-p53 treatment was 39, 9, 7, and 10 months, respectively. Until the follow-up deadline, the disease of patient No. 1 was stable with no progression. The information is summarized in Table 1. There was no control group in this study because uterine sarcoma is a rare malignant tumor of the female reproductive organ, so we have no chance to get any control group.

Table 1.

Information of basic clinical and pathological tissue specimens of four patients

Number	Age	Pathological Type	Efficacy Evaluation	Pathology Specimen Number	Specimen Location	Surgery	PFS1 (Month)	PFS2 (Month)
1	45	ESS	PR	No. 1	Bladder (metastasis)	Recurrence biopsy	2	39+
2	62	LMS	PR	No. 2	Uterus	Initial surgery	2	9
3	43	ESS	PR	No. 3	Pelvis	Initial surgery	3	7
4	61	UUS	PR	No. 4-A	Uterus	Initial surgery	6	10
4	61	UUS	PR	No. 4-B	Vaginal (relapse)	Recurrence surgery	6	10

ESS, endometrial stromal sarcoma; LMS, uterine leiomyosarcoma; PR, partial remission; UUS, undifferentiated sarcoma of the uterus; +, no progress.

Evaluation for gene mutations by next-generation sequencing and gene analysis

The samples mentioned above were sequenced by next-generation sequencing (NGS). The sequencing technology involves processes such as sample pretreatment, DNA extraction and processing, DNA library construction, probe enrichment, and high-throughput sequencing. It was performed by Nanjing GENESEEQ Biotechnology, Inc. The instrument used was Illumina Hiseq4000, and full exons and partial introns of 416 genes were sequenced.

DNA extraction and library construction

DNA was extracted by the QIAamp DNA FFPE Tissue Kit with modified protocols. The purified DNA is quantified by a Picogreen fluorescence assay using the provided lambda DNA standards (Invitrogen). Then, library construction with the KAPA Hyper DNA Library Prep Kit, containing mixes for end-repair, dA addition, and ligation, was performed in 96-well plates (Eppendorf). Dual-indexed sequencing libraries are PCR amplified for four to seven cycles.

Hybrid selection and ultradeep NGS of DNA

The PCR master mix is added to directly amplify (six to eight cycles) the captured library from the washed beads. After amplification, the samples are purified by AMPure XP beads, quantified by quantitative PCR (qPCR) (Kapa), and sized on bioanalyzer 2100 (Agilent). Libraries are normalized to 2.5 nM and pooled. Deep sequencing is performed on Illumina HiSeq 4000 using the PE75 V1 Kit. Cluster generation and sequencing are performed according to the manufacturer's protocol.

Sequence alignment and processing

Base calling was performed using bcl2fastq v2.16.0.10 (Illumina, Inc.) to generate sequence reads in FASTQ format (Illumina 1.8+ encoding). Quality control was applied with Trimmomatic.¹⁰ High-quality reads were mapped to the human genome (hg19; GRCh37 Genome Reference Consortium Human Reference 37) using modified BWA aligner 0.7.12¹¹ with Burrows-Wheeler Aligner-MEM algorithm and default parameters to create SAM files. Picard 1.119 was used to convert SAM files to compressed BAM files, which were then sorted according to chromosome coordinates. The Genome Analysis Toolkit¹² (GATK; version 3.4-0) was modified and used to locally realign the BAMs files at intervals with indel mismatches and recalibrate base quality scores of reads in BAM files.¹³

Sequencing results of 416 genes are presented in Supplementary Table S1. After gene mutation information was obtained, gene function and its relativity to the TP53 gene were analyzed in the COSMIC tumor database (Catalog of somatic mutations in cancer), GeneCards database, and STRING database.

Tumor mutational burden

Tumor mutational burden (TMB) is the average number of nonsynonymous somatic mutations in the 1 Mb (1 million bases) tumor genome with the unit of mutation/Mb.¹⁴ Nonsynonymous mutations in TMB include point mutations, insertion–deletion mutations, and shear mutations of tumor-specific mutations. But it does not include single nucleotide polymorphisms, germline mutations, copy number variation, and structural variation. The 416-gene probe developed by GENESEEQ Biotechnology completely covers the full exons, partial intron regions, and variable cleavage regions of the 416 genes, covering ∼1.2 Mb. The comprehensive and effective coverage ensures the reliability of TMB detection results. Data of >10,000 cases of Chinese TMB-based solid tumors have been collected, and TMB high/low populations have been classified according to the well-recognized quartile distribution method (Supplementary Table S2).

p53 protein expression by immunohistochemistry

The immunohistochemistry test in this study was applied in the Pathology Department of Shengjing Hospital of China Medical University. The paraffin section was deparaffinized, hydrolyzed, and repaired at high pressure. S–P method was used in immunohistochemical staining. The p53 kit was manufactured by Beijing Zhongshan Jinqiao Biotechnology Co., Ltd. The antibody used was Mouse Anti-p53 Protein Monoclonal Antibody (clone ID DO-7, 1:100). Immunolocalization was performed with the diaminobenzidine detection kit. Counterstaining was done by hematoxylin. Phosphate-buffered saline was used as negative control in place of the I antibody, and a known positive tissue section was used as the positive control.

The positive staining of p53 protein is mainly located in the nucleus with brownish-yellow particles and a clear background. Clear brownish-yellow particles in the nucleus indicate p53 protein positive expression. The samples are divided into four grades according to the percentage of positive cells. Positive cells ≤10% were considered negative; 11–25% considered (+); 26–50% considered (++); >50% considered (+++). It should be noted that the p53 protein we tested is a mutated protein.

Ethics approval

All the patients in our study signed the medical informed consent, and they all agreed with the treatment of rAd-p53 and chemotherapy. The study was approved by the Ethics Committee of Shengjing Hospital of China Medical University and was performed according to the tenets of the Declaration of Helsinki.

Results

NGS results

A total of 30 variant genes were found from the NGS test results, all of which are found in the COSMIC database. Twenty-four of the 30 genes are found in the Cancer Gene Census catalog, except ROS1, LYN, C11orf30, PKD1, CDKN1C, and GATA4. 7, 7, 8, 10, and 5 mutated genes were detected in the five samples of Nos. 1–4A and B, respectively, and the average number of mutated genes was up to 7 (Table 3). TP53 gene was mutated in two of the four patients. The p.P278R (c.C833G) missense mutation was found in the TP53 gene of patient No. 1. The cysteine141 of the TP53 gene of patient No. 4-B was mutated into a stop codon and led to a truncation mutation.

In the other two cases, there were mutations in genes CREBBP, LYN, CDKN2A, and JAK2 located upstream of the TP53 gene, which could potentiate and upregulate the antitumor activity of p53 (Table 3). Eleven additional genes are involved in the p53-related signaling pathways or interact with p53 (CCNE1,¹⁵ CDH1,¹⁶ Dnmt3A,¹⁷ SOX2,¹⁸ MDM4,^19
–21 BRCA2,^22
–24 PKD1,²⁵ RAD50, SMARCA4,^26,27 FGFR2,²⁸ MCL ^29
–31). MDM4 gene mutation was found in both No. 2 and No. 4-A. Results such as mutated gene name, mutation location, type of change, the gene pathways and the pathways and interactions with p53 or its related pathways of the mutated genes according to the search result of the COSMIC tumor database and the GeneCards database, and the Gene Ontology annotations are summarized in Table 2.

Table 2.

Next-generation sequencing results and analysis of pathological tissue specimens of patients 1–4

Gene	Mutation Location (NM_exon)	Type of Change	Pathways and Interactions with p53^Refs.	Pathways	Gene Ontology Annotations
Sample No. 1
CDH1	3	p.R90W (c.C268T)	Yes¹⁶	Immunoregulatory interactions between a lymphoid and a nonlymphoid cell	Calcium ion binding and protein phosphatase binding
DNMT3A	10	p.G550R (c.G1651A)	Yes¹⁷	Validated targets of C-MYC transcriptional repression	Identical protein binding and methyltransferase activity
FGFR4	6	p.Y301C (c.A902G)	Non	GPCR pathway and RET signaling	Transferase activity, transferring phosphorus-containing groups and protein tyrosine kinase activity
ROS1	36	p.R1948C (c.C5842T)	Non	GPCR pathway and spinal cord injury	Transferase activity, transferring phosphorus-containing groups and protein tyrosine kinase activity
TP53	4	p.P278R (c.C833G)	N/A	DNA double-strand break repair and development IGF-1 receptor signaling	Transcription factor activity, sequence-specific DNA binding, and protein heterodimerization activity
CCNE1	Amplification multiple 11.7690930431	N/A	Yes¹⁵	Chaperonin-mediated protein folding	Protein kinase binding and kinase activity
SOX2	Amplification multiple 3.1909905487	N/A	Yes¹⁸	Mesodermal commitment pathway and signaling by Wnt	Transcription factor activity, sequence-specific DNA binding, and protein heterodimerization activity
Sample No. 2
C11orf30	12	p.T563M (c.C1688T)	Non	Mesodermal commitment pathway	Protein homodimerization activity
CREBBP	9	p.R625H (c.G1874A)	Yes³⁶	CCR5 pathway in macrophages and phospholipase-C pathway	Transcription factor activity, sequence-specific DNA binding, and transcription factor binding
KDM5A	24	p.R1350X (c.C4048T)	Non	Circadian rhythm-related genes	Transcription factor activity, sequence-specific DNA binding, and transcription coactivator activity
ROS1	15	p.V726I (c.G2176A)	Non	See No. 1	See No. 1
PIK3R1	6	p.S229L (c.C686T)	Non	Development IGF-1 receptor signaling and CTLA4 signaling	GTP binding and transcription factor binding
LYN	2	p.R25H (c.G74A)	Yes³⁷	Phospholipase-C pathway and Kit receptor signaling pathway	Transferase activity, transferring phosphorus-containing groups and protein tyrosine kinase activity
MDM4	11	p.G455D (c.G1364A)	Yes^19 –21	Regulation of TP53 activity and cellular senescence	Enzyme binding
Sample No. 3
BRCA2	11	p.I1929V (c.A5785G)	Yes^22 –24	DNA double-strand break repair	Protease binding and histone acetyltransferase activity
ETV1	3	p.D38G (c.A113G)	Non	Transcriptional misregulation in cancer and p38 signaling mediated by MAPKAP kinases	Transcription factor activity, sequence-specific DNA binding
CDKN2A	1	p.L49V (c.C145G)	Yes³⁵	Cellular senescence and cyclins and cell cycle regulation	Poly(A) RNA binding and transcription factor binding
JAK2	6	p.M187I (c.G561C)	Yes³⁸	Development EPO-induced Jak-STAT pathway and RET signaling	Transferase activity
PKD1	43	p.F3910L (c.T11728C)	Yes²⁵	Cargo trafficking to the periciliary membrane and organelle biogenesis and maintenance	Protein kinase binding and protein domain-specific binding
RAD50	24	p.1243_1251del (c.3729_3752del)	Yes	DNA double-strand break repair	ATPase activity and ATP-dependent DNA helicase activity
MED12	42	p.Q2076X (c.C6226T)	Non	Regulation of Wnt-mediated beta-catenin signaling and target gene transcription	Chromatin binding and receptor activity
BAP1	13	p.Q456K (c.C1366A)	Non	DNA double-strand break repair	Chromatin binding
Sample No. 4-A
SMARCA4	12	p.G648fs (c.1944_1947del)	Yes^26,27	PEDF-induced signaling and signaling by Wnt	Chromatin binding and transcription coactivator activity
BAP1	14	p.V604L (c.G1810T)	Non	See No. 3	See No. 3
CDKN1C	1	p.A224T (c.G670A)	Non	Transcription_role of VDR in regulation of genes involved in osteoporosis	Cyclin-dependent protein serine/threonine kinase inhibitor activity
FGFR2	6	p.K405E (c.A1213G)	Yes²⁸	GPCR pathway and RET signaling	Protein homodimerization activity
FGFR4	2	p.T46I (c.C137T)	Non	See No. 1	See No. 1
GATA4	2	p.P23T (c.C67A)	Non	HOP signaling and DREAM repression and dynorphin expression	Transcription factor activity
GRIN2A	9	p.P594S (c.C1780T)	Non	RET signaling and protein–protein interactions at synapses	Calcium channel activity and ionotropic glutamate receptor activity
MDM4	3	p.V371I (c.G1111A)	Non	See No. 2	See No. 2
FGFR3	10	p.D513N (c.G1531A)	Non	GPCR pathway and RET signaling	Transferase activity
MCL1	Amplification multiple 2.49065138176	N/A	Yes	Apoptosis and autophagy and innate immune system	Protein homodimerization activity and BH3 domain binding
Sample No. 4-B
TSC1	Sequence NM_000368	D23Afs^*3	Non	Development IGF-1 receptor signaling and RET signaling	Binding and chaperone binding
SMARCA4	Sequence NM_001128844	N248Ifs^*55	Yes	See No. 4-A	See No. 4-A
SMARCA4	Sequence NM_001128844	c.1944–3_1947del	Yes	See No. 4-A	See No. 4-A
MCL1	Sequence NM_021960	Gene amplification	Yes^29 –31	See No. 4-A	See No. 4-A
ZNF217	Sequence NM_006526	Gene amplification	Non	Transcription factor activity	Transcription factor activity, sequence-specific DNA binding
TP53	Sequence NM_001126112	C141^*	N/A	See No. 1	See No. 1

GPCR, G protein coupled receptor; IGF-1, insulin-like growth factor-1.

Table 3.

TP53 gene mutation, p53 protein expression, and tumor mutational burden in various specimens

Specimen	Mutated Genes	TP53 Mutation	TP53 Upstream Gene	p53 Protein Expression	TMB (Mutation/Mb)
No. 1	7	Yes	N/A	^a	1.11
No. 2	7	No	CREBBP, LYN	−	7.78
No. 3	8	No	CDKN2A, JAK2	+++	3.33
No. 4-A	10	No	N/A	+	5.56
No. 4-B	5	Yes	N/A	^a	^a

The method or calculation was not applied because of the conditions of the specimens.

−, Negative; +, positive; TMB, tumor mutational burden.

Tumor mutational burden

According to the NGS test results, the TMB of specimens Nos. 1–4-A was calculated to be 1.11/Mb, 7.78/Mb, 3.33/Mb, and 5.56/Mb, respectively (Table 3). The median TMB was 4.44/Mb. In GENESEEQ Biotechnology's Chinese-based solid tumor TMB database, according to the generally accepted quartile method that distinguishes the high/low population of TMB, the specimens of Nos. 1–4-A were ranked at 96.3%, 38%, 82.5%, and 57.9%, respectively, and none of them were in the top 25% (Fig. 1). The TMB of specimen No. 4-B could not be calculated because of the limit of detection technique at that time.

Figure 1.

TMB quartile distribution. (A) Specimen No. 1; (B) Specimen No. 2; (C) Specimen No. 3; (D) Specimen No. 4-A. TMB, tumor mutational burden. Color images are available online.

P53 protein expression by immunohistochemistry

Three specimens from No. 2, No. 3, and No. 4-A were detected by immunohistochemistry, and the corresponding results were obtained (Table 3). Immunohistochemistry results showed that p53 protein expression was negative in specimen No. 2, strong positive (+++) in specimen No. 3, and positive (+) in specimen No. 4-A (Fig. 2). Due to the small amount of tissue in specimen No. 1 and No. 4-B, immunohistochemistry was not implemented.

Figure 2.

P53 protein under immunohistochemistry microscope (1 × ). (A) Specimen No. 2; (B) Specimen No. 3; (C) Specimen No. 4-A. Color images are available online.

Discussion

There are somatic TP53 missense mutations in ∼50% of human cancers.³² The literature reports^33,34 show that nuclear p53 protein accumulation was often in CSs (73%), LMSs (38%), and ESSs (27%). In this study, all patients probably had abnormal functions of the TP53 gene in inhibiting tumorigenesis and progression. As mentioned above, TP53 gene mutations were found in patients No. 1 and No. 4. No TP53 gene mutation was found in patient No. 2, but there were mutations in the upstream genes CREBBP and LYN, which can regulate TP53 and mutated p53 protein accumulation was also found. As patient No. 3, no TP53 gene mutation exists, but there were mutations in upstream genes CDKN2A and JAK2, which can regulate TP53, and there was a large accumulation of mutated p53 protein.

According to the GeneCards database, TP53 can be activated by its upstream genes (CDK7, PRKDC, MAPK8, EIF2AK2, MAPK13, CDKN2A, STK11, CHEK2, etc.). The roles of the four TP53 upstream genes discovered from patient No. 2 and No. 3 in this study were as follows: (1) CDKN2A ³⁵ can activate p53 by blocking MDM2-induced degradation of p53 and enhancing p53-dependent transactivation and apoptosis, and also induces G2 arrest and apoptosis in a p53-independent manner by preventing the activation of cyclin B1/CDC2 complexes. (2) CREBBP is involved in “TP53 Regulates Transcription of Cell Death Genes,” “Regulation of TP53 Activity,” “Direct p53 effectors” pathways, which are all related to p53 functions. CREBBP binds specifically to p53 as a coactivator for p53-induced transcriptional activation.³⁶ (3) LYN can upregulate p53 levels, promote p53-mediated transcription, reverse Mdm2-mediated p53 degradation, increase p53-mediated apoptosis,³⁷ and play a role in activating p53 and Mdm2-mediated regulation of p53. (4) JAK2 mutation affects p53 response to DNA damage by enhancing MDM2. ³⁸ Some think that in some cases, the p53 pathway is also inactivated in tumors carrying wild-type p53. When the amplification of MDM2/MDMX leads to p53 instability, it indirectly leads to the inactivation of the p53 pathway.^39,40

Maybe we speculate that CDKN2A, JAK2, CREBBP, and LYN all contribute to the inhibition of tumorigenesis by TP53. Mutations in these four genes in the patients in this study may probably either directly or indirectly lead to the reduction or inhibition in the related effects of TP53 in suppressing tumors. Besides, in this study MDM4 gene mutations were found in both No. 2 and No. 4-A, but the TP53 gene test results were negative and so the wild-type p53 function may be inactivated in this case. We used rAd-p53 in combination with chemotherapy for cancer treatment. A study by Wei and collegues⁴¹ showed that chemoresistance in cancer has previously been attributed to gene mutations or deficiency. On the one hand, rAd-p53 introduced an exogenous wild-type TP53 gene to repair the damage caused by the above-mentioned gene mutations and to stabilize the genome. On the other hand, chemotherapy causes severe DNA damage in tumor cells, and p53 protein exerts the function of inducing apoptosis. Li et al.'s⁷ clinical trial found that intra-arterial infusion of combined rAd-p53 and chemotherapy significantly increased the survival rate of patients with oral squamous cell carcinoma compared with intra-arterial chemotherapy. We believe that the TP53 gene and chemotherapeutic drugs play a synergistic role. Therefore, we hypothesize that mutations in TP53, CREBBP, LYN, CDKN2A, and JAK2 may be indications for the use of rAd-p53 gene therapy in combination with chemotherapy for uterine sarcoma. Therefore, exogenous rAd-p53 treatment had a good therapeutic effect.

We observe the phenomenon that the TP53 gene mutation status of No. 4-A and No. 4-B is different. It could be noted that sample extraction or tumor heterogeneity caused the failure to detect the TP53 gene mutation status. But it needs to be proved by a larger sample size. Whether TP53 gene mutation can predict the efficacy of rAd-p53 therapy remains to be explored.

Inactivation of the TP53 gene during tumorigenesis is mostly caused by point mutations in the DNA-binding region, which in turn causes accumulation of a stable, nonfunctional, or abnormally functional protein in the tumor cell.^42,43 Some studies suggest that immunohistochemistry detection of mutated p53 protein can be used as a moderately sensitive, highly specific marker for predicting mutations in the TP53 gene.⁴⁴ We have applied immunohistochemistry in three samples. Unfortunately, immunohistochemistry was not implemented for specimen No. 1 and No. 4-B with TP53 gene mutation because of the lack of tissues. The p53 protein immunohistochemistry result of No. 3 and No. 4-A was positive; instead, the NGS result for the TP53 gene was negative. For instance, specimen No. 3 shows a large accumulation of mutated p53 protein without detection of the TP53 mutation, however. Although no TP53 gene mutation was detected, there was a mutation in the upstream gene CDKN2A of TP53, which directly affected the function of the TP53 gene. Therefore, exogenous rAd-p53 treatment had a good therapeutic effect. But this still cannot explain the phenomenon of accumulation of mutated p53 protein. The inconsistent results of the TP53 gene mutation status may be due to sample extraction or tumor heterogeneity. So it needs to be proved by a larger sample size. Whether p53 protein accumulation can predict the efficacy of rAd-p53 therapy remains to be explored.

The number of mutated genes carried by tumors has recently been the topic of great interest.^45,46 The number of mutant genes in different tumor type tissues varies, ranging from <1 in thyroid cancer and testicular cancer to >10 in endometrial cancer and rectal cancer. On average, each tumor type carries four mutated genes.⁴⁷ In this study, patients with uterine sarcoma carry an average of seven mutations unexpectedly, which exceeds the average reported in the literature.

We know the TP53 gene is called “genome guardian.” The stabilization of “genome guardian” is caused by various cellular stresses such as irradiation, oncogenic activation, and exposure to genotoxic chemicals. Since most of these processes damage DNA, there are various DNA repair mechanisms to correct the damage incurred, and TP53 has been shown to play an important role in several of these repair mechanisms, including nucleotide excision repair,^48,49 base excision repair,^48,50 mismatch repair,⁵¹ homologous recombination repair, and nonhomologous end-joining. We speculated that the more types of tumor mutations, the more unstable the genome environment, the better the implication of malfunction of the “genome guardian” TP53. Fifteen of the 30 mutated genes we detected were involved in p53-related signaling pathways or interacted with p53. The relationship between p53 and the remaining 15 genes has not been found in the current GeneCards database. It is reported that ∼50% of tumor-causing mutations are not in the known oncogene spectrum.⁴⁷ Due to the complexity and variety of genomes, we know very little about it, and many things cannot be explained by the existing data and cognition and need to be further explored. A study from 13 groups of high-throughput sequencing revealed 3,509 target genes of p53.⁵² With the continuous expansion of the database and the deep research of the TP53 gene, it is believed that the links between them can be discovered in the future.

In our research, we detected 15 mutated genes that are involved in p53 signaling pathways or interact with p53. And most of the 15 mutated genes are involved in pathways that relate to the occurrence and development of tumors such as DNA double-strand break repair, RET (Ret Proto-Oncogene), G protein coupled receptor pathway, development insulin-like growth factor-1 receptor signaling, and cellular senescence. Based on the role of the TP53 gene in genomic stability and the efficacy of our patients, it is inferred that the number of tumor mutated genes can be used as a genomic characterization to predict the efficacy of rAd-p53 combined with chemotherapy in the treatment of uterine sarcoma.

Studies have shown that the higher the tumor mutation load, the higher the degree of microsatellite instability and the lack of mismatch repair, the better outcome of anti-PD-1 immunotherapy.^53,54 Yarchoan et al. found that the higher the mutation load of a cancer type, the greater the likelihood of responding to checkpoint inhibitors. The extent of >50% of cancer's response to checkpoint inhibitors can be explained by the mutational load of cancer.⁵⁵ Besides, studies have shown that if the tumor mutation load increases, the types of mutation will also increase, but the two are not linearly related. The tumor mutation burden of the four patients with uterine sarcoma in this study is low, with a median TMB of 4.44/Mb, consistent with the reported conclusion that the TMB of the primary soft tissue sarcoma is not high,⁵⁶ the median TMB of which is 5.4 mutations/Mb.⁵⁷ In conclusion, we speculate that the TMB of uterine sarcoma is not high, and it is not a genomic characterization for rAd-p53 combined with chemotherapy.

There are also several limitations to our research. As we know uterine sarcoma is a rare, aggressive, and lethal gynecological cancer accounting for 3–4% of all uterine corpus cancers.^58,59 As mentioned above, our previous study only collected 12 cases of uterine sarcoma treated with rAd-p53 combined with chemotherapy from 2009 to 2015 due to the low incidence rate of uterine sarcoma. Results show that the treatment resulted in one CR, seven PR, three SD, and one PD, respectively. Besides for some reason, we only tested five samples from four PR patients in this study. We always know that more samples are needed to make sure our results are reliable. But there were a total of four patients, one CR and three PR, which accepted radical hysterectomy in another hospital including the foreign hospital rather than our hospital. As the other two PR patients underwent surgery a long time ago, the department of pathology cannot provide their specimens anymore. So it was difficult to get their samples above.

Also, there is no control group due to the low incidence rate of uterine sarcoma. There was only one PD patient in our previous clinical study, and the sample from the PD patient should be tested to compare the difference between PR and PD patients in this study. But the PD patient accepted radical hysterectomy in another hospital, so it was also difficult to get the sample. Although the number of cases in our study is currently the largest for uterine sarcoma treated with clinical gene therapy, there are only five samples involved, which is fewer than other cancer types. Since randomized trials are scarce for uterine sarcoma patients, personalized therapy approaches should be supported. And further new treatment strategies are desperately needed to improve the 5-year survival rate of uterine sarcomas.

There have been many recent, successful applications of NGS in establishing the etiology and guiding therapeutic decision making of neoplastic diseases. NGS has been utilized as a promising diagnostic tool with its advantages of accuracy, sensitivity, and high throughput compared with traditional clinical testing. However, there are also major, but not insurmountable, obstacles to the increased clinical implementation of NGS, such as quality assurance. Different sequencing platforms vary in their ability to identify variants, even when sequencing the same genome.^60
–62 Next, we should consider validating our NGS results by more methods to assess the sensitivity, specificity, and accuracy of the assay. For instance, in the next plan, we could use real-time qPCR platform or Sanger sequencing to validate the gene variants. Because a total of 30 variant genes were found from the NGS test in this study, it must be hard to finish the alternate methods within a short period. Anyway, Xie et al.'s study⁶³ showed that NGS-based molecular diagnostic test is more sensitive in detecting genomic alterations in cancer, and supported a direct clinical use for this method to guide targeted therapy.

Although we did not establish and validate by alternate methods, we compare our results with the COSMIC database, which is the world's largest and most comprehensive resource for exploring the impact of somatic mutations in human cancer. Through searching we found that the cancer types of ESS, undifferentiated sarcoma of uterus, and LMS are usually a part of other cancer types. Together with the rarity of uterine sarcoma, this contributes to just a few sequenced cases, which are far less than other cancer types. And no suitable reference genes have been identified in uterine sarcoma. The common mutant genes of ESS are TP53 (mutation frequency 24%), MED12 (10%), ASXL1 (15%), KMT2C (10%), KRAS (24%), etc. The common mutant genes of undifferentiated sarcoma of the uterus are MED12 (50%) and TP53 (50%). The common mutant genes of uterine LMS are TP53 (33%), MED12 (8%), ATRX (23%), RB1 (16%), etc. Regrettably, our NGS results overlap slightly with the database due to their highly divergent genetic aberrations. The four overlapping genes are CREBBP and ROS1 (from patient No. 2), MED12 (from patient No. 3), and TP53 (from patient No. 4). Even so, our work is the first report on the genomic characterization of uterine sarcoma for gene therapy. The purpose of this study was to select appropriate genomic characterization for these tumors to guide targeted therapy for direct clinical use. Above all, this research has raised many questions, and further studies with larger cohorts of uterine sarcoma are required in the future.

In summary, the TP53 gene controls the entire gene network and maintains the homeostasis of the genome. It is considered to be the most important tumor suppressor gene and plays the role of the guardian of the genome.^64,65 It can induce apoptosis of cells carrying the mutated gene or initiate gene repair, thereby avoiding harmful consequences of the mutated gene.⁶⁶ Although the termination of apoptosis is not a necessary condition for tumor growth, the absence of wild-type p53 can maintain the survival of those cells that carry a wide range of genetic errors⁶⁷ and increase the frequency of new gene mutations. These may be the causes of cancer occurrence, drug resistance and may even cause more serious recurrence of tumors.^68,69 “Silencing” and “disability” of the TP53 gene imply that the entire “gene network” is defective and disordered. Treatment of rAd-p53 introduces exogenous wild-type TP53 genes and activates the TP53-regulated transcriptional response. But instead of activating a gene pathway, it activates a network that inhibits cancer. The cascade reactions generated cannot be explained by the fact that DNA damage at the single-cell level activates TP53 to cause gene expression. Based on the good effect rAd-p53 has in the group of uterine sarcoma with high degree of malignancy and radiotherapy and chemotherapy insensitivity published in our previous article, combined with the preliminary results of this study, we speculate that the mutation of TP53 gene itself and the upstream gene mutations such as CREBBP, LYN, CDKN2A, and JAK2, and the number of tumor mutated genes can be the genomic characterization of uterine sarcoma for rAd-p53 combination chemotherapy treatment.

We studied the role of clinical next-generation sequencing in precision medicine and the relationship between genomic variation and disease.^70,71 Necessarily, genomic data will influence the medical decision. But we are still facing the problems such as unexplained drug resistance, genomic heterogeneity of tumors, insufficient means for monitoring responses and tumor recurrence, and limited knowledge about the use of drug combinations.⁷⁰ In the future, we will expand the samples for further verification of our prediction.

Footnotes

Acknowledgments

The authors thank the doctors in the Department of Obstetrics and Gynecology, Shengjing Hospital of China Medical University and the collaborators of Nanjing GENESEEQ Biotechnology, Inc. for their encouragement and collaboration with this research.

Author Disclosure

No competing financial interests exist.

Funding Information

No funding was received for this article.

Supplementary Material

Supplementary Table S1

Supplementary Table S2

References

Iyer

, Hanrahan

, Milowsky

, et al. Genome sequencing identifies a basis for everolimus sensitivity. Science, 2012; 338:221.

Kris

, Johnson

, Berry

, et al. Using multiplexed assays of oncogenic drivers in lung cancers to select targeted drugs. JAMA, 2014; 311:1998–2006.

Turnbull

. Introducing whole-genome sequencing into routine cancer care: the Genomics England 100 000 Genomes Project. Ann Oncol, 2018; 29:784–787.

Lin

, Wang

, Zhang

, et al. Applications of virus vector-mediated gene therapy in China. Hum Gene Ther, 2018; 29:98–109.

Zhang

, Li

, et al. The first approved gene therapy product for cancer Ad-p53 (Gendicine): 12 years in the clinic. Hum Gene Ther, 2018; 29:160–179.

, Li

, Zhang

, et al. In vitro and clinical studies of gene therapy with recombinant human adenovirus-p53 injection for oral leukoplakia. Clin Cancer Res, 2009; 15:6724–6731.

, Li

, Wang

, et al. Selective intra-arterial infusion of rAd-p53 with chemotherapy for advanced oral cancer: a randomized clinical trial. BMC Med, 2014; 12:16.

, Huang

, Wang

, et al. 47. A combination therapy of selective intraarterial gendicine infusion with chemotherapy for locally advanced head and neck carcinoma. Mol Ther, 2006; 13:S20.

Xia

, Du

, Wang

, et al. Treatment of uterine sarcoma with rAd-p53 (Gendicine) followed by chemotherapy: clinical study of TP53 gene therapy. Hum Gene Ther, 2018; 29:242–250.

10.

Bolger

, Lohse

, Usadel

. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics, 2014; 30:2114–2120.

11.

, Durbin

. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics, 2009; 25:1754–1760.

12.

McKenna

, Hanna

, Banks

, et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res, 2010; 20:1297–1303.

13.

Van der Auwera

, Carneiro

, Hartl

, et al. From FastQ data to high confidence variant calls: the Genome Analysis Toolkit best practices pipeline. Curr Protoc Bioinformatics 2013;43:11 10 11–33.

14.

Zehir

, Benayed

, Shah

, et al. Mutational landscape of metastatic cancer revealed from prospective clinical sequencing of 10,000 patients. Nat Med, 2017; 23:703–713.

15.

Montazeri

, Bouzari

, Azadmanesh

, et al. Overexpression of cyclin E and its low molecular weight isoforms cooperate with loss of p53 in promoting oncogenic properties of MCF-7 breast cancer cells. Asian Pac J Cancer Prev, 2015; 16:7575–7582.

16.

Fricke

, Keller

, Becker

, et al. Relationship between E-cadherin gene mutation and p53 gene mutation, p53 accumulation, Bcl-2 expression and Ki-67 staining in diffuse-type gastric carcinoma. Int J Cancer, 2003; 104:60–65.

17.

Wang

, Kamarova

, Shen

, et al. DNA methyltransferase-3a interacts with p53 and represses p53-mediated gene expression. Cancer Biol Ther, 2005; 4:1138–1143.

18.

Samulin Erdem

, Skaug

, Bakke

, et al. Mutations in TP53 increase the risk of SOX2 copy number alterations and silencing of TP53 reduces SOX2 expression in non-small cell lung cancer. BMC Cancer, 2016; 16:28.

19.

Chen

, Gilkes

, Pan

, et al. ATM and Chk2-dependent phosphorylation of MDMX contribute to p53 activation after DNA damage. EMBO J, 2005; 24:3411–3422.

20.

Jin

, Dai

, Lu

, et al. 14-3-3gamma binds to MDMX that is phosphorylated by UV-activated Chk1, resulting in p53 activation. EMBO J, 2006; 25:1207–1218.

21.

Shvarts

, Bazuine

, Dekker

, et al. Isolation and identification of the human homolog of a new p53-binding protein, Mdmx. Genomics, 1997; 43:34–42.

22.

McAlpine

, Porter

, Kobel

, et al. BRCA1 and BRCA2 mutations correlate with TP53 abnormalities and presence of immune cell infiltrates in ovarian high-grade serous carcinoma. Mod Pathol, 2012; 25:740–750.

23.

Rajagopalan

, Andreeva

, Rutherford

, et al. Mapping the physical and functional interactions between the tumor suppressors p53 and BRCA2. Proc Natl Acad Sci U S A, 2010; 107:8587–8592.

24.

Verma

, Rao

. p53 suppresses BRCA2-stimulated ATPase and strand exchange functions of human RAD51. J Biochem, 2013; 154:237–248.

25.

van Bodegom

, Roessingh

, Pridjian

, et al. Mechanisms of p53-mediated repression of the human polycystic kidney disease-1 promoter. Biochim Biophys Acta, 2010; 1799:502–509.

26.

, Zhang

, Chen

. The activity of p53 is differentially regulated by Brm- and Brg1-containing SWI/SNF chromatin remodeling complexes. J Biol Chem, 2007; 282:37429–37435.

27.

Hsu

, Lee

, Yeh

, et al. 58-kDa microspherule protein (MSP58) is novel Brahma-related gene 1 (BRG1)-associated protein that modulates p53/p21 senescence pathway. J Biol Chem, 2012; 287:22533–22548.

28.

Rodriguez-Enfedaque

, Bouleau

, Laurent

, et al. FGF1 nuclear translocation is required for both its neurotrophic activity and its p53-dependent apoptosis protection. Biochim Biophys Acta, 2009; 1793:1719–1727.

29.

Chaturvedi

, Sitailo

, Qin

, et al. Knockdown of p53 levels in human keratinocytes accelerates Mcl-1 and Bcl-x(L) reduction thereby enhancing UV-light induced apoptosis. Oncogene, 2005; 24:5299–5312.

30.

Pietrzak

, Puzianowska-Kuznicka

. p53-dependent repression of the human MCL-1 gene encoding an anti-apoptotic member of the BCL-2 family: the role of Sp1 and of basic transcription factor binding sites in the MCL-1 promoter. Biol Chem, 2008; 389:383–393.

31.

Yao

, Mi

, Gong

, et al. Anti-apoptosis proteins Mcl-1 and Bcl-xL have different p53-binding profiles. Biochemistry, 2013; 52:6324–6334.

32.

Soussi

, Ishioka

, Claustres

, et al. Locus-specific mutation databases: pitfalls and good practice based on the p53 experience. Nat Rev Cancer, 2006; 6:83–90.

33.

Kobayashi

, Uekuri

, Akasaka

, et al. The biology of uterine sarcomas: a review and update. Mol Clin Oncol, 2013; 1:599–609.

34.

Nordal

, Kristensen

, Stenwig

, et al. Immunohistochemical analysis of p53 protein in uterine sarcomas. Gynecol Oncol, 1998; 70:45–48.

35.

Stott

, Bates

, James

, et al. The alternative product from the human CDKN2A locus, p14(ARF), participates in a regulatory feedback loop with p53 and MDM2. EMBO J, 1998; 17:5001–5014.

36.

Mujtaba

, He

, Zeng

, et al. Structural mechanism of the bromodomain of the coactivator CBP in p53 transcriptional activation. Mol Cell, 2004; 13:251–263.

37.

Ren

, Cao

, Zhu

, et al. Lyn tyrosine kinase inhibits nuclear export of the p53 tumor suppressor. Cancer Biol Ther, 2002; 1:703–708.

38.

Nakatake

, Monte-Mor

, Debili

, et al. JAK2(V617F) negatively regulates p53 stabilization by enhancing MDM2 via La expression in myeloproliferative neoplasms. Oncogene, 2012; 31:1323–1333.

39.

Marine

, Dyer

, Jochemsen

. MDMX: from bench to bedside. J Cell Sci, 2007; 120:371–378.

40.

Brooks

, Gu

. p53 ubiquitination: Mdm2 and beyond. Mol Cell, 2006; 21:307–315.

41.

Wang

, Guo

, Zhou

, et al. Mitochondrial p53 phosphorylation induces Bak-mediated and caspase-independent cell death. Oncotarget, 2015; 6:17192–17205.

42.

Borresen-Dale

. TP53 and breast cancer. Hum Mutat, 2003; 21:292–300.

43.

Petitjean

, Achatz

, Borresen-Dale

, et al. TP53 mutations in human cancers: functional selection and impact on cancer prognosis and outcomes. Oncogene, 2007; 26:2157–2165.

44.

Takami

, Yoshida

, Fukushima

, et al. Revisiting TP53 mutations and immunohistochemistry—a comparative study in 157 diffuse gliomas. Brain Pathol, 2015; 25:256–265.

45.

Martincorena

, Roshan

, Gerstung

, et al. Tumor evolution. High burden and pervasive positive selection of somatic mutations in normal human skin. Science, 2015; 348:880–886.

46.

Tomasetti

, Marchionni

, Nowak

, et al. Only three driver gene mutations are required for the development of lung and colorectal cancers. Proc Natl Acad Sci U S A, 2015; 112:118–123.

47.

Martincorena

, Raine

, Gerstung

, et al. Universal patterns of selection in cancer and somatic tissues. Cell, 2017; 171:1029–1041 e1021.

48.

McKay

, Ljungman

, Rainbow

. Potential roles for p53 in nucleotide excision repair. Carcinogenesis, 1999; 20:1389–1396.

49.

Amundson

, Patterson

, Do

, et al. A nucleotide excision repair master-switch: p53 regulated coordinate induction of global genomic repair genes. Cancer Biol Ther, 2002; 1:145–149.

50.

Zurer

, Hofseth

, Cohen

, et al. The role of p53 in base excision repair following genotoxic stress. Carcinogenesis, 2004; 25:11–19.

51.

Lin

, Ramamurthi

, Mishima

, et al. p53 interacts with the DNA mismatch repair system to modulate the cytotoxicity and mutagenicity of hydrogen peroxide. Mol Pharmacol, 2000; 58:1222–1229.

52.

Fischer

. Census and evaluation of p53 target genes. Oncogene, 2017; 36:3943–3956.

53.

, Durham

, Smith

, et al. Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science, 2017; 357:409–413.

54.

, Uram

, Wang

, et al. PD-1 blockade in tumors with mismatch-repair deficiency. N Engl J Med, 2015; 372:2509–2520.

55.

Yarchoan

, Hopkins

, Jaffee

. Tumor mutational burden and response rate to PD-1 inhibition. N Engl J Med, 2017; 377:2500–2501.

56.

Cancer Genome Atlas Research

Network

, Electronic address edsc, Cancer Genome Atlas Research

Network

. Comprehensive and integrated genomic characterization of adult soft tissue Sarcomas. Cell, 2017; 171:950–965 e928.

57.

Chalmers

, Connelly

, Fabrizio

, et al. Analysis of 100,000 human cancer genomes reveals the landscape of tumor mutational burden. Genome Med, 2017; 9:34.

58.

Abeler

, Royne

, Thoresen

, et al. Uterine sarcomas in Norway. A histopathological and prognostic survey of a total population from 1970 to 2000 including 419 patients. Histopathology, 2009; 54:355–364.

59.

Amant

, Coosemans

, Debiec-Rychter

, et al. Clinical management of uterine sarcomas. Lancet Oncol, 2009; 10:1188–1198.

60.

Nothnagel

, Herrmann

, Wolf

, et al. Technology-specific error signatures in the 1000 Genomes Project data. Hum Genet, 2011; 130:505–516.

61.

Moore

, Hu

, Singleton

, et al. Global analysis of disease-related DNA sequence variation in 10 healthy individuals: implications for whole genome-based clinical diagnostics. Genet Med, 2011; 13:210–217.

62.

Rieber

, Zapatka

, Lasitschka

, et al. Coverage bias and sensitivity of variant calling for four whole-genome sequencing technologies. PLoS One, 2013; 8:e66621.

63.

Xie

, Lu

, Wu

, et al. Capture-based next-generation sequencing reveals multiple actionable mutations in cancer patients failed in traditional testing. Mol Genet Genomic Med, 2016; 4:262–272.

64.

Vogelstein

, Lane

, Levine

. Surfing the p53 network. Nature, 2000; 408:307–310.

65.

Levine

, Oren

. The first 30 years of p53: growing ever more complex. Nat Rev Cancer, 2009; 9:749–758.

66.

Lane

. Cancer. p53, guardian of the genome. Nature, 1992; 358:15–16.

67.

Lowe

, Jacks

, Housman

, et al. Abrogation of oncogene-associated apoptosis allows transformation of p53-deficient cells. Proc Natl Acad Sci U S A, 1994; 91:2026–2030.

68.

McGranahan

, Swanton

. Clonal heterogeneity and tumor evolution: past, present, and the future. Cell, 2017; 168:613–628.

69.

Johnson

, Mazor

, Hong

, et al. Mutational analysis reveals the origin and therapy-driven evolution of recurrent glioma. Science, 2014; 343:189–193.

70.

Collins

, Varmus

. A new initiative on precision medicine. N Engl J Med, 2015; 372:793–795.

71.

Adams

, Eng

. Next-generation sequencing to diagnose suspected genetic disorders. N Engl J Med, 2018; 379:1353–1362.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.92 MB

0.02 MB