ARTICLE IN PRESS Cancer Letters ■■ (2016) ■■–■■

Contents lists available at ScienceDirect

Cancer Letters j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / c a n l e t

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

Q2 Mini-review

Animal models of colorectal cancer with liver metastasis Q1 Bo Young Oh a, Hye Kyung Hong b, Woo Yong Lee a,c, Yong Beom Cho a,c,d,* a

Department of Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea Samsung Biomedical Research Institute, Samsung Medical Center, Sungkyunkwan University, Seoul, Republic of Korea c Department of Health Sciences and Technology, SAIHST, Sungkyunkwan University, Seoul, Republic of Korea d Department of Medical Device Management & Research, SAIHST, Sungkyunkwan University, Seoul, Republic of Korea b

A R T I C L E

I N F O

Keywords: Colorectal cancer Liver metastasis Mouse model Orthotopic model Genetically engineered mouse model Xenografts

A B S T R A C T

Liver metastasis is a leading cause of death in patients with colorectal cancer. Investigating the mechanisms of liver metastasis and control of disease progression are important strategies for improving survival of these patients. Liver metastasis is a multi-step process and relevant models representing these steps are necessary to understand the mechanism of liver metastasis and establish appropriate treatments. Recently, the development of animal models for use in metastasis research has greatly increased; however, there is still a lack of models that sufficiently represent human cancer. Thus, in order to select an optimal model for of a given study, it is necessary to fully understand the characteristics of each animal model. In this review, we describe the mouse models currently used for colorectal cancer with liver metastasis, their characteristics, and their pros and cons. This may help us specify the mechanism of liver metastasis and provide evidence relevant to clinical applications. © 2016 Published by Elsevier Ireland Ltd.

33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

Introduction Colorectal cancer is the third most common malignancy and a leading cause of cancer-related death worldwide [1–4]. The survival of patients with colorectal cancer has improved; however, the mortality rate still ranks third due to metastasis or recurrence [4,5]. Metastases are found in about 20–25% of patients with colorectal cancer at diagnosis, and will develop in about 50% of patients during the course of the disease [6,7]. The liver is the most common site of metastasis for colorectal cancer, and only 15–20% of patients with liver metastases are suitable candidates for surgical resection [8–10]. Several adjunctive therapies have been applied in patients with unresectable liver metastasis, but survival rates remain unfavorable [10]. Thus, investigating the mechanisms of liver metastasis and control of disease progression are important strategies for improving survival of these patients. Metastasis is a multistep process which involves an epithelial–mesenchymal transition, suppression of apoptosis, local invasion and cell migration, angiogenesis and intravasation, and extravasation into distant organs [11]. During these processes, tumor cells continuously interact with the microenvironment of the host cell [12]. Relevant models representing the metastatic characteristics of tumor cells are necessary in understanding these metastatic processes. In particular, organ-specific metastasis models can help us to specify the mechanism of metastasis and, accordingly, establish appropriate treatments.

59 60 61 62

* Corresponding author. Tel.: +82 2 3410 0217; fax: +82 2 3410 1655. E-mail address: [email protected] (Y.B. Cho).

Liver metastasis models have been developed both in vitro and in vivo. In vitro metastasis models are easy to manipulate, inexpensive, and reproducible. These models allow for the manipulation of each step in metastasis, but there are limitations with respect to a comprehensive analysis of the whole process of metastasis [13,14]. In contrast, in vivo metastasis models using animals may more accurately represent the metastatic process and can be genetically manipulated to mimic human cancer. Thus, animal models have been widely used in metastasis research. However, they are associated with ethical concerns and cost burdens [15–17]. Currently, a number of animal models have been developed for the evaluation of the mechanism of liver metastasis and for the establishment of therapeutic strategies [12,15–19]. Such preclinical studies have provided evidence relevant to clinical applications and have contributed to improving the survival of patients with colorectal cancer with liver metastasis. In this review, we focus on the mouse models currently used for colorectal cancer with liver metastasis. Overview of the current understanding of animal models Ideal animal models need to satisfy several conditions in order to be suitable for metastasis research. Animal models of metastasis should allow for optimal tumor take of both primary and metastatic tumors. The tumor take should be predictable and reproducible. They should recapitulate the entire process of human metastatic diseases. Furthermore, they should reflect the genetic alterations with subsequent changes occurring in the metastatic process. Ultimately, they should be available to test and validate novel therapeutics for cancer patients [14,19–21]. However, as no model

http://dx.doi.org/10.1016/j.canlet.2016.01.048 0304-3835/© 2016 Published by Elsevier Ireland Ltd.

Please cite this article in press as: Bo Young Oh, Hye Kyung Hong, Woo Yong Lee, Yong Beom Cho, Animal models of colorectal cancer with liver metastasis, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.01.048

63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91

ARTICLE IN PRESS 2

92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134

B.Y. Oh et al./Cancer Letters ■■ (2016) ■■–■■

meets all these conditions, models should be chosen according to the specific experimental purpose. Several animal models have been used in metastasis research, with mouse models being most commonly used. Mouse models have potential advantages such as anatomical and biological similarities to humans, small size, ease of handling, breeding capacity, short gestation time, cost-effectiveness, and ease of genetic manipulation [14,22–24]. Hence, mouse models have evolved according to the specific experimental purpose required to study the biology of human cancer metastasis. Most metastasis models have been obtained by transplanting cancer cell lines or tissues into mice. The most commonly used types of mice in transplantable models are nude mice and severely compromised immunodeficient (SCID) mice. Nude mice have a mutation of the Foxn1 gene leading to an athymic state. Thus, they exhibit depletion of T cells and impaired T and B cell function. SCID mice have a mutation of the Prkdc gene, resulting in deficiencies in the number and function of both T and B cells. However, they retain intact innate immune system components such as natural killer (NK) cells and macrophages [14,19,25–27]. In addition, SCID-beige mice, non-obese diabetic (NOD)-SCID mice, NOD/ Shi-scid IL2rγnull (NOG) or NOD-SCID gamma (NSG) mice, and recombination activating genes (RAG) mice have been used as mouse models. These mice have different genomic mutations leading to various levels of immunodeficiency. Genetically engineered mice (GEM), which, unlike transplantable models, are generated through alterations in the expression of genes of interest, have also been developed. In these models, organ-specific oncogenes or tumor suppressor genes are manipulated, leading to the spontaneous development of cancer [14,16,27,28]. Transplantable mouse models are divided into syngeneic and xenograft models. Syngeneic models refer to models developed through the inoculation of murine cancer cell lines or tissues into another mouse which has an identical genetic background [27,29]. Syngeneic models are valuable for evaluating the interaction between tumors and their microenvironments. The graft is seldom rejected and they are cost-effective and convenient. However, these models lack the genetic heterogeneity of human tumors [14,27]. In contrast to syngeneic models, xenograft models usually refer to models developed through the inoculation of human cancer cell lines or tissues into immunodeficient mice [14,19]. These models have been widely used in human cancer research, although they have limitations such as deficiencies in immune system function and low rates of engraftment [30–32]. In particular, patient-derived xenograft (PDX)

models have received attention with the rise of personalized medicine. PDX models can reflect the characteristics of each patient and are useful for predicting response to new therapeutics [16,33]. Types of mouse models for colorectal cancer with liver metastasis Spontaneous liver metastasis models There are several ways to establish liver metastasis models (Fig. 1). The first approach for developing liver metastasis is the implantation of cancer cell lines or tissues into the colon (orthotopic transplantation) or subcutaneous layer (ectopic transplantation), which results in the formation of primary tumors at the injection site, followed by the spontaneous development of liver metastasis. The most commonly used cell lines are HCT116 and HT29; other cell lines used include SW480, SW620, and Lovo [12,18,34–37]. Ectopic transplantation is the easiest method for tumor engraftment. However, this model mostly fails to develop liver metastasis because the subcutaneous microenvironment is quite different from that of the colon [38–40]. In contrast, the orthotopic transplantation model is similar to human cancer in terms of histology, vascularity, gene expression, and the metastatic process [41–43]. Thus, this model is mainly used in spontaneous liver metastasis models. In this model, colon cancer cell lines or human tumor tissues can be transplanted into the colonic wall (usually the cecum) of the immuno-compromised mice by subserosal injection or surgical implantation (Fig. 2A and B) [14,19,27]. To ensure the effective development of liver metastasis, in vivo selection of highly metastatic tumor cells is an important strategy. The metastatic potential of the engrafted tumor can be increased by serial passage. In this method, tumorigenic samples are subsequently expanded and re-implanted into the cecum for several generations to obtain higher tumorigenicity and metastatic abilities [19]. Such orthotopic transplantation models with highly metastatic potential have been demonstrated to effectively developed liver metastasis [38,44,45]. However, this selection process and the subsequent in vitro cell culture result in adaptation of tumor cells to growth in a two-dimensional environment rather than in the normal three-dimensional environment [27]. A major advantage of the orthotopic spontaneous liver metastasis model is that metastatic dissemination follows a natural course and utilizes a mechanism like that used in humans. Thus, this model

135

136

Fig. 1. Types of mouse models for colorectal cancer with liver metastasis according to the formation method.

Please cite this article in press as: Bo Young Oh, Hye Kyung Hong, Woo Yong Lee, Yong Beom Cho, Animal models of colorectal cancer with liver metastasis, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.01.048

137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179

ARTICLE IN PRESS B.Y. Oh et al./Cancer Letters ■■ (2016) ■■–■■

180 181

3

Fig. 2. Transplantable models for colorectal cancer with liver metastasis. (A) Orthotopic model with serosal injection of cancer cell lines, (B) Orthotopic model with surgical implantation of cancer tissue, and (C) Experimental model with intra-splenic injection.

182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209

is suitable for predicting drug response in human tumors. This model represents the entire metastatic process from the formation of the primary site to the developing liver metastasis. Thus, it enables the investigation of all steps in the metastatic cascade. It mimics the human tumor microenvironment and provides a realistic level of heterogeneity among tumor cells. In addition, the number of passages required to isolate highly metastatic tumor cells in the spontaneous metastasis model is less than that required for the experimental metastasis model [14,19,28,46–48]. Disadvantages of this model are the low predictability and reproducibility of spontaneous metastasis from colon to liver. It is also difficult to assess the contribution of the immune system to metastatic progression as this model due to the lack of a functional immune system in the xenograft. In addition, it takes a long time to develop liver metastasis, and sometimes resection of the primary tumor is needed to reduce tumor burden and allow for the formation of metastasis [18,19,28,30]. Additionally, cecal injection is technically difficult and carriers a risk of tumor cell leakage or intraluminal spillage (Table 1). Experimental liver metastasis models The second approach to developing liver metastasis is the injection of tumor cells directly into systemic circulation, which induces liver metastasis experimentally. Experimental metastases occur in different organs, depending on the site of injection and the tropism of the injected cell. In experimental liver metastasis models, colon

cancer cells are directly injected into the spleen or the portal vein (Fig. 2C). The tumor cells reach hepatic microcirculation without undergoing the steps of primary tumor growth and intravasation, and are then arrested by size or specific retention factors. These arrested cells undergo cellular extravasation and enter the liver parenchyma or they may grow out while remaining within the vasculature. Some of the injected cells eventually form liver metastases and the rest disappear without colonization in the liver. To increase the efficiency of liver metastasis, the previously described serial passage method is necessary in experimental models as in spontaneous models. There are several reports that highly malignant cells develop liver metastases when human colon cancer cells are injected into the spleens of immunocompromised mice [12,14,19,27,49]. Experimental metastasis models provide several advantages for metastasis research. The tumor cells are injected directly into the systemic circulation, so it takes a short time for metastases to develop. Furthermore, these models are highly reproducible and consistently lead to metastasis formation. Metastasis formation in these models is more efficient compared to the spontaneous metastasis models. Use of these models enables control over the number and types of tumor cells introduced into the circulation, which is important in achieving the experimental endpoint. In addition, the metastatic organ can be targeted to specific sites [14,27,28,50,51]. A major disadvantage of the experimental metastasis models is that these models cannot represent the entire metastatic process. They represent only the late stage of the metastasis cascade, since

210 211 212

Table 1 Advantages and disadvantages of models for liver metastasis according to the type of mouse model.

213

Model types

214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236

Spontaneous model

Experimental model

Patient-derived xenograft model

Genetically engineered model

Advantages ● ● ● ● ● ● ● ● ● ● ● ● ● ●

Natural course of metastatic dissemination Suitability for predicting drug response Identification of all steps in the metastatic cascade Similarity to the human tumor microenvironment Similarity to human disease progression Maintenance of realistic heterogeneity of tumor cells Minimal number of passages required to obtain high metastatic potential Short time required to develop liver metastasis High reproducibility and consistency of metastasis formation Targetable metastatic organs Control over the number of cells delivered Suitability for examining patient-specific drug response Preservation of intratumoral heterogeneity of original tumor Effectiveness in evaluating the initiation and progression of metastasis

● ● ● ● ●

Identification of specific genetic mutation during carcinogenesis Effectiveness in evaluating the early steps in tumorigenesis Similarity of defined mutations to those of in human tumors Intact immune system of mice Species-specific microenvironment

Disadvantages ● ● ● ● ● ●

Low predictability and reproducibility of metastasis Functional immune system deficiency in xenograft Length of time required to develop liver metastasis Possibility of primary tumor resection due to tumor burden Technical difficulties Asynchronous development of liver metastasis

● ● ● ● ● ● ● ● ● ● ● ● ●

Inability to represent the entire metastatic process Occurrence of only the late stage of the metastasis cascade Necessity of serial passage to enhance tissue-specific ability Artificial route of metastatic dissemination Low incidence of metastasis Length of time required to develop liver metastasis High cost and labor-intensive Technical difficulties Immune system deficiency in xenograft Low predictability and reproducibility of liver metastases Length of time required for the development of metastasis Low incidence of metastasis Possibility of embryonic lethality, severe developmental defects, or sterility ● Difficulty in evaluating therapeutic response ● High cost and labor-intensive

Please cite this article in press as: Bo Young Oh, Hye Kyung Hong, Woo Yong Lee, Yong Beom Cho, Animal models of colorectal cancer with liver metastasis, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.01.048

237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263

ARTICLE IN PRESS B.Y. Oh et al./Cancer Letters ■■ (2016) ■■–■■

4

264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329

they bypass the early stages, including primary tumor growth and intravasation of tumor cells. Tumor cells should be isolated through serial passage to enhance their tissue-specific abilities. Metastatic dissemination follows an artificial route, rather than its natural course (Table 1) [14,19,27,51]. PDX liver metastasis models PDX liver metastasis models are established by implanting cancer cells or tissues from patients into immunodeficient mice, in which liver metastasis can then be induced spontaneously or experimentally, as described above. Xenografts derived from cell lines are reproducible, easy to manipulate, and well characterized; however, they do not exhibit tumor heterogeneity or the histopathologic and genetic characteristics of the original tumor [52–54]. In contrast, PDX models better reflect the characteristics and genetic diversity of the original tumor; thus, these models are currently the best preclinical models for testing drug response, especially for patients with refractory cancer. In addition, PDX models can be used for further genomic and pharmacologic studies on personalized treatments [53,55]. Mouse types commonly used in PDX models include nude mice, SCID mice, and NOD/SCID humanized mice [14,16]. PDX models typically have a low engraftment rate, so co-injection of stromal components like Matrigel or human stromal cells have been used to improve the success rate [14]. Subcutaneously transplanted PDX models rarely develop liver metastasis. The most commonly used PDX liver metastasis model is a patient-derived orthotopic xenograft model [45,56]. Hoffman and colleagues found that histologically intact colon cancer tissues obtained surgically from patients and orthotopically implanted into the cecum of nude mice led to the development of liver metastases in PDX models after 10 passages [44,45,57]. Other studies also reported that liver metastasis developed in orthotopic PDX models of colorectal cancer [55,58,59]. A major advantage of PDX metastasis models is that these models can examine patient-specific therapeutic responses to drugs and predict disease course, providing crucial information regarding personalized treatment [16,60]. PDX models also preserve the intratumoral heterogeneity of the original tumor at the cellular and genetic levels compared with xenograft models derived from cancer cell lines [33,52,55]. In addition, PDX metastasis models provide the opportunity to track the initiation and progression of metastasis [60]. Despite these advantages, PDX metastasis models have several disadvantages in metastasis research. PDX models show variable engraftment rates and low liver metastasis rates [14,60]. The necrotic area often present in patient tissues makes engraftment success even more challenging [55]. These models are also time consuming, expensive, and technically challenging [16]. In addition, the immune system is impaired in PDX models, which influences cancer progression. This drawback is partially overcome by using NOD/SCID humanized mice via injection of peripheral blood or bone marrow cells, but even in humanized mouse models the immune system is not fully restored (Table 1) [16,55,60]. Genetically engineered liver metastasis models The other approach to developing liver metastasis is the use of genetically engineered mouse models (GEMMs), which may serve as alternatives to human cancer xenografts. Xenografts have been used to gain an understanding of human cancer progression. However, the use of GEMMs in immunocompetent mice has been suggested depending on the emphasis on the importance of the tumor microenvironment on tumor progression. GEMMs are generated through alterations in genomic expression, such as the activation of oncogenes and the inactivation of tumor suppressor genes. They recapitulate the histologic, genomic, transcriptomic, and

proteomic spectrum of human cancers. Therefore, they provide an understanding of entire carcinogenic progress and the mechanisms of specific cancer-related genes [23,61–63]. However, they usually cannot fully reproduce the genetic complexity of human tumors. GEMMs of colorectal cancer usually use germline or tissuewide genetic modification, which are useful in the study of hereditary colorectal cancer [17]. However, somatic mutations develop in sporadic colorectal cancer, which corresponds to about 80% of all colorectal cancer. Thus, somatically engineered mouse models are needed to investigate the carcinogenesis of colorectal cancer [64]. Most GEMMs for colorectal cancer have been generated by inducing mutations of adenomatous polyposis coli (APC) upon loss of heterozygosity [65]. APC, a tumor suppressor, negatively regulates beta-catenin concentrations and interacts with E-cadherin [66]. Although APC mutations can lead to the formation of invasive colorectal cancer, liver metastases have not been demonstrated in these models [65,67]. Recently, a GEMM of colorectal cancer was established that develops liver metastases with considerable efficiency. This model was generated by Adeno-Cre injection into the colon of LSL-KRASG12V/ Apcflox/flox mice. Adenoviral administration of Cre into the colon induces the loss of the APC tumor suppressor, activation of oncogenic KRASG12V, and the generation of sporadic colorectal cancer followed by liver metastases. This model limits the overgrowth of cancer cells, therefore avoiding premature lethality, and leads to tumor progression and liver metastases [64]. There are significant advantages to the use of GEMMs. These models provide useful information about the effects of specific genetic mutations during carcinogenesis. Compared with transplantable models, GEMMs more accurately represent the natural course of tumor development and the interaction between tumor cells and tissue microenvironments [14,16,68]. Moreover, they are effective for evaluating the early steps in tumorigenesis. In addition, these models have intact immune systems and speciesspecific microenvironments [69]. A major disadvantage of genetically engineered liver metastasis models is that liver metastases seldom develop and their occurrence is not predictable [19]. Even if liver metastases occur, they require a long period of time to do so [27,70]. Mutation of specific genes can lead to embryonic lethality, severe developmental defects, or sterility prior to the development of metastases [24,71]. A potential limitation of GEMMs is that it is difficult to evaluate therapeutic responses because these models show a low incidence of and take a long time to develop metastasis [72,73]. In addition, they are expensive and labor-intensive (Table 1) [16,68].

330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375

Monitoring metastasis in animal models Histologic evaluation Tumor growth is measured by histologic analysis (Fig. 3). The extent of tumor growth is determined by measuring the area of liver tissue that has been replaced by metastatic tumor tissue [74,75]. This method can quantitatively determine the extent of liver metastasis; however, it does have several disadvantages. A major disadvantage is that it can only be performed postmortem. Thus, it cannot be used to sequentially evaluate the process of tumor growth and it is difficult to assess therapeutic response. In addition, this is a labor-intensive and time-consuming method [15,19]. Histologic analysis is also useful in the molecular characterization of liver metastasis. This method can validate similarities in the genetic or histologic features between the primary and metastatic tumors. It is particularly suitable for investigating human tumor behavior in mice [52,76]. This supports the importance of animal models as preclinical tools for metastatic biology.

Please cite this article in press as: Bo Young Oh, Hye Kyung Hong, Woo Yong Lee, Yong Beom Cho, Animal models of colorectal cancer with liver metastasis, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.01.048

376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394

ARTICLE IN PRESS B.Y. Oh et al./Cancer Letters ■■ (2016) ■■–■■

395 396

5

Fig. 3. Monitoring liver metastasis in animal models. (A) Histologic evaluation (macroscopic and microscopic) and (B) Imaging evaluation (MRI, PET, and bioluminescence imaging).

397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433

Imaging evaluation

Conclusion

Several imaging techniques have been developed for use in animal models and, consequently, the limitations of histologic analysis can be overcome (Fig. 3). In contrast to histologic analysis, imaging techniques are noninvasive and sensitive, so they can be performed several times. These techniques allow visualization of the entire process of metastasis and the response to therapy [14,19]. Imaging is a relevant method for monitoring metastasis in animal models. Conventional imaging techniques, such as magnetic resonance imaging (MRI), computed tomography (CT), positron emission tomography (PET), and single photon emission tomography (SPECT), have been widely used in monitoring metastasis formation in animal models [77–79]. MRI and CT provide high resolution threedimensional anatomical images, while PET provides high sensitivity and specificity for detecting metastatic lesions. MRI is particularly useful in monitoring liver metastasis and therapeutic response [52,78]. Currently, in vivo optical imaging techniques, such as bioluminescence imaging and fluorescence imaging, are being developed. Bioluminescence imaging allows for the effective detection of tumor cells expressing firefly luciferase, a light-emitting photoprotein. This imaging can be performed in a sequential manner to visualize intrahepatic tumor growth. In addition, it is very sensitive, fast, and easy to manipulate [15,19]. Many studies have reported the usefulness of bioluminescence imaging in liver metastasis animal models [15,18,80]. Fluorescence imaging has been used to detect the growth of tumors expressing the Aequorea victoria green fluorescent protein (GFP). This fluorescence marker is genetically-encoded and continuously expressed in tumor cells. Thus, it provides the ability to follow metastatic cells in real time [14,19,27]. Many studies have monitored liver metastasis of colon cancer using this method [81–83]. These optical imaging methods allow for monitoring of single metastatic cells in vivo and provide new insights into the biology of metastasis, especially during the early stage of liver metastasis.

Mouse models for colorectal cancer with liver metastasis, such as transplantable models (spontaneous and experimental models) or GEMMs, have contributed to our understanding of the metastatic progress. The ideal liver metastasis models should permit the predictable and reproducible establishment of liver metastasis. Furthermore, they should effectively represent the human tumor microenvironment and be available to validate novel therapeutics for cancer patients. However, there are still limitations in the generation of ideal models for human liver metastasis. Fortunately, the development of sequence-specific genome editing techniques, such as the clustered regularly interspaced short palindromic repeats (CRISPR)-Cas system, provides a new means for the development of liver metastasis models [84–86]. Ultimately, the goal of such animal models is to design therapeutic approaches for patients with refractory colorectal cancer. Additional research should be conducted to develop next-generation models of human cancer, as ultimately these efforts will promote further clinical investigation.

434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453

Conflict of interest None declared.

454 455 456 457

Acknowledgement This study was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and Future Planning (2015R1A2A2A01003225).

458 459 460 461 462 463 464

References [1] M.Y. Huang, et al., DPYD, TYMS, TYMP, TK1, and TK2 genetic expressions as response markers in locally advanced rectal cancer patients treated with

Please cite this article in press as: Bo Young Oh, Hye Kyung Hong, Woo Yong Lee, Yong Beom Cho, Animal models of colorectal cancer with liver metastasis, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.01.048

465 466 467 468

ARTICLE IN PRESS B.Y. Oh et al./Cancer Letters ■■ (2016) ■■–■■

6

469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554

[2]

[3]

[4] [5] [6] [7]

Q3

[8] [9]

[10]

Q4 [11] [12]

Q5

[13] [14] [15] [16] [17] [18] [19] [20] [21]

Q6

[22] [23] [24] [25] [26] [27] [28] [29] [30] [31]

[32]

[33]

[34]

[35]

[36]

[37]

fluoropyrimidine-based chemoradiotherapy, Biomed Res. Int. 2013 (2013) 931028. Z.J. Zhang, et al., Reduced risk of colorectal cancer with metformin therapy in patients with type 2 diabetes: a meta-analysis, Diabetes Care 34 (10) (2011) 2323–2328. X. Sui, et al., Use of metformin alone is not associated with survival outcomes of colorectal cancer cell but AMPK activator AICAR sensitizes anticancer effect of 5-fluorouracil through AMPK activation, PLoS ONE 9 (5) (2014) e97781. A. Jemal, et al., Global cancer statistics, CA Cancer J. Clin. 61 (2) (2011) 69– 90. B.Y. Oh, et al., Role of beta1-integrin in colorectal cancer: case-control study, Ann. Coloproctol. 30 (2) (2014) 61–70. R. Siegel, et al., Cancer statistics, 2014, CA Cancer J. Clin. 64 (1) (2014) 9–29. H. Kamiyama, et al., Molecular biomarkers for the detection of metastatic colorectal cancer cells, World J. Gastroenterol. 20 (27) (2014) 8928–8938. W. Li, et al., miRNA-99b-5p suppresses liver metastasis of colorectal cancer by down-regulating mTOR, Oncotarget 6 (27) (2015) 24448–24462. B. Lintoiu-Ursut, A. Tulin, S. Constantinoiu, Recurrence after hepatic resection in colorectal cancer liver metastasis – review article, J. Med. Life 8 Spec Issue (2015) 12–14. H. Chen, et al., Silencing of plasminogen activator inhibitor-1 suppresses colorectal cancer progression and liver metastasis, Surgery 158 (6) (2015) 1704–1713. H.J. Wanebo, et al., Meeting the biologic challenge of colorectal metastases, Clin. Exp. Metastasis 29 (7) (2012) 821–839. K. Ishizu, et al., Development and characterization of a model of liver metastasis using human colon cancer HCT-116 cells, Biol. Pharm. Bull. 30 (9) (2007) 1779–1783. D.M. van Marion, et al., Studying cancer metastasis: existing models, challenges and future perspectives, Crit. Rev. Oncol. Hematol. 97 (2016) 107–117. M. Saxena, G. Christofori, Rebuilding cancer metastasis in the mouse, Mol Oncol 7 (2) (2013) 283–296. N. Smakman, et al., Validation of bioluminescence imaging of colorectal liver metastases in the mouse, J. Surg. Res. 122 (2) (2004) 225–230. A. Richmond, Y. Su, Mouse xenograft models vs GEM models for human cancer therapeutics, Dis. Model. Mech. 1 (2–3) (2008) 78–82. M.M. Taketo, W. Edelmann, Mouse models of colon cancer, Gastroenterology 136 (3) (2009) 780–798. W.Y. Lee, et al., Comparison of colorectal cancer in differentially established liver metastasis models, Anticancer Res. 34 (7) (2014) 3321–3328. M.W. Heijstek, O. Kranenburg, I.H. Borel Rinkes, Mouse models of colorectal cancer and liver metastases, Dig. Surg. 22 (1–2) (2005) 16–25. B. Hann, A. Balmain, Building ‘validated’ mouse models of human cancer, Curr. Opin. Cell Biol. 13 (6) (2001) 778–784. M.P. Rosenberg, D. Bortner, Why transgenic and knockout animal models should be used (for drug efficacy studies in cancer), Cancer Metastasis Rev. 17 (3) (1998) 295–299. L. He, et al., Mouse models of liver cancer: progress and recommendations, Oncotarget 6 (27) (2015) 23306–23322. K.K. Frese, D.A. Tuveson, Maximizing mouse cancer models, Nat. Rev. Cancer 7 (9) (2007) 645–658. K. Maddison, A.R. Clarke, New approaches for modelling cancer mechanisms in the mouse, J. Pathol. 205 (2) (2005) 181–193. M. Cekanova, K. Rathore, Animal models and therapeutic molecular targets of cancer: utility and limitations, Drug Des. Devel. Ther. 8 (2014) 1911–1921. C.L. Morton, P.J. Houghton, Establishment of human tumor xenografts in immunodeficient mice, Nat. Protoc. 2 (2) (2007) 247–250. C. Khanna, K. Hunter, Modeling metastasis in vivo, Carcinogenesis 26 (3) (2005) 513–523. G. Francia, et al., Mouse models of advanced spontaneous metastasis for experimental therapeutics, Nat. Rev. Cancer 11 (2) (2011) 135–141. A. Fantozzi, G. Christofori, Mouse models of breast cancer metastasis, Breast Cancer Res. 8 (4) (2006) 212. J. Jonkers, P.W. Derksen, Modeling metastatic breast cancer in mice, J. Mammary Gland Biol. Neoplasia 12 (2–3) (2007) 191–203. N. Cabioglu, et al., CXCL-12/stromal cell-derived factor-1alpha transactivates HER2-neu in breast cancer cells by a novel pathway involving Src kinase activation, Cancer Res. 65 (15) (2005) 6493–6497. A. Orimo, et al., Stromal fibroblasts present in invasive human breast carcinomas promote tumor growth and angiogenesis through elevated SDF-1/CXCL12 secretion, Cell 121 (3) (2005) 335–348. B.Y. Oh, et al., Correlation between tumor engraftment in patient-derived xenograft models and clinical outcomes in colorectal cancer patients, Oncotarget 6 (18) (2015) 16059–16068. K. Flatmark, et al., Twelve colorectal cancer cell lines exhibit highly variable growth and metastatic capacities in an orthotopic model in nude mice, Eur. J. Cancer 40 (10) (2004) 1593–1598. N. Guilbaud, et al., Marked antitumor activity of a new potent acronycine derivative in orthotopic models of human solid tumors, Clin. Cancer Res. 7 (8) (2001) 2573–2580. J.Y. Park, et al., Fluorescent-antibody targeting of insulin-like growth factor-1 receptor visualizes metastatic human colon cancer in orthotopic mouse models, PLoS ONE 11 (1) (2016) e0146504. H.N. Chen, et al., PDLIM1 stabilizes the E-cadherin/beta-catenin complex to prevent epithelial-mesenchymal transition and metastatic potential of colorectal cancer cells, Cancer Res. (2015).

[38] X. Fu, et al., Extensive liver metastasis from human colon cancer in nude and SCID mice after orthotopic onplantation of histologically-intact human colon carcinoma tissue, Anticancer Res. 12 (5) (1992) 1395–1397. [39] M. Pocard, et al., Efficiency of orthotopic xenograft models for human colon cancers, In Vivo 10 (5) (1996) 463–469. [40] T. Kubota, Metastatic models of human cancer xenografted in the nude mouse: the importance of orthotopic transplantation, J. Cell. Biochem. 56 (1) (1994) 4–8. [41] M.C. Bibby, Orthotopic models of cancer for preclinical drug evaluation: advantages and disadvantages, Eur. J. Cancer 40 (6) (2004) 852–857. [42] C. Khanna, et al., An orthotopic model of murine osteosarcoma with clonally related variants differing in pulmonary metastatic potential, Clin. Exp. Metastasis 18 (3) (2000) 261–271. [43] M.V. Cespedes, et al., Mouse models in oncogenesis and cancer therapy, Clin. Transl. Oncol. 8 (5) (2006) 318–329. [44] B. Rashidi, et al., A nude mouse model of massive liver and lymph node metastasis of human colon cancer, Anticancer Res. 20 (2A) (2000) 715–722. [45] F.X. Sun, et al., An ultra-metastatic model of human colon cancer in nude mice, Clin. Exp. Metastasis 17 (1) (1999) 41–48. [46] R. Munoz, et al., Highly efficacious nontoxic preclinical treatment for advanced metastatic breast cancer using combination oral UFT-cyclophosphamide metronomic chemotherapy, Cancer Res. 66 (7) (2006) 3386–3391. [47] W. Cruz-Munoz, et al., Development of a preclinical model of spontaneous human melanoma central nervous system metastasis, Cancer Res. 68 (12) (2008) 4500–4505. [48] G. Francia, et al., Long-term progression and therapeutic response of visceral metastatic disease non-invasively monitored in mouse urine using beta-human choriogonadotropin secreting tumor cell lines, Mol. Cancer Ther. 7 (10) (2008) 3452–3459. [49] I.J. Fidler, Critical factors in the biology of human cancer metastasis: twentyeighth G.H.A. Clowes memorial award lecture, Cancer Res. 50 (19) (1990) 6130–6138. [50] A.F. Chambers, et al., Critical steps in hematogenous metastasis: an overview, Surg. Oncol. Clin. N. Am. 10 (2) (2001) 243–255, vii. [51] Y. Kang, Analysis of cancer stem cell metastasis in xenograft animal models, Methods Mol. Biol. 568 (2009) 7–19. [52] Y.B. Cho, et al., Colorectal cancer patient-derived xenografted tumors maintain characteristic features of the original tumors, J. Surg. Res. 187 (2) (2014) 502–509. [53] E. Marangoni, et al., A new model of patient tumor-derived breast cancer xenografts for preclinical assays, Clin. Cancer Res. 13 (13) (2007) 3989–3998. [54] T. John, et al., The ability to form primary tumor xenografts is predictive of increased risk of disease recurrence in early-stage non-small cell lung cancer, Clin. Cancer Res. 17 (1) (2011) 134–141. [55] I. Puig, et al., A personalized preclinical model to evaluate the metastatic potential of patient-derived colon cancer initiating cells, Clin. Cancer Res. 19 (24) (2013) 6787–6801. [56] Y. Hiroshima, et al., Establishment of a patient-derived orthotopic Xenograft (PDOX) model of HER-2-positive cervical cancer expressing the clinical metastatic pattern, PLoS ONE 10 (2) (2015) e0117417. [57] B. Rashidi, et al., An orthotopic mouse model of remetastasis of human colon cancer liver metastasis, Clin. Cancer Res. 6 (6) (2000) 2556–2561. [58] K. Morikawa, et al., In vivo selection of highly metastatic cells from surgical specimens of different primary human colon carcinomas implanted into nude mice, Cancer Res. 48 (7) (1988) 1943–1948. [59] S. Julien, et al., Characterization of a large panel of patient-derived tumor xenografts representing the clinical heterogeneity of human colorectal cancer, Clin. Cancer Res. 18 (19) (2012) 5314–5328. [60] D. Siolas, G.J. Hannon, Patient-derived tumor xenografts: transforming clinical samples into mouse models, Cancer Res. 73 (17) (2013) 5315–5319. [61] J. Carretero, et al., Integrative genomic and proteomic analyses identify targets for Lkb1-deficient metastatic lung tumors, Cancer Cell 17 (6) (2010) 547– 559. [62] E.S. Martin, et al., Development of a colon cancer GEMM-derived orthotopic transplant model for drug discovery and validation, Clin. Cancer Res. 19 (11) (2013) 2929–2940. [63] T. Van Dyke, T. Jacks, Cancer modeling in the modern era: progress and challenges, Cell 108 (2) (2002) 135–144. [64] K.E. Hung, et al., Development of a mouse model for sporadic and metastatic colon tumors and its use in assessing drug treatment, Proc. Natl. Acad. Sci. U.S.A. 107 (4) (2010) 1565–1570. [65] P. Rampetsreiter, E. Casanova, R. Eferl, Genetically modified mouse models of cancer invasion and metastasis, Drug Discov. Today Dis. Model 8 (2–3) (2011) 67–74. [66] S.D. Markowitz, M.M. Bertagnolli, Molecular origins of cancer: molecular basis of colorectal cancer, N. Engl. J. Med. 361 (25) (2009) 2449–2460. [67] J. Jonkers, A. Berns, Conditional mouse models of sporadic cancer, Nat. Rev. Cancer 2 (4) (2002) 251–265. [68] G. Dranoff, Experimental mouse tumour models: what can be learnt about human cancer immunology?, Nat. Rev. Immunol. 12 (1) (2012) 61–66. [69] B. Firestone, The challenge of selecting the ‘right’ in vivo oncology pharmacology model, Curr. Opin. Pharmacol. 10 (4) (2010) 391–396. [70] J.E. Talmadge, et al., Murine models to evaluate novel and conventional therapeutic strategies for cancer, Am. J. Pathol. 170 (3) (2007) 793–804. [71] R.D. Palmiter, R.L. Brinster, Germ-line transformation of mice, Annu. Rev. Genet. 20 (1986) 465–499.

Please cite this article in press as: Bo Young Oh, Hye Kyung Hong, Woo Yong Lee, Yong Beom Cho, Animal models of colorectal cancer with liver metastasis, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.01.048

555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640

ARTICLE IN PRESS B.Y. Oh et al./Cancer Letters ■■ (2016) ■■–■■

641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660

[72] S.G. Brodie, et al., Multiple genetic changes are associated with mammary tumorigenesis in Brca1 conditional knockout mice, Oncogene 20 (51) (2001) 7514–7523. [73] V. Stambolic, et al., High incidence of breast and endometrial neoplasia resembling human Cowden syndrome in pten ± mice, Cancer Res. 60 (13) (2000) 3605–3611. [74] E.A. te Velde, et al., Enhanced antitumour efficacy by combining conventional chemotherapy with angiostatin or endostatin in a liver metastasis model, Br. J. Surg. 89 (10) (2002) 1302–1309. [75] Y. van Hensbergen, et al., Rapid stereology based quantitative immunohistochemistry of dendritic cells in lymph nodes: a methodological study, Anal. Cell. Pathol. 22 (3) (2001) 143–149. [76] L. Resor, T.J. Bowen, A. Wynshaw-Boris, Unraveling human cancer in the mouse: recent refinements to modeling and analysis, Hum. Mol. Genet. 10 (7) (2001) 669–675. [77] P. Dubey, et al., Quantitative imaging of the T cell antitumor response by positron-emission tomography, Proc. Natl. Acad. Sci. U.S.A. 100 (3) (2003) 1232–1237. [78] J.S. Lewis, et al., Small animal imaging. Current technology and perspectives for oncological imaging, Eur. J. Cancer 38 (16) (2002) 2173–2188.

7

[79] R. Weissleder, Scaling down imaging: molecular mapping of cancer in mice, Nat. Rev. Cancer 2 (1) (2002) 11–18. [80] M.K. Nyati, et al., The potential of 5-fluorocytosine/cytosine deaminase enzyme prodrug gene therapy in an intrahepatic colon cancer model, Gene Ther. 9 (13) (2002) 844–849. [81] T. Chishima, et al., Cancer invasion and micrometastasis visualized in live tissue by green fluorescent protein expression, Cancer Res. 57 (10) (1997) 2042– 2047. [82] R.M. Hoffman, Orthotopic transplant mouse models with green fluorescent protein-expressing cancer cells to visualize metastasis and angiogenesis, Cancer Metastasis Rev. 17 (3) (1998) 271–277. [83] M. Yang, et al., Whole-body optical imaging of green fluorescent proteinexpressing tumors and metastases, Proc. Natl. Acad. Sci. U.S.A. 97 (3) (2000) 1206–1211. [84] W. Xue, et al., CRISPR-mediated direct mutation of cancer genes in the mouse liver, Nature 514 (7522) (2014) 380–384. [85] F.J. Sanchez-Rivera, T. Jacks, Applications of the CRISPR-Cas9 system in cancer biology, Nat. Rev. Cancer 15 (7) (2015) 387–395. [86] H. Mou, et al., Precision cancer mouse models through genome editing with CRISPR-Cas9, Genome Med. 7 (1) (2015) 53.

Please cite this article in press as: Bo Young Oh, Hye Kyung Hong, Woo Yong Lee, Yong Beom Cho, Animal models of colorectal cancer with liver metastasis, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.01.048

661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680