Development and evaluation of a pre-analytical device for liquid biopsy

(1)

UPTEC X 18 024

Examensarbete 30 hp

Juni 2018

Development and evaluation of

a pre-analytical device for liquid

biopsy

(2)

(3)

Teknisk- naturvetenskaplig fakultet UTH-enheten Besöksadress: Ångströmlaboratoriet Lägerhyddsvägen 1 Hus 4, Plan 0 Postadress: Box 536 751 21 Uppsala Telefon: 018 – 471 30 03 Telefax: 018 – 471 30 00 Hemsida: http://www.teknat.uu.se/student

Abstract

Development and evaluation of a pre-analytical device

for liquid biopsy

Nicole Yacoub

In this project, a novel method to discover and monitor cancer was studied. Tumour educated platelets (TEPs) have shown the ability to take up tumour-derived secreted membrane vesicles, which contain tumour RNA. Therefore they are of great interest for detecting cancer and could potentially work as biomarkers. One problem with using blood platelets as a source of biomarkers is that clinical implementation is prevented by pre-analytical procedures that are difficult to perform in clinical laboratories. In this study, human blood samples have been used to investigate whether density gradient centrifugation followed by microfiltration could result in pure platelet (PLT) fractions. The idea is to enable parallel extraction of multiple sources of biomarkers from a single liquid biopsy, namely peripheral blood mononuclear cells (PBMCs), PLTs and plasma. The procedure was done by first resuspending the buffy coat in plasma after the centrifugation, where some cells were fixed and then filtered through a combination of membranes. From that, PLT fractions free from any nucleated cell could be obtained both for fixed and unfixed cells. The next step will be to detect tumour-specific transcripts in PLTs by investigating whether the purity of the extracted mRNA is sufficient for these kinds of procedures. This method holds great promise for creating a procedure to isolate, extract and analyse several biomolecule fractions from a single liquid biopsy. Further studies could include other biomarkers of interest such as circulating tumour DNA (ctDNA), extracellular vesicles (EVs) and plasma proteins.

(4)

(5)

Populärvetenskaplig sammanfattning

(6)

(7)

Abbreviations

AO – Acridine Orange cfDNA – Cell Free DNA

CTCs – Circulating Tumour Cells ctDNA – Circulating Tumour DNA DNA – Deoxyribonucleic Acid

EDTA – Ethylenediaminetetraacetic acid EVs – Extracellular Vesicles

FDA – U.S. Food and Drug Administration miRNA – Micro RNA

mRNA – Messenger RNA MVs – Microvesicles

NIH – National Institutes of Health

PBMCs – Peripheral Blood Mononuclear Cells PBS – Phosphate Buffered Saline

PCR – Polymerase Chain Reaction PFA – Paraformaldehyde

PLTs – Platelets

qRT-PCR – Quantitative Reverse Transcriptase PCR RBCs – Red Blood Cells

RNA – Ribonucleic Acid

(9)

1 Introduction

Cancer is a disease characterized by uncontrolled growth and spread of abnormal cells. It is the leading cause of morbidity and mortality worldwide. The blood biomarker tests for early detection that currently exist for cancer have a tendency to incorrectly characterize healthy people as sick and sick people as healthy, which can cause e.g. mistreatment of the patient, high economical costs and stress (1). Every year over 8.2 million people die of cancer due to inaccessibility of the right detection methods and treatments. For long researchers have been trying to come up with different procedures that could help detect and monitor cancers. Many methods have been explored for screening, prognostication and monitoring cancer with only limited success (2). Therefore, there is a need for new and better diagnostic methods in cancer. Today tissue biopsy is the golden standard in cancer diagnostics for solid tumours because it can assist in targeted therapies and provide material for the genotyping of the cancer. However, with tissue biopsy only a single site of the tumour is investigated and only at a single point in time, which limits the assessment of the tumour for diagnosis.

In recent years, studies show that advances in the tumour biomarker field may result in more efficient ways to detect cancer earlier (3). The term biomarker can be defined as “a characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes or pharmacologic responses to a therapeutic intervention” according to the National Institutes of Health Biomarkers Definitions Working Group (4). There are biomarkers being used today in the clinical practice, but most of them do not show satisfactory clinical performance due to lack of sensitivity and specificity. Inappropriate statistical methods and study designs are part of the problem, which in turn also makes validation difficult. Therefore there are very few biomarkers that actually enter the market (5). Biomarker signatures, i.e. biomarkers used in combination with other biomarkers, have been widely recognized to have better clinical performance and yield better information than just single biomarkers (6). For that reason development of a novel liquid-biopsy method where a single liquid biopsy can yield several biomarkers that will be of help for detecting and monitoring cancer has come to interest us.

1.1 Purpose

(10)

obtaining separate fractions of plasma and PBMCs that also could be used for further studies. By this method, a single liquid biopsy can yield material for measurements of several biomarkers that will be of help for detecting and monitoring cancer.

2 Background

Platelets originate in the bone marrow and are derived from megakaryocytes. They are the second most abundant cell type in peripheral blood after red blood cells (RBCs). Whole blood from healthy individuals normally contains 200-500 million platelets per millilitre (7). During infection, cancer or bone marrow disease, the platelet populations can increase or decrease. Platelets do not contain a nucleus and are normally involved in haemostasis and inflammation. As anucleate cells they do not contain any genomic DNA, but recent studies shows that they regulate the body’s inflammatory and immune response mechanism through their endogenous RNA. The messenger RNA (mRNA) and microRNA (miRNA) profiles that exist in platelets have recently been shown to reflect disease and disease risk factors, rendering these promising for analysis of tumour biomarkers in liquid biopsy of cancer (9).

2.1 Tumour Educated Platelets

Platelets can take up derived secreted membrane vesicles, which can contain tumour-associated RNA. Such platelets are called tumour-educated platelets (TEPs), and they have been shown to have the potential to help detect and monitor cancer (10). Therefore, gene-expression analysis of isolated RNA, could provide biomarkers for cancer diagnostics (11). One problem is that the clinical implementation of platelet-based analyses is prevented by pre-analytical processing methods that are difficult to apply in clinical laboratories. When performing transcriptional studies, the low content of platelet mRNA is one of the greatest obstacles. In comparison to the RNA content of nucleated cells, platelet RNA content can range from 1:12,500 (12) to 1:100,000 (13) per cell. This means that platelet fractions must be purified from nucleated cells to not disturb the signal originating specifically from platelet RNA during analysis.

2.2 Liquid Biopsies

(11)

provides a very limited snapshot of the tumour. Tumour biopsy, because it is being costly and labour intense, prevents us from being able to do repeated sampling. This limits possibilities for follow-up examination of the patient to monitor the treatment response or development of resistance to a targeted therapy. As a minimally invasive method, liquid biopsies could potentially provide a revolution in cancer diagnostics for detecting and monitoring cancer diseases (10). It could provide e.g. prognosis for each individual, personalized treatment and most importantly a better chance for an early detection. But before liquid biopsies can be adopted into clinical practice, one major obstacle to overcome is sensitivity of detection. Since the molecules originating specifically from the tumour are present at low levels in the body fluids, there is a need for very high assay sensitivity and the amount of material collected for analysis must be sufficient for detection (14). Since liquid biopsies base detection on peripheral fluids rather than the tumour itself it is very difficult to obtain information to be able to trace the cellular sources back to the tumour. Also, even if it is possible to pick up heterogeneity in a primary tumour or metastasis it will be difficult to pinpoint which clones dominate which tumour site without maybe combining it with tissue biopsy procedures (15).

2.3 Circulating Tumour DNA

In 1948 the presence of cell-free DNA (cfDNA) in blood was discovered but it was not until 1994 that it could be used to identify tumour specific mutations (16). The cfDNA is shed into the bloodstream during apoptosis, necrosis and by active release, from both healthy and diseased cells (17). The finding of higher levels of cfDNA in cancer patients compared to healthy individuals interested researches in circulating tumour DNA (ctDNA) as a potential biomarker. The concentration of this free DNA in serum of patients with diverse types of cancer compared to healthy individuals had a mean value of 180 ng/ml and 13 ng/ml, respectively (18). However, detection has been more feasible in patients with advanced, metastatic disease compared to patients with localized cancer, indicating that ctDNA may be of limiting utility as an early stage detection biomarker (19). It is the high cell turnover rate of tumours that result in high release of ctDNA in the cancer patients. The ctDNA carries genetic information from the entire tumour genome and for that fact it can provide insight into clonal heterogeneity, which is shown by several studies. This could help understand the evolution of solid cancers present at any moment (20). It has been shown that ctDNA can be recovered through a minimally invasive method such as a liquid-biopsy (21).

2.4 Extracellular Vesicles

(12)

be done to interact with or modify the behaviour of a specific targeted cell at a close or distant site (23). It has also been shown that EVs are important carriers of cancer cell components that consist of proteins, mRNA, miRNA, DNA, lipids and other transcriptional factors (24). These have shown to mediate paracrine signalling in the tumour microenvironment, which contributes to the understanding of long-distance crosstalk between cancer cells and distant organs. Therefore, the use of EVs can have strong potential impact in liquid biopsies for cancer diagnosis (25).

2.5 Protein Biomarkers in Blood

During disease progression, protein biomarkers can serve multiple clinical purposes, both in early and late stages of cancer. Therefore, proteins attract great interest as potential biomarkers. Proteins are biological endpoints that control most biological processes and can be found either circulating in blood or expressed in tumour tissue (5). A biomarker must be measured reliably for it to be clinically useful. Proteins can be quantified efficiently with high analytical sensitivity. Carcinoembryonic antigen (CEA), carbohydrate antigen 19-9 (CA19-9) and cancer antigen 125 (CA125) are examples of biomarkers shown to be useful for monitoring cancer patients. The problem is that they either have too low sensitivity or specificity to be approved for screening purposes (26). Therefore, there are only 18 protein cancer biomarkers today that have been approved by the US. Food and Drug Administration (FDA). However, since well developed techniques for protein analyses have existed for a long time and the abilities to detect analytes at low concentrations have improved, further studies to improve assay sensitivity and efficiency may increase future possibilities to use protein biomarkers for screening. (5)

3 Material and Methods

3.1 Blood Collection and Blood Sample Processing

Blood was collected in 4 ml Vacutainer CPT Tubes (BD Biosciences, Cat. No. 362781) containing Ficoll Hypaque solution, sodium citrate and a polyester gel, from anonymous healthy blood donors at Uppsala University Hospital, Sweden.

(13)

3.2 Evaluation of Filter Setups

Next step aimed at capturing PBMCs and any RBCs that could be contaminating the sample, to produce a pure fraction containing only PLTs that can be further processed. The pure PLT fraction is then filtered in a second step to produce cell free plasma with Vivid Plasma Separation membrane (PALL Corporation).

By using filter combinations with different pore sizes, purification was carried out based on the knowledge of the greater diameter of WBCs and RBCs (7-25 µm and 7.5-8.5 µm respectively) compared to those of PLTs (2-4 µm). Previous work in the group had identified a stack of filters with different properties combined in a specific order that yields near complete separation of PBMCs from PLTs. A common membrane, Membrane A, was combined with either Membrane B or Membrane C depending on whether input samples were fixed or not. Membrane B (pore size 5 µm) had earlier been evaluated to perform well with fixed cells, therefore 3 ml of cell suspension was fixed in 300 µl of 4% (w/v) paraformaldehyde (PFA) (Merck, Cat. No. 104005) in 1X phosphate buffered saline (PBS) (Medicago), pH 7.4 for 15 min at room temperature. Similarly, Membrane C (pore size 2 µm) had been determined superior for separation of unfixed cells.

Following density gradient centrifugation, 3 ml of cell suspensions were transferred to new 15 ml Falcon tubes and then drawn into 5 ml syringes with an 18 G needle. The syringes were individually placed in a compatible filter holder that had a filter area of 3.5 cm2_{. The}

suspension was then pushed through different filter setups using a syringe pump at a flow rate of 0.2 ml/min. For analysis of the filtration efficiency, PLT samples were diluted in 1X PBS with a dilution factor of 1:40 (input) and 1:20 (output). For WBC count, the samples were diluted 1:2 in 1 µl/ml Acridine Orange (AO) (Sigma Aldrich Cat. No. A8097) in 1X PBS before and after filtration. Cell counts for both PLT and WBC were assessed from brightfield imaging in a Cellometer Vision Trio (Nexcelom), where RBCs are distinguished from WBCs by their biconcave shape.

3.3 Evaluation of 3D-Printed Filterholder

(14)

3.4 RNA Extractions from Platelets

To analyse if RNA could be recovered from the purified platelet fractions, RNA extractions were made from the filtered samples. A volume of 2 ml was collected after each filtration and used for the extraction. The cell concentrations of each sample used for the extraction are

shown in Table 1. The RNeasy Plus Mini Kit (QIAGEN) was used for RNA extraction

according to the protocol. The extracted RNA concentration was measured with NanoDrop for the first experiment and with Qubit for the last two extraction experiments.

Table 1. Cell concentrations in platelet fractions after filtration. A presentation of cell concentrations in platelet fractions obtained using filter setups Membrane B (5 µm) and Membrane C (2 µm), used for RNA extractions in 3 separate experiments. The cell counts for PLTs were assessed with brightfield imaging in a CellometerVision Trio (Nexcelom).

RNA extraction Membrane B (5 µm) Cell concentration Membrane C (2 µm) Cell concentration 1 2,97×107 cells/ml 1,54×107 cells/ml 2 3,08×107 cells/ml 1,93×107 cells/ml 3 2,08×107cells/ml 8,35×106 cells/ml

3.5 Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR)

Synthesis of cDNA was performed from cell lysates from different blood fractions using the TaqMan Gene Expression Cells-to-CT Kit (Cat# AM1728; Thermo Fisher Scientific) according to the provided protocol. Expression levels of ITGA2B, CD63 and PTPRC were then determined at transcript level in the different blood fractions. Expression analysis was performed for the three aforementioned genes using TaqMan probes (Hs01116228_m1, Hs00156390_m1, Hs04189704_m1, respectively) and for the reference gene ACTB (Hs01060665_g1) using the StepOne Plus Real-Time PCR machine.

3.6 Ethical Considerations

(15)

4 Results

In the first part of this project, blood samples were used to investigate whether density gradient centrifugation followed by microfiltration can isolate pure platelet fractions without contamination of WBCs. In the second part of the project, the possibility to analyse platelet RNA from the pure PLT fraction was investigated. The purpose is to also maintain PBMCs and plasma in order to make parallel extractions of multiple biomarkers from one single “liquid biopsy”.

4.1 Blood Collection and Density Gradient Centrifugation

From each Vacutainer CPT tube, 3 ml of plasma and buffy coat was collected after centrifugation. After running the solutions through the filter setups a final volume of 2 ml was obtained for each.

4.2 Evaluation of Filter Setups

A total of 11 blood samples were used to evaluate two different filter combination setups,

Membrane A combined with either Membrane B (5 𝜇𝑚) or Membrane C (2 𝜇𝑚). For

evaluation of filtration efficiency, mean values and standard deviations were calculated for the number of cells of PLTs counted in samples before and after filtration.

First, investigation of the ability of PLTs to pass through the filter combinations used to capture and remove PBMCs from the solution (filter step 1) was made. The recovery of platelets after removal of leukocytes by filtration is presented in Figure 1. Counting of PLTs after filtration through the 5 µm Cyclopore membrane in a 25 mm filterholder and comparison with the number of PLTs present before filtration resulted in 44% recovered platelets. For the 2 µm membrane, 24% of the input platelets were recovered after filtration. Next, measurements of how many PBMCs were present in the plasma and buffy coat before filtration were made (Figure 2), which showed complete removal of PBMCs by filtration

(16)

Figure 1. Recovery of platelets after removal of leukocytes by filtration. Plasma and buffy coat from blood samples was filtered through a standard filter (Membrane A) combined with membranes having pore size of either 5 µm (Membrane B) or 2 µm (Membrane C) in a 25 mm filter holder. Cell counts were assessed by brightfield imaging using a Cellometer Vision Trio (Nexcelom). Blue bars show platelet input; orange bars shows how many platelets were present in the solution after filtration. Mean and SD from 11 independent experiments are presented.

0 2 4 6 8 10 12 14 16 18 Pl at el et s × 1 0^7

(17)

Figure 2. Efficiency of removal of PBMCs by filtration. Plasma and buffy coat from blood samples was filtered through a standard filter (Membrane A) combined with membranes having pore size of either 5 µm (Membrane B) or 2 µm (Membrane C) in a 25 mm filter holder. Cell counts were assessed by brightfield imaging using a Cellometer Vision Trio (Nexcelom). Blue bars, PBMCs in input fraction; orange bars, PBMCs in flowthrough/output fractions. Mean and SD from 11 independent experiments are presented.

0 1 2 3 4 5 PB MC s × 1 0^6

(18)

Figure 3. Enhanced recovery of platelets after filter elution with PBS. The filter combination used for filtration 1 were washed with 2 ml 1X PBS and the number of platelets were counted by brightfield imaging using a Cellometer Vision Trio (Nexcelom). (The ‘5 µm out F1’ and ‘2 µm out F1’ fractions in this case represent the output fractions of the first filtration step). Mean and SD values for three experiments on each membrane setup are presented.

We investigated the possibility to increase the number of platelets recovered in the output fractions after filtration to remove the PBMCs. For this, each membrane setup was washed with 2 ml 1X PBS. The results showed that more PLTs could be recovered from the membranes by additional washing/elution with PBS (Figure 3). An enhanced recovery of platelets was shown with a gain of 35% in the number of recovered PLTs for the 5 µm membrane setup and a gain of 246% in recovered platelets for the 2 µm membrane setup.

0 2 4 6 8 10 12 Pl at el et s × 1 0^7

(19)

Figure 4. Platelet capture using the VIVID membrane. Platelet fractions resulting from filtration 1 were filtered through the Vivid filter (a filter for platelet capture) and cells in the input (in) and flowthrough (out) solution were counted by brightfield imaging using a Cellometer Vision Trio (Nexcelom). Mean and SD values for three experiments for each membrane setup are presented.

To be able to isolate free plasma fractions from blood samples, investigation if blood platelets could be captured and removed from the flowthrough of the first filtering step was made. Three experiments were made on filtration 1 flowthrough from each filter setup to evaluate the Vivid membranes capacity of catching platelets. In Figure 4, the number of PLTs passing through the Vivid membrane is presented. For the fraction filtered through the 5 µm pore size setup, 97% of the platelets were captured by the Vivid membrane. For the fraction filtered through the setup with 2 µm pore size, 96% of the platelets were captured by the Vivid membrane. 0 1 2 3 4 5 Pl at el et s × 1 0^7

(20)

4.3 Evaluation of 3D-Printed Filterholder

A relatively large volume of each sample was lost from analysis during the first filtration step due to dead volume trapped inside the filterholder. Therefore 3D-printed filterholders that would result in equally pure output solutions while also reducing dead volume of solution trapped inside the holder was designed and evaluated. Figure 5 shows three 3D-printed filterholders with different shapes that, in the future, could possibly be used to hold filters to capture e.g. PBMCs, PLTs or exosomes separately. In Figure 6, a picture of the 3D-printed filterholder used in the cell separation experiments is shown.

Figure 5. Filterholders in different shapes. Figure 6. Final 3D-printed filterholder. Developed 3D-printed filterholders with different

shapes for collection of different steps in the process.

(21)

Figure 7. Evaluation of the 3D-printed filterholder with regards to PLT recovery. Plasma and buffy coat from blood samples was filtered through a standard filter (Membrane A) combined with the filter with 2 µm pore size (Membrane C) in a 3D-printed filterholder. Cell counts were assessed by brightfield imaging using a Cellometer Vision Trio (Nexcelom). Mean and SD values of two experiments are presented.

Figure 8. Evaluation of the 3D-printed filterholder with regards to leukocyte removal. Plasma and buffy coat from blood samples was filtered through a standard filter (Membrane A) combined with 2 µm pore size membrane (Membrane C) in a 3D-printed filterholder. Cell counts were assessed by brightfield imaging using a Cellometer Vision Trio (Nexcelom). Mean and SD values of two experiments are presented.

(22)

4.4 Evaluations of RNA Distributions in Platelets

The next step was to evaluate whether RNA could be isolated and analysed in the purified platelet fraction. RNA was extracted from three different filtration experiments for each filter combination setup. Results for the concentration of RNA extracted from the filtered platelet fractions are presented in Table 2. Similar results were obtained for platelet RNA extracted from the filter combinations with 5 µm and 2 µm filter membranes. A very low yield of RNA was established with this method for all setups, which prevented us from continuing with the resulting material.

Table 2. Concentration of RNA extracted from platelets. The concentration of RNA extracted from platelets after filtering using filter setups with Membrane B (5 µm) and Membrane C (2 µm) was assessed with NanoDrop or Qubit measurements in 3 separate experiments.

RNA extraction

(method for concentration measurements) Membrane B (5 µm) (ng/µl) Membrane C (2 µm) (ng/µl) 1 (NanoDrop) 0,9 0,6 2 (Qubit) < 20 < 20 3 (Qubit) < 20 < 20

4.5 qRT-PCR Validation

Since not enough RNA could be extracted from the filtered platelet fractions, a different approach using a kit that produces cDNA for use in qPCR directly from lysed samples was assessed to be able to perform real-time PCR quantifications. The three genes were quantified to estimate the levels of mRNAs expressed in different cell types in the platelet fractions. A similar level of expression for ITGA2B, a biomarker for PLTs, in the Platelet fraction as compared to the Unfiltered fraction containing WBC, PLTs and exosomes was observed (Figure 9). On the contrary, almost 90% reduced expression levels could be seen for both

PTPRC and CD63, which are biomarkers for WBC and exosomes respectively, in the filtered

platelet fraction as compared to the Unfiltered fraction. From this it could be concluded that

the filter assembly used in this study, Membrane C (2 µm), was able to separate PLTs from

(23)

Figure 9. Evaluation of filtered PLT solution in qRT-PCR. Unfiltered solution (Input) and filtered solution (Platelet-fraction), using Membrane C (2 µm) was compared in qRT-PCR for validation of PLT recovery and WBC contamination by quantification of ITGA2B (marker for PLTs), CD63 (marker for WBCs) and PTPRC (marker for exosomes). Blue bars, PLTs; purple bars, WBCs and pink bars, exosomes. Data were normalized to input samples.

5 Discussion

In this project, we have studied a new approach for liquid biopsies from blood samples. The aim was mainly to evaluate if we could separate PLTs from WBCs and plasma to enable further analyses on each fraction of the sample separately. Because of the fact that WBCs has a much higher mRNA content than PLTs, pure PLT fractions are necessary for downstream analyses. A platelet contains 0,2 fg mRNA compared to one single white blood cell that contains between 2-5 pg mRNA (27). So, the mRNA content is >10 000-fold higher in leukocytes than in platelets (28). This means that mRNA from WBCs has potential to disturb analysis of platelet mRNA even if the WBC content of the platelet fraction is very low.

(24)

5.1 Data Obtained

Blood from healthy blood donors was used in our study to evaluate cell separation using different filter setups. By density gradient centrifugation followed by microfiltration we managed to produce platelet fraction free from any nucleated cell. The filtering step to remove nucleated cells was done with two different membrane combinations; Membrane B followed by two Membrane A or Membrane C followed by two Membrane A. Figure 1 and Figure 2 show that a high purity of PLT can be obtained using the developed filtering method, but that a low percentage of PLT are recovered. We could see that Membrane B with fixed cells has a greater recovery of platelets compared to Membrane C without fixed cells (44% vs. 24%, see Figure 1). Since a large fraction of platelets are lost through the first filtration step, we tried washing the membranes to improve platelet recovery for both membrane combinations. A wash carried out with 2 ml 1X PBS showed that a greater extent of PLTs could be recovered from the filters. The first filter step had a mean recovery of 2,65×107 cells/ml for fixed cells

through the 5 µm pore size membrane combination and a mean recovery of 1,51×107 cells/ml

without fixative cells for the 2 µm pore size membrane combination (Figure 3). An increase in platelets is shown with this approach, especially when using smaller pores without a fixative, where over 100% platelets could be gained compared to the membrane with larger pores with fixative cells, where 35% was gained.

This indicates that it is easier for the cells to be washed out with 1X PBS from Membrane C without fixative cells compared to Membrane B with fixative. It is possible to achieve either high purity free from nucleated cells or having good platelet recovery, rarely both at the same time.

An observation that there is still detectable expression of CD63 and PTPRC in the Platelet-fraction (Figure 9) indicates that low contamination of WBC RNA and exosomes could be present. The explanation can be that WBCs and exosomes are stuck in the membrane based filter assembly or that it is free RNA in the solution that has been detected. For downstream transcriptional studies on TEPs it holds great potential as a pre-analytical procedure.

(25)

5.2 Limitations and Future Approaches

In the future it is important to develop an understanding of the purpose of platelet RNA to be able to obtain a more complete picture of the platelets function in the body. If platelets potentially play a direct role in translational regulation or through RNA transfer, it is paramount to understand how they impact disease progression (9).

The degree of platelet activation caused by filtration is of interest since it causes release of MVs that can possibly contain RNA (29). In the future, an assessment with flow cytometer can be done, in which platelet activation could be measured using for example the CD62P marker (30).

Another important step that is crucial for the processes to completely work is the ability to separate the purified PLTs from plasma. Only a limited number of observations have been done on the Vivid membrane (Platelet capture), which is definitely worth investigating more deeply. The challenge is to obtain high purity plasma and at the same time catch all platelets. Although most of the PLTs were captured by the Vivid membrane, there are still some platelets that slipped through the membranes (Figure 4).

Another necessity that should be validated is the Vacutainer CPT tubes, which are very expensive. Before basing any procedure on them an alternative would be to try EDTA tubes. The big advantage with CPT tubes is that we are able to resuspend the plasma and buffy coat without getting contamination from the RBCs thanks to the separation provided by the gel. Finally, to be able to apply this to cancer patients in the future, we must consider patients undergoing treatments with cytotoxic drugs and counts of altered WBCs and PLTs in high concentrations. It is therefore important to see how final compositions are affected by performing these experiments on samples with different levels of baseline cells to further develop this method.

6 Conclusions

(26)

value. Development may potentially include other blood based cancer biomarkers of interest such as TEPs, PBMC, plasma proteins, circulating tumour DNA and extracellular vesicles.

7 Acknowledgements

First of all, I would like to thank my supervisor Tobias Sjöblom for this amazing opportunity and for trusting me with this project. I have learned so much from having you as a supervisor, most importantly how to become more independent, which I am very grateful for! I would also like to thank Ulf Landegren for being a part of this project.

I would like to pay my gratitude and thank all the members in the group for this time and for making me feel so welcome. A special thanks to Chatarina, Lucy and Snehangshu for guiding me and introducing me to everything in the lab, I am very thankful for your time and valuable help. Klara, Luis, Natalia and Veronica you have been so kind and always so motivational, I can’t thank you enough for your support and guidance during this time. Marina I am so glad we did this together and that I got to know you, thank you for your support and for always being so positive.

My sincere thanks goes to my classmates – Bonoshree, Caroline, Linda and Matilda – these five years would not have been possible or nearly as fun without you. I’m thankful for always having you by my side.

(27)

References

1. Ilié M, Hofman P. 2016. Pros: Can tissue biopsy be replaced by liquid biopsy?.

Translational lung cancer research 5:420-423.

2. Han X, Wang J, Sun Y. 2017. Circulating Tumor DNA as Biomarkers for Cancer Detection. Genomics Proteomics Bioinformatics 15:59-72.

3. Cheng F, Su L, Qian C. 2016. Circulating tumor DNA: a promising biomarker in the liquid biopsy of cancer. Oncotarget 7:30.

4. Strimbu K, Tavel JA. 2011. What are Biomarkers? Curr Opin HIV AIDS 5:463-466. 5. Pavlou MP, Diamandis EP, Blaustig IM. 2013. The Long Journey of Cancer

Biomarkers from the Bench to the Clinic. Clinical Chemistry 59:147-157.

6. Borrebaeck CAK. 2017. Precision diagnostics: moving towards protein biomarker signatures of clinical utility in cancer. Nature Reviews Cancer 17:199-204.

7. Best MG, Vancura A, Wurdinger T. 2017. Platelet RNA as a circulating biomarker trove for cancer diagnostics. Journal of Thrombosis and Haemostasis 15:1295-1306. 8. Rowley JW, Schwertz H, Weyrich AS. 2012. Platelet mRNA: the meaning behind the

message. Curr Opin Hematol. 19:385-391.

9. Clancy L, Freedman JE. 2015. The role of circulating platelet transcripts. Journal of

Thrombosis and Haemostasis 13:S33-S39.

10. Joosse SA, Pantel K. 2015. Tumor-Educated Platelets as Liquid Biopsy in Cancer Patients. Cancer Cell 28:552-554

11. Nilsson RJA, Balaj L, Esther Hulleman, Rijn S, Pegtel M, Walraven M, et al. 2011. Blood platelets contain tumour-derived RNA biomarkers. Blood 118:3680-2683. 12. Fink L, Hölschermann H, Kwapiszewska G, Muyal JP, Lengemann B, Bohle RM, et

al. 2003. Characterization of platelet-specific mRNA by real-time PCR after laser-assisted microdissection. Thromb Haemost. 90: 749-756.

13. Rolf N, Knoefler R, Suttrop M, Kluter H, Bugert P. 2005. Optimized Procedure for Platelet RNA Profiling from Blood Samples with Limited Platelet Numbers. Clin

(28)

14. Di Meo A, Bartlett J, Cheng Y, Pasic MD, Yousef GM. 2017. Liquid biopsy: a step forward towards precision medicine in urologic malignancies. Molecular Cancer

16:80.

15. Dominguez-Vigil IG, Moreno-Martinez AK, Wang JY, Roehrl MHA, Barrera-Saldana HA. 2018. The dawn of the liquid biopsy in the fight against cancer. Oncotarget

9:2912-2922.

16. Volik S, Alcaide M, Morin RD, Collins C. 2016. Cell-free DNA (cfDNA): Clinical Significance and Utility in Cancer Shaped By Emerging Technologies. Molecular

Cancer Research 14:898-908.

17. Stroun M, Lyautey J, Lederrey C, Olson-Sand A, Anker P. 2001. About the possible origin and mechanism of circulating DNA: Apoptosis and active DNA release. Clin

Chim Acta. 313:139-142.

18. Jahr S, Hentze H, Englisch S, Hardt D, Fackelmayer FO, Hesch RD, et al. 2001. DNA Fragments in the Blood Plasma of Cancer Patients: Quantitations and Evidence for Their Origin from Apoptotic and Necrotic Cells. Cancer Research 61: 1659-1665. 19. Christie EL, Dawson SJ, Bowtell DDL. 2016. Blood Worth Bottling: Circulating

Tumour DNA as a Cancer Biomarker. Cancer Research 76:5590-5591.

20. Sol N, Wurdinger T. 2017. Platelet RNA signatures for the detection of cancer.

Cancer Metastasis Rev. 36:263-272.

21. Murtaza M, Dawson S-J, Progrebniak K. 2015. Multifocal clonal evolution characterized using circulating tumour DNA in a case of metastatic breast cancer.

Nature Communications 6:8760.

22. Keerthikumar S, Gangoda L, Liem M, Fonseka P, Atukorala I, Ozcitti C, et al. 2015. Proteogenomic analysis reveals exosomes are more oncogenic than ectosomes.

Oncotarget. 6:15375-15396.

23. Calapre L, Warburton L, Millward M, Ziman M, Gray ES. 2017. Circulating tumour DNA (ctDNA) as a liquid biopsy for melanoma. Cancer Letters 404:62-69.

24. Bebelman MP, Smit MJ, Pegtel DM, Baglio SR. 2018. Biogenesis and function of extracellular vesicles in cancer. Pharmacology & Therapeutics.

(29)

26. Cohen JD, Javed AA, Thoburn C, Wong F, Tie J, Gibbs P, et al. 2017. Combined circulating tumour DNA and protein biomarker-based liquid biopsy for the earlier detection of pancreatic cancers. Proc Natl Acad Sci USA. 114:10202-10207. 27. Rox JM, Muller J, Pötzsch B. 2006 Platelet Transcriptome Analysis*. Transfusion

Medicine and Hemotherapy 33:177-182.

28. Rox JM, Burgert P, Muller J, Schorr A, Hanfland P, Madlener K, et al. 2004. Gene Expression Analysis in Platelets from a Single Donor: Evaluation of a PCR-Based Amplification Technique. Clinical Chemistry 50:2271-2278.

29. Preuber C, Hung L-H, Schneider T, Schreiner S, Hardt M, Moebus A, et al. 2018. Selective release of circRNAs in platelet-derived extracellular vesicles. J Extracell

Veiscles 7:1424473