The availability of a minimally invasive patient simple, capable of providing tumor information, represents a valuable clinical tool. The liquid biopsy has the potential to achieve this need. Circulating cell free DNA (ccfDNA), other circulating nucleic acids such as microRNA and circulating tumor cells (CTCs), can be obtained from a peripheral blood sample. Liquid biopsy has been particularly studied in solid tumors, specially of the epitelial origin, such as a pancreatic carcinoma and advanced breast cancer. It has been considerably less applied to the study of non-solid tumors. It represents an important source for diagnosis, prognosis and predictive information. Also it is suitable to evaluate response to therapy and drugs pharmacokinetics. It provides a unique opportunity to evaluate the disease evolution in serial blood samples collection, otherwise difficult to obtain. Liquid biopsy can be rehearsed using different circulating biological fluids such as whole blood, serum, plasma and lymph, as well as, non-circulating fluids such as urine, feces, saliva, bile and accumulated pathological fluids such as ascites. This review summarizes the current status of circulating material analysis in non-solid tunors. It is specially focused on Hodgkin Lymphoma and among Non-Hodgkin Lymphoma, it refers particularly to Diffuse Large B cell Lymphoma, the most common aggressive Non-Hodgkin Lymphoma derived from germinal center B-cells in adults. It further discusses the benefit of liquid biopsy in oncohemtaological diseases and potential clinical applications.
Liquid biopsy represents a late development of an antique discovery. The first report on CTCs (Circulating Tumor Cells) was from 1869 when an Australian physician described a case in which cells, similar to those in the tumor, were seen in blood following patient’s death [
The number of CTCs in the blood stream is considerably low. It is accepted that there might be 1 CTC/109 cells in a patient blood sample with a metastatic tu- mor. The studies on CTCs are aimed to evaluate the risk of relapse in an onco- logical patient, to select treatment and monitor response to therapy, to identify novel therapeutic targets, to detect driver mutations that might be druggable if known, to evaluate pharmacodynamic effects, to analyze resistance mechanisms or to understand molecular basis of tumor metastasis dissemination [
Different methods, based on the physical and molecular CTCs features, have been recently developed and some of them have been enumerated in
The first CTC-chip micro-spots array, containing antibody anti- EpCAM(Epithelial Cell Adhesión Molecule), was initially used with metastatic
Method | Developer | Antibodies |
---|---|---|
Immunomagnetic assays: Positive Selection | ||
Cell Search | Janssen Diagnostics | EpCAM-coated beads |
Adna Test | AdnaGen (Langenhagen, Germany) | EpCAM, MUC-1 Mesothelin, molecular profiling |
MACS Cell Separation technology | Miltenyi Bioteh (BergischGLadbach, Germany) | EpCAM, pan-CK |
Dynabeads | Life Technologies | Monoclonal antibody BerEP4 |
Cell Collector | Gilupi | Medical wire anti-EpCAM coated placed into the patient vein blood steam for 30 minutes. |
Biofluidic CTC Detection System | Biofluidica | EpCAM coated Chip |
Immunomagnetic assays: Negative Selection | ||
CTC-Chip CTC-iChip | Daniel Haber and Mehmer Toner at Dana-Farber and Massachusetts General Hospital | EpCAM and CD45/citokeratin substraction |
Based on CTCs biological and molecular features | ||
Size | ||
Size and deformability | ||
Density | ||
Physical properties | ||
Others |
Different methods have been tried to improve the isolation of the very low percentage of circulating tumor cells in plasma, serum or whole blood simple.
tumors including Non-Small Cell Lung Cancer (NSCLC), pancreatic cancer, breast carci- noma, colon cancer, localized and metastatic prostate carinoma and healthy do- nors. The number of CTCs/ml isolated was in average 155, 16 to 292, 25 to 174, 9 to 831, 5 to 176, 42 to 375 and 0 respectively [
Vimentin is usually consider a mesenchymal cell lineage marker. It has been re- ported that cell surface vimentin, can be used as an universal marker to identify CTCs from metastatic sarcomas, using a monoclonal antibody. This antibody was proven to be specific for vimentin bound to cancer cell surface. It did not show any binding toward macrophages, endothelial cells, neutrophils, platelets or apopto- tic T lymphocytes [
Extracellular DNA embraces both nuclear and mitocondrial DNA released from cells. This genetic material can be found in circulating fluids (whole blood, plas- ma, serum, lymph) non-circulating extracellular products (bile, milk, feces, urine, saliva, mucous suspension, spinal fluid, amniotic fluid) and liquid accumulated du- ring certain pathological conditions (ascites, among others). Extracellular DNA mo- lecules are found in humans, and both the animal and plant kingdoms. Also, extra- cellular DNA, can be isolated from the supernatant of tissue cultures [
The cirulating cell free DNA (ccfDNA) does not come from circulating cancer cells, considering the low number of CTCs and the amount of ccfDNA. We all have ccfDNA in physiological conditions, coming from tissues with a high cellular turn over rate, such as the bone marrow, the intestinal epithelium or the fetus during pregnancy. The mean ccfDNA concentration in healthy individuals is 2 - 5 ng/ml. An enhanced amount can be isolate from individuals with chronic inflammatory diseases. Even higher, is the ccfDNA level in patients with cancer, ranging from 10 to 1000 ng/ml. Only a low percentage of the total amount of ccfDNA corre- sponds to tumor cells. Circulating tumor DNA (ctDNA) has biochemical and bio- physical different properties as compared to ccfDNA arising from non-tumor cells, such as the double strand fragments being less stable and bearing a different GC- composition [
It has been stablished that nucleosomes positioning varies among different cell types [
There are different approaches to characterize this ccfDNA such as PCR-based techniques, digital droplet PCR and BEAMing (all of them aimed to assess muta- tions and fusion genes previously known) and massively parallel sequencing with distinct sensitivity (focused on the novo genetic alterations discovery) [
Beyond DNA, RNA and CTCs, other material can be also obtained from blood, among them we found microvesicles, exosomes, ectosomes and apoptotic bo- dies. Regarding blood ccfDNA, it has to be considered the neutrophils and eosi- nophils DNA traps, circulating DNA linked with serum proteins and cell mem- brane surface [
Exosomes originate from the membrane invagination of a subset of late endo- somes, which ends up containing a large number of small vesicles, taking the name of multivesicular bodies. Exosomes are characterized by small dimensions (30 - 100 nm) and round shape. Instead, microvesicles bud from the plasma mem- brane and are characterized by a wider size (from 100 nm to 1000 nm) and less re- gular morphology [
Although extensive data related to liquid biopsy has been lately generated, there is still lack of consensus and applied asssays standardization, to reproduce results. Also, it seems necessary to design protocols including larger number of pa- tients. Liquid biopsy has been extensively explored in patients with solid tumors, that is not being reviewed herein. Its potential benefits have been less evaluated in non-solid tumors and among these, the least investigation, has been perfor- med in pediatric oncohematological pathologies. A summary of the clinical evi- dence based utility, for the distinct circulating elements performance as potential biomarkers, provided by liquid biopsy, can be found in
As in solid tumor patients, non-solid oncopathology patients show higher ccf- DNA as compared with healthy subjects. A 2009 screening in both Hodgkin (HL) and Non-Hodgkin Lymphoma (NHL) patients found that they had a ccfDNA median concentration of 24.1 ng/ml at diagnosis, which doubled the healthy con- trol group ccfDNA median concentration. After two-years of follow up, 88% of NHL and 95% of HL patients, who developed complete response to therapy, showed a significant decrease in ccfDNA levels. ccfDNA at diagnosis correlated with serum lactate dehidrogenase (LDH) levels, tumor burden and prognosis, in terms of DFS [
Nucleic Acid Type | Clinical application | Evidence level | |
---|---|---|---|
CTCs | Prognosis | I Metastatic Breast Cancer II-2 Prostate carcinoma | |
Exosomes | Prognosis | III | |
Circulating Nucleic Acids | cfDNA | Prognosis | II-2 |
cfRNA | Therapy response | III | |
miRNA | Prognosis | Pre-clinical |
US preventative services task force.
of patients into low, intermediate-low, intermediate-high and high risk [
Diffuse Large B Cell Lymphoma (DLBCL)
B cell Non-Hodgkin Lymphoma comprises a wide range of genetically, phe- notipically and clinically distinct lymphomas. Diffuse Large B Cell Lymphoma (DLBCL), Burkitt (BL) and Follicular Lymphomas (FL), account for approxima- telly 80% of all NHL, being DLBCL the most frequent in adults [
Recently, liquid biopsy has been proven as an excellent source to genotyping DLBCL, in real time. Also it showed to be as accurately as doing it in a conven- tional biopsy. Furthermore, liquid biopsy has allowed to follow up the lympho- ma evolution and patient response to treatment, in a non invasive manner plus, uncovering novel genetic alterations. This last achievement, overcomes the hete- rogeneity of this B-cell lymphoma. Ultra-deep next generation sequencing of plasma ccfDNA, in newly diagnosed DLBCL patients before treatment, showed the known mutations associated with this lymphoid disease. It was also feasible to uncover in ccfDNA, mutations that were detected during a conventional bio- psy, due to the intrinsic heterogeneity of this B-cell lymphoma. Patients that re- sponded to the R-CHOP treatment scheme, showed clearence of the alterations found in the diagnosis sample. On the other hand, resistant R-CHOP patients, remained harboring the same mutations meanwhile acquaring new ones, as a re- sult of resistant clone selection. Camus, V. et al., developed digital PCR assays for the detection of recurrent mutations in ccfDNA DLBCL patients, focusing on those ones that woud allow patients to benefit from an available targeted therapy such as EZH2 and MYD88 mutations [
CCND2 and MYC mRNA levels were found in 14% and 10% respectively, in a study that included 42 DLBCL patients, and they correlated with worst prog- nosis in terms of shorter OS. Others mRNA levels also studied were BCL2, LMO2 and BCL6, which were found in a less percentage of patients in the studied cohort, and no association was found with OS. None of them were associated with DFS [
Roschewski M. et al., show that ctDNA is reliable to monitor disease status in DLBCL patients. They demonstrated a positive association between ctDNA and disease progression. They studied a group of 107 DLBCL patients who achieved complete remission at the end of treatment. Overall, 17 patients relapsed, during the follow-up and 16 of them developed detectable ccfDNA before clinical evi- dence of disease progression [
Relapse in DLBCL patients, most likely is due to the persistance of minimal residue disease (MRD), bellow the detection of imaging. The possibility of ana- lizing circulating material would largely improve this current situation.
Central Nervous System (CNS) relapse remains as an important issue in DLB- CL patients. This needs better understanding in order to identify the subgroup of patients at high risk, so that they require effective targetting of CNS prophy- laxis. This could be one of the clinical unsolved situations to explore in liquid bio- psy since no biomarkers are available.
So far, it is known that patients with very aggressive lymhomas such as BL or Lymphoblastyc Limphomas, have to routinely receive CNS prophylaxis due to high relapse rate, meanwhile patients with indolent lymphomas, such as Mar- ginal Lymphoma or FL, do not need it. However, it remains controversial whe- ther DLBCL patients, the most common germinal center (GC) derived B Cell lym- phoma in adults, shoud receive CNS prophylaxis. The research on this topic is extensive but still unconclusive [
CXCR4, CXCR5, CCR7, CXCL12, and CXCL13 expression have been eva- luated by inmunohistochemistry on 89 tissue samples of patients with CNS pri- mary and secondary lymphomas, as well as systemic DLBCL patients, in order to analyze lymphocytes CNS tropism. Interestingly, strong nuclear CXCR4 positi- vity correlated with systemic DLBCL, whereas strong cytoplasmic CXCR5 positi- vity, correlated with CNS involvement [
Regarding other complications, neutropenic fever is common in hematologi- cal patients, as a consecuense of aggressive chemotherapy schemes. Furthermore, these patients can take a course for the worst, developing serious infections such as sepsis. A study that included 100 oncohematological patients, showed that ccf- DNA levels measurements did not predict late complications in patients with neutropenic fever, such as sepsis or septic shock. Interestingly, if hematological malignancies subgroups were analyzed, differences became more apparent. Acute Myeloid Leukemia (AML) patients had lower ccfDNA levels than lymphoma ones. The ratio ccfDNA/leucocytes among these two groups of patients, successfully identified those ones that would develop sepsis or septic shock. This promising result deserves deeper investigation [
As referred before, while circulating genetic material has been widely investi- gated in patients with solid tumors, there are few reports on liquid or non-solid tumors. Among the non-solid tumors, there has been particularly less focus on the ccfDNA evaluation, in pediatric oncohematological diseases. Machado et al., studied total ccfDNA and EVB plasma DNA in 30 pediatric B cell NHLs. That group concluded that the analysis of ccfDNA could be useful for disease detection in pediatric EVB-associated lymphomas [
Mussolin, L. et al., studied ccfDNA in plasma of 201 italian cases of pediatric lymphomas (43 cases of HL, 45 of ALCL, 88 BL, 17 lymphoblastic and 8 DLB- CL). The median ccfDNA was 1.6 ng/ml in the healthy control group and 46 ng/ml in the lymphoma group of patients applying a real-time quantitative PCR for the POLR2 gene [
Interestingly, the group of Mussolin et al., found that the fraction of tumor DNA was higher in patients that had lower amount of total ccfDNA. Although BL patients showed the highest level of total ccfDNA, no association was found between plasma circulating DNA and lymphoma histoloy, B-symptoms, LDH levels, bulky disease or MDD. The 3-years follow-up period showed that total le- vel of ccfDNA at diagnosis, had prognosis value in terms of DFS, since it was shorter in the group of patients that had a ccfDNA concentration over 46 ng/ml [
Somatically mutated inmunoglobulin genes represent an unique marker for GC derived B cell lymphomas [
Immunoglobulin genes high-throughput sequencing, particularly the D (Di- versity) and J (Joining) genes joint, provides a unique DNA clonotype shared by all malignant B cells. This method has been used for diverse B cell malignancies including DLBCL cases [
Most of the studies published on liquid biopsy have isolated ccfDNA from plasma. It has been reported that it seems more convenient to obtain circulating DNA from plasma as compared with serum (
DLBCL, a heterogeneous B-cell malignancy, has been classified in two main mo- lecular subtypes according to gene expression profile: ABC-Like (Activated B Cell) and GCB-Like (Germinal Center B cell) based on the cell of origin [
All these reports even though, they have been usually performed in a small number of patients, are promising, specially considering that it is believed that
ccfDNA | Serum | Plasma |
---|---|---|
Quantity | > | < |
Quality | < | > |
Purity | < | > |
Detection sensibility | < | > |
CTCs | ccfDNA | |
---|---|---|
Equipment | Need of specializad equipment | No need |
Isolation | Complex | Standard DNA preparation |
Whole Genome Amplication | Yes | No |
Heterogeneity | yes | Results represent an average |
Specific markers for isolation | Dependent | No dependent |
MTS | Ideal material | No |
Diagnosis and monitoring | Stablished by clinical studies | Clinical studies pending |
only a few DLBCL patients have evidence of circulating disease [
Epigenetic modifications have also been explored in liquid biopsy as potential DLBCL biomarkers. DAPK1, DBC1, MIR34A, and MIR34B/C promoters aberrant methylation, determined by pyrosequencing, was identified in plasma ccfDNA of DLBCL patients at diagnosis. During R-CHOP treatment, patients that were sen- sitive and responsive, lost of DAPK1 promoter hypermethylation occured. On the other side, patients that did not respond, mantained or regained promoters aberrant methylation and had a poor prognosis, in terms of OS [
Hodgkin Lymphoma (HL) is one of the most common B cell lymphoma derived from the GC. Nowadays, all these patients are treated with the same chemo- therapy schedule, which combines four citotoxic drugs. Chemosensitive patients could even reach the cure. Nevertheless, the probability of relapse is 10%, and commonly take place during the first twelve months following diagnose.
The initially refractory patients, despite different rescue stratagies, remain unresponsive although improvements in adjuvant therapy (conjugated mono- clonal antibodies plus high chemotherapy and autologous bone marrow trans- plant), have been made [
In this regard it has been reported that, the E571K Exportin 1 (XPO1) mu- tation assessed by digital PCR in ccfDNA, could have prognosis value in cHL (classical Hodgkin Lymphoma) patients. The four histological subtypes of cHL account for the 95% of patients diagnosed with HL. The presence of this XPO1 mutation in plasma DNA cHL patients after treatment, correlated with shorter DFS [
It has been investigated the role of cell free circulating microRNAs (ccfmicro- RNA) in cHL patient plasma. Higher levels of ccfmiR-494, miR-1973 and miR- 21, were detected in cHL patients as compared to healthy subjects. Levels of these three microRNAs decreased when patients developed a complete response to che- motherapy. Only fluctions of miR-494 and miR-1973 levels, reflected response during therapy, with reduction being more pronunce in patients achiving com- plete versus partial response [
Novel and interesting results were obtained applying a standardized size-exclu- sion chromatography method, to isolate vesicle microRNA from plasma cHL pa- tients. Sequenced RNA candidates were validated by quantitative RT-PCR. The isolated vesicles from plasma HL patients at diagnosis before treatment were enriched for miR-24-3p, miR-127-3p, miR-21-5p, miR-155-5p and let7a-5p as compared to vesicles isolated from healthy controls. Those vesicles microRNAs levels significantly decreced or even disappeared after treatment in patients that developed complete response. If a responsive patient later relapsed, this was accom- panied by an enhanced microRNA level. Remarkably, there was a correlation between plasma circulating microRNAs and the patient status, determined by the fluorodeoxyglucose-PET before, during and once treatment was completed [
The evaluation of a large cohort of pediatric HL patients, showed higher ccf- DNA levels as compared to controls. The follow-up showed that an enhance- ment in ccfDNA after treatment correlated with worse prognosis [
This review has specially focused on Hodgkin and Non-Hodgkin Lymphomas, since other hematological diseases including leukemias and multiple mieloma (MM), have been less prominent as a subject of diagnostic, prognostic or predictive biomarkers screening in liquid biopsy.
Extracellular vesicle microRNA-155 has been analyzed as a potential diagnos- tic biomarker in several different hematological malignancies. The extracellular vesicles were isolated by differential centrifugation and its quality verified by atomic force and transmission electron microscopy. The microRNA was analy- zed by quantitative RT-PCR. High microRNA-155 levels were detected in acute myeloid leukemia (AML), chronic lymphocytic leukemia (CLL) and Waldens- trom’s Macroglobulinemia (WM) as compared to the group of healthy controls. microRNA-155 levels were significantly lower in multiple mieloma (MM), BL, FL and DLBCL [
N-ras point mutations have been determined by PCR in plasma ccfDNA, peripheral blood cells and bone marrow samples of acute myelogenous leukemia patients (AML). The study showed that the spectrum of alterations determined in plasma ccfDNA was wider as compared to the finding observed in the bone marrow samples [
Similar to the results obtained from ccfDNA in AML patients, the analysis of mutations in paired ccfDNA and bone marrow biopsy DNA in MM patients, showed that more information is obtained from the first sample. NRAS, KRAS, BRAF and p53 mutations were analyzed in newly diagnosed and relapsed/ refractory MM patients showing higher frequency of mutations in the first group of patients. They observed that activating RAS mutation, was higly prevalent in MM patients, as it has been previously described. This work showed that ccfDNA analysis improved the mutational pattern characterization in MM patients as well as allowed monitoring response to therapy [
In a recently published pilot study, the myeloma VDJ rearrangement was investigated by next generation sequencing in ccfDNA and circulating myeloma cells. Therapy responder patients, evidenced a prompt clearence of the VDJ mye- loma rearrangement, meanwhile it persisted in refractory patients after treat- ment [
Long non-coding RNAs (lncRNA), are the least circulating nucleic acids, inves- tigasted as potential biomarkers in cancer; including both solid and nonsolid tumors. In any case, further studies are necessary to elucídate their role in onco- pathology.
A study of plasma lncRNAs, including TUG1, LincRNA-p21, MALAT1, HOTAR and GAS5 by real-time PCR, in Chronic Lymphocytic Leukemia (CLL) and MM patients was carried out. It was only observed significantly higher LincRNA-p21 levels in the CLL group of patients as compared to healthy control individuals. Regarding MM patients, they found an enhanced TUG1 level [
Liquid biopsy present the unique opportunity to examine th disease biology at diagnosis and to monitor the disease as it evolves over time, minimizing patient discomfort. There is a deep need of moving forward after the biomarker discovery, to the description of its performance including the device that is used, the assay that is being carried out, the software applied for results analysis and the range of conditions under which the measurement gives reproducible and accurate results. Then, there should be trials from phase 1 to phase 3, designed to demonstrate that the candidate biomarker is “fit for the purpose”, for a specific context of use. To start it is necessary to examine the frequency at which the biomarker is present in the illness for which is considered a candidate and the group of patients that migh potentially benefit from its determination. Lastly, it has to be showed that the test result, has an impact on the patient management and therapecutic secheme selection.
Liquid biopsy promises to contribute better to patient prognosis, both before and after treatment, as compared to diagnosis. Despite the terrific technological advancement, the analytical validation, clinical validation and clinical utility for the ccfDNA and CTCs analysis methods need to be standardized.
The study of tumor genetic alterations has traditionally been performed on tissue biopsy material. However, many tumors are difficult to biopsy, for various reasons. Tumor localization could be accessible only using fine needle aspirates, obtaining as a consequence a quite small sample of usable material, just enough to approch a diagnosis. In this context it might be difficult ore ven imposible to move forward to molecular analysis. Also, tumors can be located either in unknown sites. There is a 10% of patients in whom, the onset of the oncology disease, occurs with metastatic lesions in multiple sites bearing an unknown primary tumor. Evenmore, the tumor couls be located in challenging and risky sited to be successfully accessed such as it occurs in CNS. The traditional methods for genetic alterations tumor assessment, fail to capture the heteroge- neity of the disease, especially during its progression. The Liquid Biopsy can profoundly improve the way clinicians face and respond to these issues. The Liquid biopsy, with a simple blood drawn, can simultaneously sample different sites compromised by the disease or distinct primary tumor sections, finally building a “fluid” picture of the tumor cell alterations. Liquid biopsy can be re- peated as needed, specially for effective monitoring of the disease evolution and the patient response to therapy. As a detailed picture of the mutations that drive onco- genesis emerges, new avenues for researchers, clinicians, and drug developers are currently being opened up.
Ranuncolo, S.M. (2017) Liquid Biopsy in Liquid Tumors. Journal of Cancer Therapy, 8, 302-320. https://doi.org/10.4236/jct.2017.83026