The role and definition of cancer cells in blood

(1)

(2)

The role and definition

of cancer cells in blood

(3)

Samenstelling promotiecommissie:

Prof. dr. H. Hilgenkamp Universiteit Twente (voorzitter)

Prof. dr. L.W.M.M. Terstappen MD Universiteit Twente (promotor)

Prof. dr. H.B.J. Karperien Universiteit Twente

Prof. dr. D. Stamatialis Universiteit Twente

Prof. dr. C.J.M van Rijn Universiteit Wageningen

Prof. dr. H.J.M Groen MD Rijksuniveristeit Groningen

Dr. G. Attard MD Royal Marsden NHS foundation trust

This work was nancially supported by: EU ERC starting grant 282276 EU FP7 CTCTrap grant 305341 Janssen Diagnostics, LLC

transmitted, in any form or by any means, electronically or mechanically, including photocopying, microlming, and recording, or by any

information storage or retrieval system, without prior written permission of the author.

ISBN 978-90-365-3831-2 DOI 10.3990/1.9789036538312

Cover: Cells stained by immunouorescence in a CellSearch cartridge (front) and on a microsieve (back).

(4)

The role and definition

of cancer cells in blood

Proefschrift

ter verkrijging van

de graad van doctor aan de Universiteit Twente, op gezag van de rector magnicus,

prof. dr. H. Brinksma,

volgens besluit van het College voor Promoties in het openbaar te verdedigen

op vrijdag 30 januari 2015 om 12:45 uur

door Guus van Dalum geboren op 8 maart 1981

(5)

Dit proefschrift is goedgekeurd door:

(6)

Introduction

Thesis motivation

Cancer is the leading cause of morbidity and mortality in high developed countries and will be a major cause of morbidity and mortality for the en-tire world in the near future1_{. It is therefore also a disease which is and} has been subject to extensive research. Cancer is slowly moving away from being lethal to becoming a disease that is treatable. Ideally all cancer types should be treatable as a chronic disease. With the increased understanding of cancer it has become clear that cancer is a very heterogeneous disease in many ways. A large number of risk factors including numerous genetic predispositions and environmental factors have been identied. Also the cancer cells within a patient can be very heterogeneous. Because not all cells are similar, it is possible that some cells respond dierently to the same treatment. This is one of the reasons that a tumor can develop drug resistance. This raises the question, which of these cell types cause the spread of cancer through the body and form distant metastasis. Not all steps in the metastatic cascade from primary tumor to the formation of a distant metastasis are fully understood. But to spread, the cells have to leave the primary tumor and enter the blood circulation. Cancer cells that rst spread through the lymphatic system will also end up in the blood cir-culation. Although technically challenging, these cancer cells in the blood stream can be isolated and are called circulating tumor cells. They can provide real time information of the metastatic process and knowledge of the composition of these tumor cells should be ideal to select the most appropriate therapy. The presence of tumor cells in blood is known to be prognostic for poor survival in patients with disseminated disease and can be used to evaluate the response of patients to treatment. Heterogeneity of tumor cells and changes in time giving rise to resistance to therapy requires access to tumor cells throughout the course of the disease. Identication, dierentiation and characterisation of cancer cells in blood is the subject of this thesis

1_{Bray, F. et al. Global cancer transitions according to the Human Development}

Index (2008-2030): a population-based study. Lancet Oncol. 13, 790801 (Aug. 2012).

(13)

xii

CONTENTS

Thesis outline

This thesis has the following content:

The relation between circulating tumor cells (CTC) and the metastatic process is discussed in chapter 1. This chapter also contains an overview of several CTC enumeration and characterization methods, together with the assumptions made about the CTC phenotype when using each method.

In chapter 2 and 3 the prevalence of CTCs in non-metastatic colorectal and breast cancer and it's association with recurrence free survival and overall survival is investigated. In these studies the CTC are enumerated with the CellSearch assay before surgery and at several time points after surgery.

Not all cells found in a sample after a CellSearch test can be identied as either a tumor cell or a leukocyte. In chapter 4 the frequency of unidentied cells by CellSearch is quantied. The number of nucleated events and the proportion of nucleated events identied as leukocytes by the expression of CD45 was retrospectively analyzed for several studies. In this chapter additional markers are explored to increase the number of identied cells in healthy donor and patient samples.

In chapter 5 methods are explored to quantify therapy targets on CTC. The CellSearch assay allows for the detection of additional markers on the tumor cells, but quantication of the expression level of these markers is hard to do manually. In this chapter an image analysis algorithm is used to evaluate the nuclear expression of the androgen receptor in circulating tumor cells from castration resistant prostate cancer patients.

An alternative to the EpCAM based enrichment of CTC as done in CellSearch is enrichment based on cell size and exibility using ltration membranes. This ltration process through pores with cell size diameters can be modeled and a theoretical model is described in chapter 6. The various ltration parameters inuencing this ltration process are investi-gated in chapter 7 and in chapter 8 the lter characteristics are compared. These chapters resulted in an EpCAM agnostic enrichment method for cir-culating tumor cells. In chapter 9 a method was developed to collect the blood discarded by CellSearch after immunomagnetic depletion of cells ex-pressing EpCAM. This blood was ltered using the method described in chapter 8 and after optimization of a protocol to uorescently label and identify the ltered cells the presence of EpCAM+ and EpCAM- CTCs was investigated in blood samples from lung cancer patients.

(14)

CHAPTER

1

Metastasis and Circulating

Tumor Cells

G. van Dalum, L. Holland and L.W.M.M. Terstappen JIFCC 23.3 (2012): 1-11.

Abstract

Cancer is a prominent cause of death worldwide. In most cases, it is not the primary tumor which causes death, but the metastases. Metastatic tumors are spread over the entire human body and are more dicult to remove or treat than the primary tumor. In a patient with metastatic disease, circulating tumor cells (CTCs) can be found in venous blood. These circulating tumor cells are part of the metastatic cascade. Clinical studies have shown that these cells can be used to pre-dict treatment response and their presence is strongly associated with poor survival prospects. Enumeration and characterization of CTCs is important as this can help clinicians make more informed decisions when choosing or evaluating treatment. CTC counts are being included in an increasing number of studies and thus are becoming a bigger part of disease diagnosis and therapy management. We present an overview of the most prominent CTC enumeration and characterization methods and discuss the assumptions made about the CTC phenotype. Exten-sive CTC characterization of for example the DNA, RNA and antigen expression may lead to more understanding of the metastatic process.

(15)

2 1.1. INTR ODUCTION

1.1 Introduction

Cancer is the worlds third cause of death and the leading cause of death in economically developed countries[1]. In cancer most often the metastases are the cause of death and not the primary tumor. When a patient is diagnosed with cancer before it has spread outside of the primary tumor, it improves the chance of survival. But spotting a primary tumor before it has the chance to metastasize is dicult. Physical examination and traditional imaging methods such as MRI, PET, CT, X-ray or ultrasound have a detection limit which is not sucient to detect smaller metastasis. For breast cancer this detection limit is for example 6mm or larger [2]. This makes it hard to spot small lesions or micro-metastases. It is still unclear exactly when and how the metastatic process begins and which factors drive the process, but it is known that tumor cells spread via the lymphatic system and subsequently into the blood circulation or are shed directly into the blood. These Circulating Tumor Cells (CTCs) are associated with poor progression free and overall survival [310]. CTC are rare and require multiple steps to enumerate, but we catch metastasizing cells in the act and thus they may increase our understanding of the metastatic process. CTC may also provide a way to monitor disease progression more directly than traditional imaging methods. Here we review the current state of the CTC detection eld and the extra information these cells can provide us with now and in the future.

Figure 1.1: Formation of metastasis. Panel A shows a primary tumor after its ception. The heterogeneity of the tumor is in-dicated with dierent colors. Panel B shows blood vessels providing nutrients to the tu-mor (angiogenesis) leading to further growth and diversity. In this process either the tu-mor cells or the endothelial cells will need to penetrate the basal lamina. At this time tumor cells can enter the blood. The arrow depicts a CTC attached to the blood vessel wall of a distant organ. Panel C shows the formation of a metastasis after extravasation of the CTC. Only few CTC will have the char-acteristics necessary to create a metastasis.

(16)

3 CHAPTER 1. MET AST ASIS AND CIR CULA TING TUMOR CELLS

Figure 1.2: Dierentiation into heterogeneous metastases. A pri-mary tumor in the breast generally spreads to the liver (panel A), lung (panel B) or bone (panel C). Due to tumor heterogeneity some tumor cells have a preference for a certain organ indicated by their color. Panel D shows a primary tumor in the prostate which only spreads to bone.

1.2 Metastatic process

Cancer occurs after a cell is progressively genetically damaged and turns into a cell bearing a malignant phenotype. These cells are able to undergo uncontrolled abnormal mitosis, which leads to an increase of these can-cerous cells at that location. In absence of regular control mechanisms a heterogeneous population of cells is created and these cancerous cells to-gether form the primary tumor [11]. A tumor is considered benign if it lacks the ability to invade other tissue. When cells acquire the ability to penetrate and inltrate surrounding normal tissues, the cancer is consid-ered malignant and has the potential to metastasize. Before tumor cells can start to metastasize, they need to succeed in stimulating angiogenesis. In this way tumor cells gain direct access to the blood circulation. This leads to improved access to the nutrients and oxygen carried by the blood, but also an opportunity for the tumor cells to enter the blood stream. This process is shown in gure 1.1. An alternative route for tumor cells to end up in the blood circulation is through the lymphatic system.

Tumor cells circulating in the blood can reach in principle most sites of the body, but dierent kinds of cancer create metastasis at dierent sites. For example breast cancer generally creates metastases in liver, lung and

(17)

4 1.3. CIR CULA TING TUMOR CELLS

bone while prostate cancer most often metastasizes in bone as illustrated in gure 1.2. This preference is driven by two processes. The rst is mechani-cal of nature, a large amount of CTC arrests in the rst capillary bed they encounter. The second is more biological, the CTCs will form a metastasis in tissue only if they are able to extravasate out of the blood stream and the local environment is suitable for them to grow. This preference has been noted for the rst time by Stephen Paget and is known as the seed and soil hypothesis. Tumor cells thus have a preference for a certain site, and this opens an interesting research eld to identify the cell surface molecules on the tumor cells and the endothelial cells aligning the capillaries at the specic sites [1215].

1.3 Circulating tumor cells

Figure 1.3: The frequency of ery-throcytes, platelets, leukocytes and circu-lating tumor cells in blood of metastatic carcinoma patients and their cumulative probability[16] The rst observation of tumor cells in blood was

made by Thomas Ashworth in 1869 [17]. In sub-sequent reports CTCs were only observed in blood when present in high numbers [1822]. As tech-nology advanced it became possible to detect the presence of CTC in a much lower concentration. For example various PCR techniques can be used to detect CTC in blood but are less suitable for enumeration [23]. The combination of ow cytom-etry with a magnetic enrichment step allowed for CTC enumeration down to 1 cell per mL [24]. CTC in peripheral blood of patients with metastatic dis-ease turned out to be very rare, and range from 0 to 10000 CTCs per mL of whole blood [25]. Their frequency compared to other cells present in blood is shown in gure 1.3.

Prospective clinical studies in breast, colon and prostate cancer [35] showed that the presence of CTCs in 7.5 mL of blood strongly correlated with progression free and overall survival. The relation between the number of CTC in 7,5 mL of blood from 294 metastatic breast and prostate cancer pa-tients and survival is illustrated in a Kaplan-Meier plot in gure 1.4. The blood samples were taken after the patients received the rst cycle of chemo therapy. The gure clearly shows the relation be-tween the CTC load and survival. Intuitively the dierence between <1 CTC and 1-4 CTC should make a larger dierence, but because not the com-plete volume of blood is measured the

(18)

concentra- 5 CHAPTER 1. MET AST ASIS AND CIR CULA TING TUMOR CELLS

tion of CTC contains a Poison error due to sampling. This together with the chance of a false positive makes it hard to distinguish between none and a few CTC. If error free detection would be possible even one CTC in 7,5 mL would already lead to a worse prognosis [26]. Lower concentrations of CTC can be found in patients when the analysis of a higher volume is possible. If the concentration of CTC currently found in patients is plotted as a cumulative distribution function a distribution can be plotted through it to extrapolate the lower concentrations [27]. The result is a distribution such as seen in gure 1.3, where 99% of the patients with metastatic disease have at least 1 CTC in 5L of blood before initiation of therapy. The com-bination of low numbers and the risk of identifying a false positive makes the selection of a good threshold for the separation of patients important. The denition of a CTC also has a strong inuence on this threshold. For example in immunouorescent microscopy a more loose denition of what is a CTC will result in higher counts in patients and in healthy controls, but also in a lower distinguishing power [28, 29]. It is important to only select the CTC or objects which inuence outcome. Comparisons between CTC and other predictors of prognosis or response to therapy such as serum tu-mor markers and imaging modalities have shown that CTC perform well and are independent predictors of outcome [4, 5, 3034].

1.4 Current CTC detection methods

Of all the methods to enumerate CTC most have some form of enrichment to make the number of cells that have to be analyzed manageable. But what exactly is the denition of a circulating tumor cell? A nucleated cell of non-hematopoietic origin, a nucleated cell of epithelial cell origin, or a cell with an aberrant genotype? To nd CTC all sample preparation meth-ods make certain assumptions, for erythrocyte lysis the assumption is made that CTC are not lysed, for separation based on density, size or exibility that they have the chosen properties, for depletion of the hematopoietic cells that they are not aberrantly expressing the hematopoietic antigens or are bound to the hematopoietic cells and for enrichment based on antigen expression that they bear the antigens chosen for separation. To identify CTC in the enriched sample similar assumptions will have to be made. A PCR method on the CTC enriched sample targeting for example cytoker-atins assumes that the expression of these kercytoker-atins is conned to tumor cells. Moreover a large heterogeneity of the number of cytokeratin copies is present in cells prohibiting enumeration of the CTC. Detection of extracel-lular and intracelextracel-lular antigens by means of uorescence labeled antibodies is the most frequently used method. The combination of antigens and the sensitivity by which the antigens are detected will greatly inuence the number of cells that are identied as CTC. A more strict criterion will result in more specic but less sensitive detection method. The choice of

(19)

6 1.4. CURRENT CTC DETECTION METHODS

Figure 1.4: Kaplan-Meier plots for overall survival of samples from metastatic breast (a), colon (c), and prostate (d) cancer patients with 0, 14, 524, and >25 CTC at the start of therapy and at follow up for metastatic breast (b). The number of patients at risk is listed at every time point of measurement.

(20)

uorochromes and the method to detect them will also inuence the sen-sitivity and specicity of the assay. Whereas owcytometry is in general more sensitive to detect uorescence as compared to mercury arc based u-orescence microscopy the measurement of light scatter properties by owcy-tometry will not provide the same level of condence on cell morphology as microscopic images provide. The ideal CTC enumeration assay has: high sensitivity, high specicity, is reproducible and is somewhat independent of sample lead time. For CTC characterization it would be ideal to have vi-able CTC with intact morphology and preserved cellular content to detect extracellular and intracellular antigens, RNA and DNA.

1.5 Antibody based CTC enrichment

Although there are many dierent detection assays described in literature only one is cleared by the FDA, this is the CellSearch system from Veridex. The CellSearch system uses anti-EpCAM antibodies conjugated to coated ferrouids in combination with a strong magnetic eld to selectively enrich cells expressing the EpCAM antigen as is illustrated in gure 1.5A. After enrichment the cells are permeabilised and stained with DAPI, anti-CD45-APC and anti-CK-PE (CK 8, 18 and 19). The stained cells are then placed in a sample holder with a magnetic eld that uniformly distributes the cells across the imaging area. The sample is then imaged with an automated uorescent microscope and an image processing algorithm identies pos-sible CTC. These candidates are displayed in a gallery of thumbnails as illustrated in gure 1.6A for review by an operator [35, 36]. When look-ing into the CellSearch CTC denition it stipulates that: a cell should be suciently large (>16 µm2_{), with an intact intracellular nucleus, does not} stain with CD45-APC and does stain with keratins 8, 18 and 19 enveloping the nucleus. Expression of EpCAM is likely as the cells were immunomag-netically pulled to the surface, cells of hematopoietic origin will however also be present.

A method that samples a much larger volume is the functionalized and structured medical wire [37]. One end of the wire is functionalized with human anti-EpCAM antibody. A schematic representation is illustrated in gure 1.5B. The wire is inserted into the cubital vein of a patient for 30 minutes, so cells can be captured in vivo. After which it is removed and stained for DNA, EpCAM, cytokeratins 4, 5, 8, 11 ,18 and CD45. Imaging is done using conventional uorescence microscopy. The frequency of EpCAM positive epithelial cells among the captured cells has not been reported yet. As a larger blood volume is sampled the method has the potential to be more sensitive compared to the CellSearch system [27].

The Epispot method uses an antibody coated surface to capture the proteins secreted by a circulating tumor cell [38, 39]. Cells are deposited in a petri dish and incubated to permit the cells to secrete proteins. The

(21)

8 1.5. ANTIBOD Y BASED CTC ENRICHMENT

Figure 1.5: Examples of dierent CTC enrichment methods. Panel A) CellSearch which uses iron particles coupled to anti-EpCAM antibodies to enrich cells from epithelial origin. Cells are kept in a magnetic eld during wash steps. After staining they are presented in a MagNest to pull all EpCAM positive cells to the imaging area. Panel B) Functionalized nano wire, an EpCAM functionalized probe is placed directly in the blood stream to capture CTC. After 30 minutes the probe is removed from the blood and the attached cells are stained and xed. The cells are investigated while still on the probe. Panel C) Rosettesep, white blood cell depletion is achieved by forming aggregates between erythrocytes and CD45 positive cells. After which the aggregates are separated from the CTC using density separation. Panel D) passing whole blood through a sieve with 8 µm pores. The erythrocytes and white blood cells will pass the pores while some of the bigger cells will stay on the membrane surface. The cells are xed and stained on the lter.

(22)

proteins secreted by the cells are selectively bound to the surface, after the cells are washed o the target proteins are stained with a second antibody targeting the same protein. This method is only sensitive to the target antigens of the spotted antibodies that are secreted by live cells and does not allow direct interaction with the CTC as they have been washed away.

Microuidic devices have been described [4042] with a functionalized surface to capture CTC. The cells collide with EpCAM functionalized sur-face to enrich CTC. After the cells are captured in the microuidic device they are stained while they remain on the chip. The ability to do further analysis of the captured CTC in a microuidic device is one of the main ad-vantages of using the lab on a chip approach [43]. The main disadvantage of microuidic approaches is that all devices presented so far are limited in the amount of uid that can be processed. The time for processing one milliliter of whole blood per hour [41] will make the analysis of larger blood volumes unpractical. The reports of higher number of CTC detected with these systems is likely due to a lesser stringency of the criteria used to assign objects as CTC. Comparison of CellSearch CTC detection with detection by owcytometry using EpCAM-PE and CD45-PerCP and a nu-cleic acid dye after erythrocyte lysis showed a maximum of 3 fold increase in sensitivity [27].

1.6 Density based enrichment

Traditionally density based enrichment techniques such as Ficoll with its density of 1.077 is used to separate the mononuclear cell fraction from blood. It also has been used to enrich CTC although the range of densities in which CTC appear is not known. Other density based separation such as Percoll or Oncoquick permit a well dened range of densities, The Rosette-sep method uses antibodies in combination with density Rosette-separation for the depletion of CD45 positive cells as illustrated in gure 1.5C. The antibod-ies present form aggregates of red blood cells and the CD45 positive cells, thus increasing the density of all white blood cells. After a density sep-aration using Ficoll, the mononuclear cell population can be cytospinned on a glass and stained or run on a ow cytometer. Although 62% of cells derived from tumor cell lines can be recovered after spiking in whole blood using Rosettesep the implicit assumption is that CTC will have a similar density to peripheral blood mononuclear cells and cells derived from cell lines [44]. A novel density based method is RareCyte which uses a oater. The oater has the same density as the mononuclear cells. The oater is designed such that there is only a small gap of a few µm between the tube and the oater. This causes a thin layer of nucleated cells to be formed near the surface of the custom tube so all the nucleated cells can be imaged [45]. In general when comparing traditional density separating methods with EpCAM specic methods the latter are more sensitive, but might miss the

(23)

10 1.7. SIZE OR FLEXIBILITY BASED ENRICHMENT EpCAM negative CTC [4648].

1.7 Size or exibility based enrichment

CTC are slightly larger than white blood cells [49] and this property is leveraged by several methods which select on size or exibility. The whole or lysed blood is ltered through a membrane with distinct pore sizes vary-ing from 5 to 8 µm such as illustrated in gure 1.5D. The premises is that CTC stay on the lter while white blood cells pass the pore. There are dif-ferent fabrication methods for such membranes. Track etched membranes are made of polycarbonate with a random pore distribution and used as in several systems such as ISET and ScreenCell [50, 51] or directly [52] Micro fabricated membranes from parylene [5355], silicon [49, 56, 57] or other materials [58, 59] have a regular pore spacing and well dened pore dimensions. They can also have a more complex topology.

In for example the ISET system the sample is rst mixed with a buer that lyses the erythrocytes and xes all other cells. The sample is loaded in a disposable with a lter membrane on the bottom. The sample ows through the membrane in 3 minutes [53]. After ltration the sample is stained and enumerated. The dierent ltration methods all perform well using cells derived from cell lines with relatively large size and are able to catch EpCAM negative CTC. The parameters under which CTC ltration is done however varies greatly between methods with variation in xation, erythrocyte lysis, sample dilution and operating pressure [56]. One of the main advantages of ltration is that the CTC remain quite easily accessible which facilitates both reanalysis of the whole sample and individual cells. It is also possible to lter unxed cells which results in viable enriched cells. Enriching CTC based on size has also been applied in a micro uidic device [60, 61]. Here 900 cups consisting of three posts placed in a triangle position with a 5 µm gap in between the posts. The gaps allow normal cells to pass and capture CTC. The triangles are spaced 20 µm from each other so that cells can ow past them if a cup is occupied. Their unique capturing structure allows for the release of captured cells from the chip. So far this chip by Clearbridge Biomedics is the only commercially available microuidic CTC isolation device.

1.8 CTC detection after erythrocyte lysis

CTC can be detected by owcytometry [62, 63] or laser scanning cytome-try [64] after erythrocyte lysis and staining with uorescently labeled anti-bodies. The combination of EpCAM-PE, CD45-PerCP and a nucleic acid indeed can be used to identify CTC in 100 µL of blood [62], increasing the blood volume however results in a large background in healthy controls [62, 63]. Using this ow cytometry approach the maximum increase in yield

(24)

of CTC was estimated at 3.3 fold, compared to the number of CTC found in CellSearch [27]. This increase is however awed as by ow cytometry no detailed information about morphology is provided, resulting in events falsely identied as CTC.

A technology introduced by Epic Sciences uses specially prepared slides to place the blood after erythrocyte lysis. The cells are allowed to settle to create an evenly distributed mono layer of cells. This allows for the imaging of all non-lysed cells present in blood. Identication of CTC is done using a dedicated image analyzer with a dedicated image processing technique [65] making the analysis of large amount of cells possible. Having no enrichment is advantageous for cell loss but will make the technique more sensitive to unspecic immunouorescent staining.

1.9 Comparing dierent enumeration methods

Because of the low numbers of CTC found in patient samples new methods are being explored to increase sensitivity. Each CTC detection and isola-tion technique however makes its own assumpisola-tions about what constitutes a CTC. This is important to keep in mind when comparing techniques, especially considering the fact that changing the denition a CTC in the same technique can already have a large impact on the prognostic power of the assay [29]. Finding more CTC does not automatically imply that the distinguishing power of the assay increases. The recovery of spiked cells gives some information about the eciency of a technique. For patient samples the biggest hurdle is that the number of CTC that are present is unknown and new techniques are frequently compared with CellSearch. When drawing conclusions from this comparison it is important to note that the phenotype of the cells found with the new method might not be the same as found by CellSearch. When the same phenotype is measured a cor-relation between the two techniques could indicate that the same prognostic value might apply to the CTC found. Ultimately for real conclusions about their prognostic value, prospective clinical studies are needed. These stud-ies would not only provide interesting data about the enrichment method, but will also give an idea of which CTC phenotype is the most important for the prognosis of the patient.

1.10 Detection of treatment targets on CTC

Besides enumeration, CTC can also serve as a means to detect the presence or absence of treatment targets. For example in breast cancer the Her2/neu expression is of importance when considering Herceptin therapy and can be assessed on CTC both at the protein and gene level as illustrated in gure 1.6B and 1.6E, [6871]. In prostate cancer the expression of the androgen receptor, PTEN and ERG is important and can also be detected on CTC

(25)

12 1.10. DETECTION OF TREA TMENT T AR GETS ON CTC

Figure 1.6: Panel A, thumbnail images of CTC detected in metastatic cancer patient using CellSearch Row 1 one DAPI+ cell, CK+, CD45- => CTC. Row 2 four DAPI+ cells, one CK+, CTC, one CKdim, CD45-CTC, one CK-, CD45+ leukocyte and one CK-, CD45- with no proof of origin. Panel B and C, detection of treatment targets Her2 and AR on CTC using immunouorescence. Panel B, a Her2+ and Her2- CTC, Panel C, an AR+ and AR- CTC. For weak signals such as the androgen receptor a bright uorophore is needed in this case the CK antibody is labeled with FITC and the AR labeled antibody with Phycoerythrin. Panels D and E, CTC can be restained for FISH to asses chromosomal abnormalities [66], panel D, or specic genes HER2, AR, PTEN, ERG [67][68]

Figure 1.6C and 1.6E [67]. Clear advantage of CTC above traditional biopsies is that it can provide a real time assessment of the tumor to be

(26)

treated. CTC give a more recent and thus sometimes dierent picture than the information from biopsies [69, 70, 72, 73]. Clinical studies will however need to be conducted to prove that the detection of treatment targets on CTC is more benecial to the treatment outcome than the assessment of treatment targets on a biopsy. Most important is that the event detected is indeed a tumor cell and can be assessed by for example ploidy status as illustrated on a CTC in Figure 1.6D [66]. Ideally CTC should be used as a tool to guide therapy: if a patient's CTC are not eliminated after the rst cycles of therapy a switch to another therapy should be considered as the current one is not working [4, 5, 74, 75]. The use of CTC in this manner is currently being investigated in a multicenter trial for breast cancer by the south west oncology group (SWOG 0500 - NCT00382018). The type of therapy to be administered can be obtained from the detected treatment targets on CTC as is currently investigated in the DETECT III trial (NCT01619111).

1.11 The future

For CTC not only enumeration is of importance, but also the ability to examine them for the absence or presence of treatment targets. To achieve this goal proteins, RNA and DNA that contain information pertinent to treatment should be preserved in CTC and most important we should be able to isolate and characterize them in all patients with metastatic disease whether detected with present technology or not. Given the fact that in approximately 50% of patients with metastatic disease no CTC can be detected with the currently available validated CellSearch system and in a signicantly lower portion of patients with primary disease with a risk of recurrence, it will be quite a challenge to develop technology that can benet all cancer patients.

References

1. Jemal, A. et al. Global cancer statistics. CA. Cancer J. Clin. 61, 6990 (2011). 2. Kelly, K. M., Dean, J., Lee, S. J. & Comulada, W. S. Breast cancer detection:

Ra-diologists' performance using mammography with and without automated whole-breast ultrasound. Eur. Radiol. 20, 25572564 (2010).

3. Cristofanilli, M. et al. Circulating tumor cells, disease progression, and survival in metastatic breast cancer. N. Engl. J. Med. 351, 78191 (Aug. 2004).

4. Cohen, S. J. et al. Relationship of circulating tumor cells to tumor response, progression-free survival, and overall survival in patients with metastatic colorectal cancer. J. Clin. Oncol. 26, 321321 (July 2008).

5. De Bono, J. S. et al. Circulating tumor cells predict survival benet from treatment in metastatic castration-resistant prostate cancer. Clin. cancer Res. 14, 63029 (Oct. 2008).

6. Hiltermann, T. J. N. et al. Circulating tumor cells in small-cell lung cancer: a predictive and prognostic factor. Ann. Oncol. 23, 293742 (Nov. 2012).

(27)

14

REFERENCES

7. Rao, C. et al. Circulating melanoma cells and survival in metastatic melanoma. Int. J. Oncol. 38, 75560 (Mar. 2011).

8. Liu, M. C. et al. Circulating tumor cells: a useful predictor of treatment ecacy in metastatic breast cancer. J. Clin. Oncol. 27, 51539 (Nov. 2009).

9. Matsusaka, S. et al. Circulating tumor cells as a surrogate marker for determining response to chemotherapy in patients with advanced gastric cancer. Cancer Sci. 101, 106771 (Apr. 2010).

10. Krebs, M. M. G. et al. Evaluation and prognostic signicance of circulating tumor cells in patients with non-small-cell lung cancer. J. Clin. Oncol. 29, 155663 (Apr. 2011).

11. Marusyk, A. & Polyak, K. Tumor heterogeneity: causes and consequences. Biochim. Biophys. Acta (BBA)-Reviews Cancer 1805, 105117 (2010).

12. Abdel-Ghany, M. & Cheng, H. The breast cancer β4 integrin and endothelial hu-man CLCA2 mediate lung metastasis. J. Biol. Chem. 276, 2543825446 (2001). 13. Felding-Habermann, B. et al. Integrin activation controls metastasis in human

breast cancer. Proc. Natl. Acad. Sci. U. S. A. 98, 18531858 (2001).

14. Wang, H. et al. Tumor cell alpha3beta1 integrin and vascular laminin-5 mediate pulmonary arrest and metastasis. J. Cell Biol. 164, 93541 (Mar. 2004). 15. Gassmann, P., Kang, M.-L., Mees, S. T. & Haier, J. In vivo tumor cell adhesion in

the pulmonary microvasculature is exclusively mediated by tumor cellendothelial cell interaction. BMC Cancer 10, 177 (2010).

16. Ligthart, S. Redening circulating tumor cells by image processing PhD thesis (University of Twente, May 2012).

17. Ashworth, T. R. A case of cancer in which cells similar to those in the tumors were seen in the blood after death. Aust. Med. J. 14, 146149 (1869).

18. Carey, R. W., Taft, P. D., Bennett, J. M. & Kaufman, S. Carcinocythemia (carci-noma cell leukemia). An acute leukemia-like picture due to metastatic carci(carci-noma cells. Am. J. Med. 60, A77 (Feb. 1976).

19. Myerowitz, R. L., Edwards, P. A. & Sartiano, G. P. Carcinocythemia (carcinoma cell leukemia) due to metastatic carcinoma of the breast: report of a case. Cancer 40, 310711 (Dec. 1977).

20. Gallivan, M. V. & Lokich, J. J. Carcinocythemia (carcinoma cell leukemia). Report of two cases with English literature review. Cancer 53, 11002 (Mar. 1984). 21. Yam, L. & Janckila, A. Immunocytodiagnosis of carcinocythemia in disseminated

breast cancer. Acta Cytol. 31, 6872 (1986).

22. Glaves, D., Huben, R. P. & Weiss, L. Haematogenous dissemination of cells from human renal adenocarcinomas. Br. J. Cancer 57, 325 (Jan. 1988).

23. Zippelius, A. & Pantel, K. RT-PCR-based detection of occult disseminated tu-mor cells in peripheral blood and bone marrow of patients with solid tutu-mors. An overview. Ann. N. Y. Acad. Sci. 906, 110123 (2000).

24. Racila, E. et al. Detection and characterization of carcinoma cells in the blood. Proc. Natl. Acad. Sci. U. S. A. 95, 458994 (May 1998).

25. Allard, W. J. et al. Tumor cells circulate in the peripheral blood of all major carcinomas but not in healthy subjects or patients with nonmalignant diseases. Clin. cancer Res. 10, 6897904 (Oct. 2004).

26. Tibbe, A. G. J., Miller, M. C. & Terstappen, L. W. M. M. Statistical considerations for enumeration of circulating tumor cells. Cytometry. A 71, 15462 (Mar. 2007).

(28)

27. Coumans, F. A. W., Ligthart, S. T., Uhr, J. W. & Terstappen, L. W. M. M. Challenges in the Enumeration and Phenotyping of CTC. Clin. cancer Res. 18, 57118 (Oct. 2012).

28. Coumans, F. A. W., Doggen, C. J. M., Attard, G., de Bono, J. S. & Terstappen, L. W. M. M. All circulating EpCAM+CK+CD45- objects predict overall survival in castration-resistant prostate cancer. Ann. Oncol. 21, 18517 (Sept. 2010). 29. Ligthart, S. T. et al. Unbiased and automated identication of a circulating tumour

cell denition that associates with overall survival. PLoS One 6, e27419 (Jan. 2011).

30. Pierga, J.-Y. et al. High independent prognostic and predictive value of circulating tumor cells compared with serum tumor markers in a large prospective trial in rst-line chemotherapy for metastatic breast cancer patients. Ann. Oncol. 23, 61824 (Mar. 2012).

31. Bidard, F.-C. et al. Assessment of circulating tumor cells and serum markers for progression-free survival prediction in metastatic breast cancer: a prospective ob-servational study. Breast Cancer Res. 14, R29 (2012).

32. Scher, H. I. et al. Circulating tumour cells as prognostic markers in progressive, castration-resistant prostate cancer: a reanalysis of IMMC38 trial data. Lancet Oncol. 10, 233239 (2009).

33. Budd, G. T. et al. Circulating tumor cells versus imagingpredicting overall sur-vival in metastatic breast cancer. Clin. cancer Res. 12, 64039 (Nov. 2006). 34. De Giorgi, U. et al. Circulating tumor cells and [18F]uorodeoxyglucose positron

emission tomography/computed tomography for outcome prediction in metastatic breast cancer. J. Clin. Oncol. 27, 330311 (July 2009).

35. Coumans, F. A. W. Catching metastasizing cells in the act PhD thesis (University of Twente, 2012).

36. Kagan, M. et al. A sample preparation and analysis system for identication of circulating tumor cells in J. Clin. Ligand Assay 25 (2002), 104110.

37. Saucedo-Zeni, N. et al. A novel method for the in vivo isolation of circulating tumor cells from peripheral blood of cancer patients using a functionalized and structured medical wire. Int. J. Oncol. 41, 12411250 (2012).

38. Alix-Panabières, C. et al. Characterization and enumeration of cells secreting tu-mor markers in the peripheral blood of breast cancer patients. J. Immunol. Meth-ods 299, 17788 (Apr. 2005).

39. Alix-Panabieres, C., Riethdorf, S. & Pantel, K. Circulating tumor cells and bone marrow micrometastasis. Clin. Cancer Res. 14, 50135021 (2008).

40. Nagrath, S. et al. Isolation of rare circulating tumour cells in cancer patients by microchip technology. Nature 450, 12359 (Dec. 2007).

41. Stott, S. L. et al. Isolation and characterization of circulating tumor cells from patients with localized and metastatic prostate cancer. Sci. Transl. Med. 2, 25ra23 (Mar. 2010).

42. Gleghorn, J. P. et al. Capture of circulating tumor cells from whole blood of prostate cancer patients using geometrically enhanced dierential immunocapture (GEDI) and a prostate-specic antibody. Lab Chip 10, 279 (Jan. 2010). 43. Den Toonder, J. Circulating tumor cells: the Grand Challenge. Lab Chip 11, 3757

(Feb. 2011).

44. He, W. et al. Quantitation of circulating tumor cells in blood samples from ovar-ian and prostate cancer patients using tumor-specic uorescent ligands. Int. J. Cancer 123, 19681973 (2008).

(29)

16

REFERENCES

45. Campton, D., Ramirez, A., Nordberg, J. & Blau, A. High-recovery multiplex anal-ysis of circulating tumor cells by density-based enrichment, automated platform immunouorescence staining, and digital microscopy. Cancer Res. (2014). 46. Balic, M. et al. Comparison of two methods for enumerating circulating tumor

cells in carcinoma patients. Cytometry B. Clin. Cytom. 68, 2530 (Nov. 2005). 47. Königsberg, R. et al. Circulating tumor cells in metastatic colorectal cancer:

E-cacy and feasibility of dierent enrichment methods. Cancer Lett. 293, 117123 (2010).

48. Königsberg, R. et al. Detection of EpCAM positive and negative circulating tumor cells in metastatic breast cancer patients. Acta Oncol. 50, 700710 (2011). 49. Coumans, F. A. W., van Dalum, G., Beck, M. & Terstappen, L. W. M. M.

Fil-ter characFil-teristics inuencing circulating tumor cell enrichment from whole blood. PLoS One 8, e61770 (Jan. 2013).

50. Desitter, I. et al. A new device for rapid isolation by size and characterization of rare circulating tumor cells. Anticancer Res. 31, 427441 (2011).

51. Vona, G. et al. Isolation by size of epithelial tumor cells : a new method for the im-munomorphological and molecular characterization of circulatingtumor cells. Am. J. Pathol. 156, 5763 (2000).

52. Kahn, H. J. et al. Enumeration of circulating tumor cells in the blood of breast cancer patients after ltration enrichment: correlation with disease stage. Breast Cancer Res. Treat. 86, 23747 (Aug. 2004).

53. Zheng, S. et al. Membrane microlter device for selective capture, electrolysis and genomic analysis of human circulating tumor cells. J. Chromatogr. A 1162, 15461 (Aug. 2007).

54. Lin, H. K. et al. Portable lter-based microdevice for detection and characteriza-tion of circulating tumor cells. Clin. Cancer Res. 16, 50118 (Oct. 2010). 55. Zheng, S. et al. 3D microlter device for viable circulating tumor cell (CTC)

en-richment from blood. Biomed. Microdevices 13, 203213 (2011).

56. Coumans, F. A. W., van Dalum, G., Beck, M. & Terstappen, L. W. M. M. Filtration parameters inuencing circulating tumor cell enrichment from whole blood. PLoS One 8, e61774 (Jan. 2013).

57. Van Rijn, C. J. M. et al. Microsieves made with laser interference lithography for micro-ltration applications. J. Micromechanics Microengineering 9, 170172 (June 1999).

58. Adams, D. L. D. et al. The systematic study of circulating tumor cell isolation using lithographic microlters. RSC Adv. 4, 43344342 (2014).

59. Adams, D. L. et al. Circulating giant macrophages as a potential biomarker of solid tumors. Proc. Natl. Acad. Sci. U. S. A. 111, 35149 (2014).

60. Tan, S. J., Yobas, L., Lee, G. Y. H., Ong, C. N. & Lim, C. T. Microdevice for the isolation and enumeration of cancer cells from blood. Biomed. Microdevices 11, 883892 (2009).

61. Tan, S. J. et al. Versatile label free biochip for the detection of circulating tumor cells from peripheral blood in cancer patients. Biosens. Bioelectron. 26, 17015 (Dec. 2010).

62. Rao, C. G. et al. Expression of epithelial cell adhesion molecule in carcinoma cells present in blood and primary and metastatic tumors. Int. J. Oncol. 27, 4957 (July 2005).

63. Hoeppener, A. E. L. M., Swennenhuis, J. F. & Terstappen, L. W. M. M. Im-munomagnetic separation technologies. Recent results cancer Res. 195, 4358 (Jan. 2012).

(30)

64. Pachmann, K. et al. Monitoring the response of circulating epithelial tumor cells to adjuvant chemotherapy in breast cancer allows detection of patients at risk of early relapse. J. Clin. Oncol. 26, 12081215 (2008).

65. Marrinucci, D. et al. Fluid biopsy in patients with metastatic prostate, pancreatic and breast cancers. Phys. Biol. 9, 016003 (Feb. 2012).

66. Swennenhuis, J. F., Tibbe, A. G. J., Levink, R., Sipkema, R. C. J. & Terstappen, L. W. M. M. Characterization of circulating tumor cells by uorescence in situ hybridization. Cytometry. A 75, 5207 (June 2009).

67. Attard, G. et al. Characterization of ERG, AR and PTEN gene status in circulating tumor cells from patients with castration-resistant prostate cancer. Cancer Res. 69, 29128 (Apr. 2009).

68. Lang, J. E. et al. HER2 status predicts the presence of circulating tumor cells in patients with operable breast cancer. Breast Cancer Res. Treat. 113, 5017 (Feb. 2009).

69. Riethdorf, S. et al. Detection and HER2 expression of circulating tumor cells: prospective monitoring in breast cancer patients treated in the neoadjuvant Gepar-Quattro trial. Clin. Cancer Res. 16, 263445 (May 2010).

70. Ignatiadis, M. et al. HER2-positive circulating tumor cells in breast cancer. PLoS One 6, e15624 (Jan. 2011).

71. Ligthart, S. T. et al. Unbiased quantitative assessment of Her-2 expression of cir-culating tumor cells in patients with metastatic and non-metastatic breast cancer. Ann. Oncol. 24, 12318 (May 2013).

72. Fehm, T. et al. Determination of HER2 status using both serum HER2 levels and circulating tumor cells in patients with recurrent breast cancer whose primary tumor was HER2 negative or of unknown HER2 status. Breast Cancer Res. 9, R74 (Jan. 2007).

73. Meng, S. et al. uPAR and HER-2 gene status in individual breast cancer cells from blood and tissues. Proc. Natl. Acad. Sci. U. S. A. 103, 173615 (Nov. 2006). 74. Hayes, D. F. et al. Circulating tumor cells at each follow-up time point during

therapy of metastatic breast cancer patients predict progression-free and overall survival. Clin. cancer Res. 12, 42184224 (July 2006).

75. Coumans, F. A. W., Ligthart, S. T. & Terstappen, L. W. Interpretation of changes in circulating tumor cell counts. Transl. Oncol. 5, 48691 (2012).

(31)

(32)

CHAPTER

2

Importance of circulating

tumor cells in newly

diagnosed colorectal cancer

G. van Dalum, M.R. de Groot, G.J. Stam, L.F.A. Scholten, W.J.B. Mastboom, I. Vermes, A.G.J. Tibbe, L.W.M.M. Terstappen International Journal of Oncology. in press

Abstract

Background: Presence of circulating tumor cells (CTC) is associ-ated with poor prognosis in patients with metastatic colorectal cancer (CRC). This study was conducted to determine if the presence of CTC prior to surgery and during follow-up in patients with newly diagnosed non-metastatic CRC can identify patients at risk for disease recurrence. Methods: In a prospective single center study 183 patients with newly diagnosed non-disseminated CRC scheduled for surgery were enrolled and followed up for a median of 5.1 years. CTC were enumerated with the CellSearch System in 4 aliquots of 7.5 mL of blood before surgery and at several time points during follow-up after surgery. Results: 1 CTC/ 30 mL of blood were detected in 44 (24%) patients before surgery. Patients with CTC before surgery had a signicant decrease in Recurrence Free Survival (RFS, log-rank test p=0.014) and Colon Can-cer Related Survival (CCRS, p=0.002). Five year RFS dropped from 75% to 61% and ve year CCRS from 83% to 69% for patients with CTC before surgery. The presence of CTC and positive lymphnodes re-mained signicant factors in multivariate analysis for Recurrence Free Survival (RFS). Surprisingly the presence of CTC weeks after surgery was not signicantly associated with RFS and Colon Cancer Related

(33)

20 2.1. INTR ODUCTION

Death (CCRD) whereas CTC 2-3 years after surgery was again signif-icantly associated with RFS and CCRD. Conclusion: The presence of CTC in patients with stage I-III CRC before surgery is associated with a signicant reduction RFS and CCRS. These ndings suggest a role of CTC detection to assess, which patients need adjuvant treatment.

2.1 Introduction

The lifetime risk to develop colorectal carcinoma (CRC) is more than 5% and because current therapy is not curative in all cases this disease is one of the leading causes of death worldwide [1]. Thirty to fty percent of all curative resected CRC patients will develop local recurrence or distant metastatic disease [2, 3]. Adjuvant chemotherapy is used in high risk pa-tients, often dened as patients above stage IIB according to the TNM classication, angio-invasive growth, tumor perforation or obstruction and less than 10 detectable lymph nodes. Adjuvant chemotherapy results in a relative risk reduction of approximately 30% in disease recurrence [2, 4]. Because of a lack of 100% sensitivity and specicity of the known risk factors for disease recurrence, numerous patients receive adjuvant therapy without having presumed micro-metastasis. On the other hand, a subgroup of patients classied as having a low risk for disease recurrence, thus not receiving adjuvant treatment will develop disease recurrence. Better tools are necessary to discriminate between these patient groups. The presence of tumor cells in blood of cancer patients may help to discriminate between these patient groups. The introduction of a validated system for the enu-meration of circulating tumor cells (CTC) [5] enabled prospective clinical studies in both the metastatic and non-metastatic setting. Data from the multicenter prospective studies of CTC in metastatic breast [6], prostate [7] and colorectal [8] and a single center prospective study at diagnosis of breast cancer [9, 10] were reported earlier. These studies demonstrated that CTC are an independent predicting factor for disease free survival and overall survival and these ndings were conrmed by other studies [1116]. Here we report on a single center prospective study in newly diagnosed CRC that was initiated at the same time as the original study in the metastatic colorectal cancer. The frequency of CTC in patients with metastatic col-orectal cancer is extremely low, no CTC were detected with the CellSearch system in 52% of these patients using the FDA cleared protocol for 7.5 mL of blood [8]. In this study we investigate CTC in newly diagnosed patients without overt metastasis, the incidence of detectable CTC is expected to be lower and therefor larger blood volumes will need to be analyzed to detect signicant amount of CTC. To explore whether or not the presence of CTC in newly diagnosed patients could predict recurrence 30 mL of blood was

(34)

21 CHAPTER 2. IMPOR T ANCE OF CTC IN NEWL Y DIA GNOSED CR C

analyzed for the presence of CTC taken before surgery, after surgery and at several time points during a four-year follow-up.

2.2 Materials and methods

2.2.1 Study design and patients

In this double blind single center cohort study 216 patients with colorectal malignancy and 58 patients with benign colorectal disease were enrolled at Medisch Spectrum Twente (MST), Enschede, the Netherlands. The ethics board of Medisch Spectrum Twente, approved the study protocol and all patients provided written informed consent. Patients were included be-tween September 2003 and November 2008. Inclusion criteria were dened as patients age above 18 years, newly diagnosed colorectal cancer without metastases and scheduled for surgery, ECOG performance state 0-1. The main exclusion criterion was presence of malignancy in the ve years before inclusion in the medical history (excluding non-melanoma skin carcinoma or cervix carcinoma in-situ). Thirty-three patients were excluded from analysis because of the following reasons: 15 patients were diagnosed with distant metastasis peri-operative, 15 patients had a malignancy in their medical history, and 3 patients did not have CTC data at the inclusion time point. This resulted in nal cohort of 183 patients. The control group con-sisted of 58 patients undergoing colonoscopy or abdominal surgery in which no malignancy was detected. These patients were included throughout the study period to prevent bias in the sta performing the laboratory CTC analysis. CTC enumeration during follow-up was done coinciding with a routine follow-up visit according to the Dutch guidelines for treatment of colorectal cancer [17]. Treatment intention of all included patients with col-orectal cancer was curative surgery. Peri-operative ndings and pathologic outcome would dene adjuvant therapy. Presence of CTC was blinded and did not inuence adjuvant therapy. Patient records were reviewed in June of 2013 to record whether or not disease recurrence had occurred and if so when and whether or not the patient died and if this was related to colon cancer. The primary end point of the study was to determine a correlation between the presence of CTC prior to surgery and recurrence free survival (RFS). Secondary end points were dened as a correlation of CTC prior to surgery with Colon Cancer Related Death (CCRD) and correlation of blood draws during follow-up after surgery with RFS and CCRD.

2.2.2 Blood collection and CTC detection

Four peripheral blood samples were drawn by venipuncture into a 10 mL CellSave Preservative Tube (Veridex, Raritan NJ). Time points of blood draw were before surgery or colonoscopy (Draw A), after surgery/before adjuvant therapy (Draw B), after adjuvant therapy (Draw C), after one

(35)

22 2.2. MA TERIALS AND METHODS

Table 2.1: Characteristics of the 183 patients and their relation to recurrence free survival (RFS) and colon cancer related death (CCRD) using a log-rank test.

N % %CTC≥1 RFS(p) CCRD(p) T Stage 0.008 0.001 T4 19 10 8a T3 108 59 8a T2 39 21 28a T1 13 7 37a Unknown 4 2 N Stage 0.025 0.006 N2 26 14 27 N1 41 22 34 N0 112 61 19 Unknown 4 2 M Stage M1 0 M0 183 100 24 Histology Grade 0.115 0.125 Poor 11 7 8 Moderate 133 73 28 Good 13 6 37 Unknown 26 14 Adjuvant therapy <0.001 <0.001 Yes 54 30 23 No 129 70 28 Sex 0.510 0.537 Male 116 63 28 Female 67 37 18

Continues Mean Min-Max

Age 66 37-85

Follow-up 60 1-109

a_{Denotes a signicant dierence in coincidence of unfavorable CTC using a χ}2

test.

year (Draw D), after two years (Draw E), after three years (Draw F) and after four years (Draw G). Four aliquots of 7.5 mL were examined for the presence of CTC with the CellSearch system (Veridex). The CTC number was the total of the number found in the four aliquots. Analysis took place within 72 hours after the blood draw. CTC during follow-up was dened for RFS as having one or more CTC in any draw from C until G but before recurrence, for CCRD this was dened as one or more CTC in any draw from C until G. The CellSearch system enriched CTC using antibodies directed against the epithelial cell adhesion antigen (EpCAM) coupled to ferrouids. The enriched cells were uorescently labeled with the nucleic acid dye 4,6-diaminodino-2-phenylindole (DAPI) and phycoerythrin (PE) labeled monoclonal antibodies against cytokeratin 8, 18 and 19 and Allophycocyan (APC) labeled antibodies directed against CD45. Images of CTC candidates were captured by the CellTracks Analyzer II and presented

(36)

to experienced operators for classication and assigned as CTC when the objects were larger than 4 µm, stained with DAPI and cytokeratin, lacked CD45 and had morphological features consistent with that of a cell (5). The operators were blinded to the clinical status of the patient.

2.2.3 Statistical analysis

All patient data were collected in an Access database including demo-graphic parameters such as age and sex, and pathological ndings including histological grade and TNM staging. Follow-up ndings included recur-rence date, adjuvant therapy and last outpatient control visit. The Access database was merged at the moment of nal analysis with the database of analyzed CTC measurements using SPSS version 20.0 and R [18]. A p-value smaller than 0.05 was considered to indicate a signicant dier-ence. All tests were 2-sided. When dividing patients into a favorable and unfavorable group using CTC counts, unfavorable was considered one or more CTC. Kaplan Meier curves for RFS and CCRD were compared using the log-rank test. Between-group dierences in categorical variables were tested by the Pearson Chi-square test. The following signicant univariate prognostic factors were included in a multivariate Cox proportional regres-sion model: T stage, N stage and CTC status. Due to the low numbers of patients with T1, T1 and T2 were grouped together in the multivariate model. The proportional hazard assumption was checked for all factors included in the model. Factors were removed from the multivariate model using stepwise elimination, using P > 0.10 as criteria.

2.3 Results

2.3.1 Patient characteristics and univariate analysis

The follow-up ranged from 1 - 109 months with a mean of 60 months and a median of 61 months. Recurrence of disease was observed in 48 of 183 (26%) patients, 36 (20%) patients died of causes related to colorectal cancer and 23 (12%) patients died of other causes. The median follow-up of the patients alive at the end of the follow-up was 66 months. Patient characteristics and their relation with CTC, RFS and CCRD is shown in table 2.1. Univariate analysis showed a signicant relationship between RFS and CCRD with T-stage, N stage and adjuvant therapy. The other variables histological grade, tumor size and sex were not signicant. T-stage (p=0.016) showed a signicant dierence in coincidence with unfavorable CTC counts.

2.3.2 Frequency of Circulating Tumor Cells

In 44 out of the 183 patients (24%) CTC were detected before surgery. This decreased to 29 (19.9%) in the sample drawn weeks after surgery (and

(37)

24 2.3. RESUL TS Table 2.2: Prev alence of circulating tumor cells before colon cancer surgery an at sev eral time poin ts after surgery . Benign disease A before surgery B after surgery C after adjuv an t therap y D one year after surgery E tw o years after surgery F three years after surgery G four years after surgery CDEF G com bined if before recurrence CDEF G com bined before death N=58 N=183 N=146 N=42 N=116 N=82 N=47 N=16 N=131 N=135 CTC N % N % N % N % N % N % N % N % N % N % 0 50 86 139 76 117 80 32 76 96 83 74 90 42 89 13 81 96 73 97 72 ≥ 1 8 14 44 24 29 20 10 24 20 17 8 10 5 11 3 19 35 27 38 28 1 7 12 28 15 17 12 6 14 11 9 3 4 3 6 2 13 18 14 18 13 2 1 2 11 6 5 3 3 7 3 3 1 1 0 0 0 0 6 5 6 4 3 0 0 0 0 2 1 0 0 0 0 1 1 0 0 0 0 1 1 1 1 4 0 0 0 0 0 0 0 0 1 1 0 0 1 2 0 0 0 0 1 1 >4 0 0 5 3 5 3 1 2 5 4 3 4 1 2 1 6 10 7 11 8

(38)

Table 2.3: Relation between the presence of CTC and re-currence free survival (RFS) or colon cancer related death (CCRD). For each time point during follow-up (FU) the RFS and CCRD for the before surgery draw of the same patients is shown in italic Draw N Unfavorable % RFS(p) CCRD(p) Before surgery 183 24 0.014 0.001 After surgery 143 20 0.940 0.425 (before surgery) 22 0.252 0.029 After adjuvant TX 42 24 0.027 0.009 (before surgery) 24 0.056 0.016 After 1 year 116 27 0.772 0.584 (before surgery) 20 0.056 0.001 After 2 years 42 10 0.001 <0.001 (before surgery) 22 0.194 0.196 After 3 years 47 11 0.091 0.007 (before surgery) 17 0.838 0.717 After 4 years 16 19 0.034 0.004 (before surgery) 25 1.000 0.595 FU before recurrance 131 27 0.018 0.003 (before surgery) 20 0.018 0.004 FU before death 135 28 0.001 <0.001 (before surgery) 21 0.024 0.004

before the initiation of adjuvant therapy when indicated). 10 of the 29 patients with CTC after surgery had CTC before surgery. The number of patients with one or more CTC found at the dierent time points are provided in Table 2.2

2.3.3 Relation between Circulating Tumor Cells and

Recurrence Free Survival or Colon Cancer Related

Death

Patients were divided into those with Favorable (0 CTC) and Unfavorable (>1 CTC) CTC for the dierent time points. Table 3 shows the result of the log-rank test for each of the dierent blood draws for RFS and CCRD. The number of patients participating in the follow-up CTC measurements decreased and to verify potential bias the before surgery results of the same patients are provided in italics below each draw. When comparing Unfavorable CTC to Favorable CTC counts the risk of recurrence (Hazard Ratio=2.07, p=0.016) and CCRD (Hazard Ratio=2.74, p=0.003) was sig-nicantly increased before surgery. The associated Kaplan Meier curves for RFS and CCRD are shown in gure 2.1. The frequency of CTC 1-12 weeks after surgery decreased only slightly (24 to 20%), but their presence was no longer signicant, likewise CTC were not signicant 1 year after

(39)

26 2.3. RESUL TS A B CTC = 0 CTC ≥ 1 p = 0.014 0.00 0.25 0.50 0.75 1.00 0 10 20 30 40 50 60 70 80 90 100 Time (months) Su rv iv a l Pro b a b ili ty

RFS for CTC before surgery

139 126 118 109 95 81 64 35 22 11 6 44 36 32 28 22 20 13 7 6 4 1

CTC = 0 CTC ≥ 1

Number of patients at risk

CTC = 0 CTC ≥ 1 p = 0.002 0.00 0.25 0.50 0.75 1.00 0 10 20 30 40 50 60 70 80 90 100 Time (months) Su rv iv a l Pro b a b ili ty

CCRD for CTC before surgery

138 133 129 121 113 98 78 50 31 16 7 44 39 35 33 29 27 19 10 8 5 1

CTC = 0 CTC ≥ 1

Number of patients at risk

Figure 2.1: Kaplan Meier graphs of Recurrence Free Survival (panel A) and Colon Cancer Related Death (panel B) of colorectal cancer patients with 0 CTC / 30 mL and >1 CTC / 30 mL of blood before surgical intervention. surgery. For the subgroup of patients that received adjuvant therapy the presence of CTC after completion of therapy was signicant for RFS and CCRD, as listed in table 3. A more detailed analysis of the subgroups with and without adjuvant therapy can be found in supplemental table 1, sup-plemental gure 1 and 2. Two and three years after surgery the presence of CTC again was highly signicant for CCRD as shown in the Kaplan Meier curves of gure 2.2. For the before surgery curves for each time point see supplemental gure 1. Table 3 also lists Unfavorable CTC at any of the follow-up time points of blood draw C through F before recurrence or death, both were highly signicant for RFS and CCRD.

2.3.4 Multivariate analysis

Only the presence of CTC before surgery was included in the multivariate analysis. The univariate signicant parameters CTC, N-stage, and T-stage were also included in a Cox proportional hazards model. The multivari-ate regression was performed with a conditional stepwise elimination of the least signicant parameters. For RFS and CCRD CTC and N-stage remained as signicant predictors in the model, see table 2.4. In gure 2.3 both CTC and lymphnodes were combined to dene at risk status in Kaplan Meier curves for RFS and CCRD.

The role and definition of cancer cells in blood

The role and definition

of cancer cells in blood

The role and definition

of cancer cells in blood

Proefschrift

Contents

Introduction

Thesis motivation

Thesis outline

CHAPTER

1

Metastasis and Circulating

Tumor Cells

1.1 Introduction

1.2 Metastatic process

1.3 Circulating tumor cells

1.4 Current CTC detection methods

1.5 Antibody based CTC enrichment

1.6 Density based enrichment

1.7 Size or exibility based enrichment

1.8 CTC detection after erythrocyte lysis

1.9 Comparing dierent enumeration methods

1.10 Detection of treatment targets on CTC

1.11 The future

References

CHAPTER

2

Importance of circulating

tumor cells in newly

diagnosed colorectal cancer

2.1 Introduction

2.2 Materials and methods

2.2.1 Study design and patients

2.2.2 Blood collection and CTC detection

2.2.3 Statistical analysis

2.3 Results

2.3.1 Patient characteristics and univariate analysis

2.3.2 Frequency of Circulating Tumor Cells

2.3.3 Relation between Circulating Tumor Cells and

Recurrence Free Survival or Colon Cancer Related

Death

2.3.4 Multivariate analysis

1.7 Size or exibility based enrichment

1.9 Comparing dierent enumeration methods