Detection of genes mutations in cerebrospinal fluid circulating tumor DNA from central nervous system lymphomas patients using next generation sequencing

doi:10.21203/rs.3.rs-3448695/v1

Purpose

To evaluate the potential clinical value of cerebrospinal fluid (CSF) circulating tumor DNA (ctDNA) in the diagnosis and monitors the central nervous system (CNS) lymphomas.

Methods

This was a prospective study of 17 consecutive patients with B-cell lymphoma: 10 patients with CNS lymphomas and 7 patients with B-cell lymphomas at high clinical risk of CNS relapse. Genomic profiles were performed on the CSF and plasma samples of patients by next generation sequencing (NGS).

Results

In patients with CNS lymphomas, ctDNA was detected in 70.0% of CSF and 60.0% of plasma. The detection rate and gene mutation abundance of cerebrospinal fluid were higher than plasma (P=0.016). CSF had a unique genetic profile. Furthermore, we newly found that gene mutations consistent with plasma or lymphoma-related were also detected in the CSF of the high-risk group without CNS involvement. Analysis of paired plasma and CSF samples from three patients at different time points, changes of CSF ctDNA abundance occurred at the same time or earlier than clinical disease changes, which could timely monitor the therapeutic response and relapse trend.

Conclusions

The detection rate of ctDNA in CSF is higher than that in plasma. The dynamic monitoring of ctDNA in CSF has hint significance for therapeutic response of CNS lymphoma patients.

Central nervous system (CNS) lymphomas

circulating tumor DNA (ctDNA)

cerebrospinal fluid (CSF)

next generation sequencing (NGS)

Central nervous system (CNS) involvement remains to be the difficulty in the treatment of B-cell lymphomas. It occurs in about 2–6% of patients with diffuse large B-cell (DLBCL) and Mantle Cell Lymphoma (MCL) or high-grade B-cell (HGBCL) [1]. Patients with CNS involvement often have a very poor prognosis, and the majority of patients die within 6 months after diagnosis. Therefore, early recognition of CNS involvement and high-risk groups, followed by timely intervention is key to improving patient outcomes. Unfortunately, the predictive capabilities of current clinical prognostic models are still limited, such as the CNS-International Prognostic Index (IPI), as almost half of CNS involvement occurs in medium or low-risk groups [2]. The diagnosis of CNS lymphomas is typically based on MRI, brain tissue biopsies/ resections, and cytopathological examination of the cerebrospinal fluid (CSF) [3]. However, MRI and CSF cytopathological examination have poor specificity and sensitivity, respectively, and brain biopsies are invasive. Thus, there is a pressing need to find less invasive and more reliable methods for diagnosing and monitoring patients with CNS lymphomas.

A promising innovative approach for analyzing circulating tumor cells or circulating nucleic acids in biofluids is known as liquid biopsy. In B-cell lymphomas, circulating tumor DNA (ctDNA) in the blood plasma or CSF is one of the analytes that is currently being studied the most [4]. The ctDNA comprises short DNA fragments derived from tumor cells and theoretically represents the entire spectrum of tumor genome aberrations [5]. Improvements in polymerase chain reaction (PCR) methods and next-generation sequencing (NGS) technologies based on the detection of tumor genomic aberrations in ctDNA provide new methods for early detection of minimal residual disease (MRD) during or after therapy, allowing for early prediction of disease relapse and timely therapeutic intervention [6].

In CNS lymphomas, CSF is in close contact with the tumor cells. Only a limited number of current studies are available of the CSF ctDNA. In this study, we sought to determine whether analysis of CSF ctDNA can be used to detect individual mutations in CNS lymphoma patients and evaluate therapeutic response.

Patients

A total of 17 patients with B-cell lymphomas who were diagnosed and treated in the Hematology Department at Fujian Medical University Union Hospital were included in the study. The histologic subtype of each lymphomas sample was determined according to World Health Organization (WHO) criteria. All clinical data were obtained by consulting medical records and contacting patients by telephone. This study was executed under the approval of the Institutional review board of the Fujian Medical University Union Hospital in accordance with the Declaration of Helsinki (number 2019YF042-01), and informed consent was obtained from each patient who participated.

Patients were recruited into two groups: The first group included patients (n = 10) with CNS lymphoma; The second group included patients (n = 7) without CNS disease but risk factors for CNS relapse. Risk factors for CNS relapse defined according to the National Comprehensive Cancer Network (NCCN) guidelines. Prophylactic intrathecal (IT) methotrexate (MTX) was administered every 3 weeks together with systemic chemotherapy in patients with risk factors for CNS relapse. Evaluation of response was assessed using imaging examination at the end of treatment and/or at suspicion of disease progression.

DNA extraction, library preparation and target sequencing

As previously described [7], both CSF and plasma-derived cfDNA and genomic DNA from peripheral blood lymphocytes was extracted with the Magnetic Serum/Plasma DNA Maxi Kit (DP710, TIANGEN Biotechnology, China). DNA was quantified using the Qubit dsDNA HS Assay Kit (Invitrogen) and sheared into 150 to 200 bp fragments was subjected to library construction in prior to fragment length analysis using Qsep100 (Bioptic, Taiwan, China). Fragmented DNA libraries were constructed using nano DNA library prep kit (for Illumina®) (Nanodigmbio, Nanjing, China) for end-repairing, adding A-tailing, and adapter ligation. DNA libraries were then subjected to PCR amplification and purification. The fragment quantification was performed using Qsep100 automatic nucleic acid analysis system (BiOptic, Taiwan, China). Subsequent, the libraries were captured hybrid with a targeted sequencing gene panel, including 127 genes that are recurrently mutated in lymphoma (Lymcare®, Oriomics Biotech Inc, Zhejiang, China) using xGen® Hybridization and Wash Kit (IDT). The barcoded duplex adapters were eventually blocked by NadPrep NanoBlockers (for illumina) (Nanodigmbio, Nanjing, China).). The libraries were sequenced on a NovaSeq 6000 platform (Illumina) with a mean coverage depth of about 9,000X for cfDNA samples.

Bioinformatics analysis

Sequencing data was processed as previously described [7]. In brief, the data was first demultiplexed and subjected to FASTQ file quality control to remove low quality data or N bases. All clean paired-end reads were aligned to the human reference genome (19) using Burrows-Wheeler Aligner (BWA). Variant calling was performed using Mutect2 from Genome Analysis Toolkit (GATK 3.4.0, https://software.broadinstitute.org/gatk/; Broad Institute, Cambridge, MA, USA). Then, the detected mutations were further selected following established criteria: somatic mutations with allelic fractions of 1% and/or supported by 2 reads were disregarded. EXAC_EAS or genomAD_EAS should be less than 0.05. DNAs from leukocytes of the same patients were sequenced as germline controls.

CNVkit was selected to identify copy number variations (CNVs). The cutoff of log2 ratio was set at ± 0.6 for copy number changes (corresponding to 1.5-fold copy number gain and 0.65-fold copy number loss).

Statistical analysis

Statistical analyses were conducted using R package (version 3.61). Two-sided P value < 0.05 was considered to represent a statistically significant difference.

Baseline Characteristics of Patients

Table 1 displays the baseline clinical characteristics of the two groups of patients. The median age of all patients was 46 years (range, 22–69 years), and the male to female ratio was 12:5. Histologically, 15 (88.2%) patients were diagnosed with DLBCL, one (5.9%) patients with MCL, and one (5.9%) with HGBL. Twelve (70.6%) of 17 patients had stage IV lymphoma, and 4 patients (23.5%) had IPI scores of 3–5. The majority of patients received the R-COEP regimen as first-line therapy. Of the 17 patients, 11 had complete/partial response (CR/PR), 1 had stable disease (SD) and 5 had progressive disease (PD).

Table 1

Characteristics of patients at the time of enrolment and outcome
ID	Age	Gender	CNS involvement	Histology	COO classification	Stage	IPI	Outcome
High risk CNS involvement lymphoma
P005	69	M	N	DLBCL	GCB	II	2	CR
P006	38	F	N	DLBCL	GCB	IV	1	CR
P008	44	F	N	DLBCL	non-GCB	IV	1	CR
P010	66	M	N	DLBCL	non-GCB	I	1	CR
P012	59	M	N	DLBCL	non-GCB	III	2	CR
P013	22	M	N	DLBCL	non-GCB	I	0	CR
P018	34	M	N	DLBCL	non-GCB	IV	1	CR
CNS lymphoma
P001	67	M	Y	MCL	NA	IV	4	PD
P002	50	M	Y	HGBCL	NA	IV	2	PD
P003	60	M	Y	DLBCL	GCB	IV	3	CR
P004	40	M	Y	DLBCL	GCB	IV	1	PD
P007	46	M	Y	DLBCL	non-GCB	IV	2	CR
P009	57	F	Y	DLBCL	NA	IV	3	PD
P011	40	M	Y	DLBCL	GCB	IV	2	CR
P014	45	F	Y	DLBCL	non-GCB	I	0	PD
P015	55	M	Y	DLBCL	non-GCB	IV	3	SD
P019	44	F	Y	DLBCL	non-GCB	IV	2	CR
COO: cell of origin. F: female; GCB: Germinal Center B-cell-like; ID: patient identification number; M: male; N: no; NA: not available; Y: yes.

CSF is superior in identifying gene mutations of CNS lymphoma than plasma

Here, we first sought to analyze the presence of ctDNA in the CSF and plasma from 10 patients with CNS disease at time of enrollment. The results showed that ctDNA was detected in 70.0% (7/10) of CSF and 60.0% (6/10) of plasma. The detection rate of ctDNA in the CSF samples were higher than the paired plasma samples. In CSF, PIM1 (30%), MYD88 (30%), KMT2D (20%) and CD79B (20%) were the most common mutated genes, whereas in plasma, PIM1 (30%), MYD88 (30%) and CD79B (20%) were the most common mutated genes (Fig. 1). A total of 46 genetic mutations were detected in CSF while only 37 genetic mutations were detected in plasma (Fig. 2). The gene mutation abundance of CSF were significantly higher than plasma (P = 0.016) (Figs. 3 and 4). Moreover, we identified CSF ctDNA had 35 unique gene mutations, which mainly included BIRC3, KMT2D, PIM1, and TSC2, among others. A full list of unique mutations and gene functions was shown in Supplementary Tables 1 and 2. The most frequent mutation in 11 shared mutations is the MYD88 X160R mutation.

CSF ctDNA detection might indicate the CNS involvement with high risk lymphomas.

The majority of CNS relapses occurred within one year of initial remission in patients with a higher risk of CNS relapse. We then analyzed 7 patients with systemic lymphoma without CNS involvement but with high risk for CNS relapse to see if CSF ctDNA could detect the signal of CNS disease. Much to our surprise, mutations in CSF ctDNA were detected in 6 of 7 cases. KMT2D (28.6%) and TET2 (28.6%) were the most frequently mutated genes (Fig. 5). Notably, in one case (P012) we detected MYD88 V172F mutation in both CSF and Plasma. All the mutations detected in the CSF ctDNA were showed in the Table 2. These results indicate that ctDNA analysis of the CSF can improve the detection CNS involvement by conventional techniques.

Table 2

**Gene mutations identified in CSF of patients with high risk CNS involvement lymphoma** AF: allele fractions
ID	Gene	Mutations	AF(%)
P005	BTG1	p.S137N	6.25
P005	CREBBP	p.A222D	3.6
P005	IDH2	p.G173V	2.29
P005	KMT2C	p.T820I	3.72
P005	KMT2D	p.E446D	1.58
P005	KMT2D	p.A4513S	2.82
P005	SYK	p.R22W	4.24
P005	TSC2	p.A971S	8.75
P006	KMT2A	p.E883X	12.12
P008	MYD88	p.V172F	5.97
P010	KMT2D	p.E830G	8.51
P010	KMT2D	p.S837P	9.62
P010	KMT2D	p.P2955S	17.5
P013	SPEN	p.P3317S	4.42
P018	KMT2A	p.D2680E	37.5
P018	STAT5B	p.P129L	33.33

CSF ctDNA can be used to assess therapeutic response

To determine whether the CSF ctDNA could be a useful tool for evaluating the therapy response, we simultaneously analyzed sequential samples of CSF and plasma. Both blood and CSF samples were collected before starting treatment, or when progression or relapse of the disease was noticed. Tumor burden was evaluated by imaging examination.

Two individuals in our cohort (P004 and P012), we observed the CSF ctDNA levels decreased with response to treatment and undetectable after complete remission. In P004, following a disease-free interval of 11 months, the patient relapsed and ctDNA reappeared in CSF but not in plasma (Figs. 7 and 8). The P012 showed recurrence of the lesion after a 4-month interval, ctDNA was detected in both plasma and CSF (Figs. 8 and 9). These results are in line with what the imaging findings.

Moreover, We also observed that the increase in the CSF ctDNA level can precede the detection of relapses by imaging evaluation and clinical manifestations. Another case of high risk CNS involvement lymphoma (P013) treatment with Orelabrutinib + R-CEOP/R-GEMOX achieved a CR, although the CSF ctDNA levels remained elevated. The P013 patient's CSF GNA13 gene mutation abundance increased to 6.76% at the third review, and the imaging examination and clinical manifestations results suggested disease recurrence (Figs. 10 and 11). This finding may indicate that rigorous clinical follow-up should be done to monitor disease progression and recurrence in such patients.

B-cell lymphoma can infiltrate the CNS with a poor prognosis, especially at relapse [8]. Diagnosing and monitoring can be challenging in CNS lymphoma. MYD88 L265P mutation strongly suggests a diagnosis of CNS lymphoma. Many previous studies have confirmed this viewpoint [9–11]. In CSF ctDNA from patients with CNS lymphoma, the MYD88 L265P mutation was found, demonstrating that the CSF ctDNA may be employed as a complementary diagnostic tool. ctDNA was detectable in almost all cerebrospinal fluid specimens in this study and demonstrated that CSF ctDNA has the potential to improve both current diagnostics and therapeutic response monitoring in CNS lymphoma.

In terms of CNS lymphoma, CSF is in intimate contact with tumor cells and tumor cells release ctDNA fragments shed into the CSF directly. The blood-brain barrier could prevent ctDNA from entering the circulation, resulting in the low detectability of ctDNA in plasma [12]. Moreover, a significant quantity of free DNA from normal tissue exists in peripheral blood, and ctDNA accounts for a relatively small proportion of the test, which can be easily interfered [13]. It has been published that the CSF ctDNA could more comprehensively represent the genomic alterations of intracranial lesions in patients with CNS tumors, to help guide clinical targeted treatment [14]. In our study, we observed that the detection rate and gene mutation abundance of cerebrospinal fluid were higher than plasma. Although the sample size of this study was small, this trend was consistent with the results of previous studies [15–16]. Therefore, CSF sampling is a more invasive method of fluid biopsy than a simple blood test but is still necessary to detect CNS lymphoma by CSF ctDNA.

In our research, lymphoma-related gene mutations could be detected by CSF in the high-risk group without CNS involvement in imaging. This finding suggests that CSF ctDNA testing may may be able to identify molecular level changes earlier than imaging in the high-risk group. According to three separate investigations, serial ctDNA monitoring of DLBCL patients in CR helps the diagnosis of lymphoma relapse in the vast majority of cases, with a ~ 3–6 month lead time prior to radiographic imaging [17–19]. Imaging studies are often noninformative or slow to reflect progression and provide no genetic information [5]. The short half-life of ctDNA (< 1.5 hours) makes it possible to deliver more dynamic and real-time data on tumor progression [20]. Patients are exposed to radiation from repeated imaging tests, while ctDNA monitoring is noninvasive. Beyond that, a previous study has confirmed that cancer containing about 50 million malignant releases sufficient DNA for detection in the circulation, whereas cancer of that size is far below that required for definitive imaging [21]. A better sensitivity of CSF ctDNA (100%) than imaging examination (62.86%) was also demonstrated by Zhao Y et al [5]. Thus, CSF ctDNA as liquid biopsy technology can be used as a supplement to diagnose CNS relapse with negative imaging and is helpful for the elaboration of therapeutic protocols.

Monitoring the response to therapeutic treatments remains a challenging question in CNS lymphoma. Tumor biomarkers and imaging are widely used to assess the effectiveness of chemotherapy treatment. However, these methods may not be very accurate in evaluating therapeutic responses and do not provide information about genetic alterations in tumors [13]. Six patients with brain tumors had their CSF-ctDNA examined in studies by Mattos-Arruda et al., and they discovered that the mutant allelic frequencies (MAFs) of DNA decreases with surgical resection or response to systemic therapy and increased with the tumor progressed [22]. Similarly, our study found that CSF ctDNA is a promising biomarker. Dynamic detection of gene mutations in CSF may be used to assess the therapeutic response in real-time and guide the choice of the therapeutic regimen.

Taken together, CSF ctDNA, as a liquid biopsy technique, has open a new avenue for CNS lymphoma identification in biomarker research. This technology is valuable for the diagnosis, treatment and prognosis of CNS lymphoma. With the technological improvements that can be foreseen in the near future, the various dimensions of ctDNA as an analyte will be broadly explored.

In conclusion, CSF is a better source of ctDNA than plasma, and can be used to assess therapeutic response.

Financial disclose There are no financial conflicts of interest to disclose.

Conflicts of interest The authors declares that they has no conflict of interest.

Funding This work was financially supported by the the National Natural Science Foundation of China (82274268), the Fujian Science and Technology Department social development projects (2022Y0040), the Joint Funds for the Innovation of Science and Technology of Fujian Province (2018Y9033; 2019Y9069), the Natural Science Foundation of Fujian Province (2022J01732), the Traditional Chinese Medicine Scientific Research Project (202169-003), and the Fujian University of Traditional Chinese Medicine Key Disciplines Construction Project (XJG2023004).

Olszewski AJ, Chorzalska AD, Petersen M, et al. Detection of clonotypic DNA in the cerebrospinal fluid as a marker of central nervous system invasion in lymphoma. Blood Adv. 2021;5(24):5525-5535. https://doi.org/10.1182/bloodadvances.2021004512
Schmitz N, Zeynalova S, Nickelsen M, et al. CNS International Prognostic Index: A Risk Model for CNS Relapse in Patients With Diffuse Large B-Cell Lymphoma Treated With R-CHOP. J Clin Oncol. 2016;34(26):3150-3156. https://doi.org/10.1200/JCO.2015.65.6520
Li H. A statistical framework for SNP calling, mutation discovery, association mapping and population genetical parameter estimation from sequencing data. Bioinformatics. 2011;27(21):2987-2993. https://doi.org/10.1093/bioinformatics/btr509
Cirillo M, Craig AFM, Borchmann S, Kurtz DM. Liquid biopsy in lymphoma: molecular methods and clinical applications. Cancer Treat Rev. 2020;91:102106. https://doi.org/10.1016/j.ctrv.2020.102106.
Zhao Y, He JY, Zou YL, et al. Evaluating the cerebrospinal fluid ctDNA detection by next-generation sequencing in the diagnosis of meningeal Carcinomatosis. BMC Neurol. 2019;19(1):331. Published 2019 Dec 19. https://doi.org/10.1186/s12883-019-1554-5
Lauer EM, Mutter J, Scherer F. Circulating tumor DNA in B-cell lymphoma: technical advances, clinical applications, and perspectives for translational research. Leukemia. 2022;36(9):2151-2164. https://doi.org/10.1038/s41375-022-01618-w
Fu H, Zhou H, Qiu Y, et al. SEPT6_TRIM33 Gene Fusion and Mutated TP53 Pathway Associate With Unfavorable Prognosis in Patients With B-Cell Lymphomas. Front Oncol. 2021;11:765544. Published 2021 Dec 1. https://doi.org/10.3389/fonc.2021.765544
Yamamoto W, Tomita N, Watanabe R, et al. Central nervous system involvement in diffuse large B-cell lymphoma. Eur J Haematol. 2010;85(1):6-10. https://doi.org/10.1111/j.1600-0609.2010.01438.x
Hickmann AK, Frick M, Hadaschik D, et al. Molecular tumor analysis and liquid biopsy: a feasibility investigation analyzing circulating tumor DNA in patients with central nervous system lymphomas. BMC Cancer. 2019;19(1):192. Published 2019 Mar 1. https://doi.org/10.1186/s12885-019-5394-x
Hiemcke-Jiwa LS, Minnema MC, Radersma-van Loon JH, et al. The use of droplet digital PCR in liquid biopsies: A highly sensitive technique for MYD88 p.(L265P) detection in cerebrospinal fluid. Hematol Oncol. 2018;36(2):429-435. https://doi.org/10.1002/hon.2489
Escudero L, Martínez-Ricarte F, Seoane J. ctDNA-Based Liquid Biopsy of Cerebrospinal Fluid in Brain Cancer. Cancers (Basel). 2021;13(9):1989. Published 2021 Apr 21. https://doi.org/10.3390/cancers13091989
Bettegowda C, Sausen M, Leary RJ, et al. Detection of circulating tumor DNA in early- and late-stage human malignancies. Sci Transl Med. 2014;6(224):224ra24. https://doi.org/10.1126/scitranslmed.3007094
Yan W, Xu T, Zhu H, Yu J. Clinical Applications of Cerebrospinal Fluid Circulating Tumor DNA as a Liquid Biopsy for Central Nervous System Tumors. Onco Targets Ther. 2020;13:719-731. Published 2020 Jan 23. https://doi.org/10.2147/OTT.S229562
Ma C, Yang X, Xing W, Yu H, Si T, Guo Z. Detection of circulating tumor DNA from non-small cell lung cancer brain metastasis in cerebrospinal fluid samples. Thorac Cancer. 2020;11(3):588-593. https://doi.org/10.1111/1759-7714.13300
De Mattos-Arruda L, Mayor R, Ng CKY, et al. Cerebrospinal fluid-derived circulating tumour DNA better represents the genomic alterations of brain tumours than plasma. Nat Commun. 2015;6:8839. Published 2015 Nov 10. https://doi.org/10.1038/ncomms9839
Zorofchian S, Iqbal F, Rao M, Aung PP, Esquenazi Y, Ballester LY. Circulating tumour DNA, microRNA and metabolites in cerebrospinal fluid as biomarkers for central nervous system malignancies. J Clin Pathol. 2019;72(4):271-280. https://doi.org/10.1136/jclinpath-2018-205414
Scherer F, Kurtz DM, Newman AM, et al. Distinct biological subtypes and patterns of genome evolution in lymphoma revealed by circulating tumor DNA. Sci Transl Med. 2016;8(364):364ra155. https://doi.org/10.1126/scitranslmed.aai8545
Roschewski M, Dunleavy K, Pittaluga S, et al. Circulating tumour DNA and CT monitoring in patients with untreated diffuse large B-cell lymphoma: a correlative biomarker study [published correction appears in Lancet Oncol. 2015 May;16(5):e199]. Lancet Oncol. 2015;16(5):541-549. https://doi.org/10.1016/S1470-2045(15)70106-3
Kurtz DM, Green MR, Bratman SV, et al. Noninvasive monitoring of diffuse large B-cell lymphoma by immunoglobulin high-throughput sequencing. Blood. 2015;125(24):3679-3687. https://doi.org/10.1182/blood-2015-03-635169
Diehl F, Schmidt K, Choti MA, et al. Circulating mutant DNA to assess tumor dynamics. Nat Med. 2008;14(9):985-990. https://doi.org/10.1038/nm.1789
Diaz LA Jr, Williams RT, Wu J, et al. The molecular evolution of acquired resistance to targeted EGFR blockade in colorectal cancers. Nature. 2012;486(7404):537-540. https://doi.org/10.1038/nature11219
De Mattos-Arruda L, Mayor R, Ng CKY, et al. Cerebrospinal fluid-derived circulating tumour DNA better represents the genomic alterations of brain tumours than plasma. Nat Commun. 2015;6:8839. Published 2015 Nov 10. https://doi.org/10.1038/ncomms9839

No competing interests reported.

TableS1.docx
Table S1 Unique mutations identified in CSF
stable.docx
Table S2 Gene function and clinical significance

Detection of genes mutations in cerebrospinal fluid circulating tumor DNA from central nervous system lymphomas patients using next generation sequencing

Status:

Version 1

Abstract

Figures

Introduction

Materials and methods

Patients

DNA extraction, library preparation and target sequencing

Bioinformatics analysis

Statistical analysis

Results

Baseline Characteristics of Patients

CSF is superior in identifying gene mutations of CNS lymphoma than plasma

CSF ctDNA can be used to assess therapeutic response

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1