Multi-parametric magnetic resonance imaging evaluation of a novel ex ovo chick chorioallantoic membrane model for cancer research

Introduction

Animal experiments play a crucial role in the clinical development of novel drugs and technologies. Adhering to the principles of 3Rs—replacement, reduction, and refinement—in animal experimentation is essential for minimizing animal usage whenever possible (1-3). Chicken embryos are not considered sentient in the early to mid-stages of development; experiments on them are conducted without ethical concerns, even in later stages. The chick embryo’s immune system experiences delayed maturation, and even after hatching, certain immune functions continue to be underdeveloped, categorizing the embryo as an inherently immunodeficient organism (4,5). In the 1910s, Rous et al. achieved successful transplantation of chicken sarcoma tissue onto the chorioallantoic membrane (CAM) (6,7), signifying the commencement of a new era of CAM transplantation experiments. Throughout the last century, it has been shown that nearly all human tumor cells can be effectively engrafted in CAM with remarkable success rates (8). The CAM model has gained increasing attention in cancer research over recent years due to its numerous advantages over traditional models such as nude mice or zebrafish, including accelerated tumor growth, higher tumor take rates, reduced costs, and exemption from ethical review processes (9-11). Consequently, this CAM assay exhibits considerable promise for future applications and has the potential to serve as a partial substitute for mammalian animal experiments in subsequent research endeavors.

The current research is chiefly concerned with the methodologies employed in establishing the model, yet there is a significant paucity of systematic exploration regarding the growth characteristics of nodules within said model. This gap poses potential challenges for future applications of the model. For instance, beyond the initial 7-day developmental period of the CAM, and considering its gradual atrophy during the later stages of chicken embryo development, CAM-based studies typically have a relatively short experimental timeframe, concentrated in the early to mid-developmental stages (9,12). Therefore, it is imperative to ascertain the optimal time points for conducting experiments. For example, in drug sensitivity studies, administering drugs before the development of vascularization in the nodule may not only fail to demonstrate the efficacy of the drug but also potentially increase embryo mortality. Conversely, delaying drug administration could impede the observation of long-term effects caused by the drug. Additionally, when conducting imaging studies with the CAM tumor model, it is essential to consider and account for any residual Matrigel in nodules since overlooking its effects may compromise the scientific validity of experiments (12,13).

Currently, imaging studies of the CAM model primarily focus on evaluating nodules at a single fixed time point (14,15), with limited comparative analysis across different time points. These investigations often utilize the limited availability of specialized equipment, such as 7.0 Tesla (T) magnetic resonance imaging (MRI) scanners. However, there is a noticeable lack of exploration into functional imaging techniques (14,16). It remains uncertain whether conventional field-strength MRI (1.5–3.0 T) can accurately visualize nodules in the CAM model and whether functional imaging techniques can provide additional insights into these nodules. In this study, conventional 3.0 T MRI was utilized to perform multi-parametric imaging and pathological comparisons of CAM tumor models at different stages of growth (the study flow is shown in Figure 1). The primary objective of this study was to investigate the growth characteristics and imaging features of the nodules in the CAM model, with a secondary focus on quantitatively analyzing these differences to establish a theoretical basis for future research. We present this article in accordance with the ARRIVE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2024-2612/rc).

Figure 1 A timeline of the experiment. Fertilized eggs were incubated in ovo for 3 days before being transferred to a custom-designed chamber for the following experiment. Tumor cell implantation was performed on day 7. Model examinations were conducted on days 3, 5, 7, and 9 following tumor implantation. *, examination time point. MRI, magnetic resonance imaging.

Methods

Chick embryo and tumor model preparation

This study was conducted in accordance with the People’s Republic of China National Standard (GB/T 35892-2018), which categorizes chick embryos as non-animal, thereby exempting ethical approval. A total of 50 fertilized eggs of comparable weight were purchased from a local animal research facility (Pukou Laifu Animal Breeding Farm, Nanjing, China). The eggs were incubated in a cell incubator (Wiggens, Beijing, China) for three days under optimal conditions (39 ℃ and 60–80% humidity). Following this preliminary incubation period, the embryos were transferred to a customized chamber (Figure 2, A1-A3) for the remaining experiments, where the same environmental conditions were maintained. Daily viability checks were performed, with deceased embryos removed.

Figure 2 Diagram of the customized chamber and procedures for embryo transplantation (A1-A3), and tumor implantation (B1-B4). (A1) Components of the chamber: a1.1: inner ring (9 cm diameter); a1.2: outer ring (10 cm diameter); a1.3: snap button (2.5 cm diameter); a1.4: latex film (Beilile condom, China); a1.5: holder cup (9.5 cm diameter, 2.5 cm height, with a 2-cm central hole for the snap button). (A2) Sterilization process for chamber: a2.1: clean the surface coating of the latex using soapy water; a2.2: disinfect using an alcohol spray; a2.3: place the chamber in a laminar flow cabinet to air-dry, followed by 30 minutes of ultraviolet irradiation for additional disinfection. (A3) After 3 days of culture, carefully transplant the chick embryo into the chamber then covered with a 10-mm petri dish (Thermo Fisher Scientific, Waltham, MA, USA), ensuring the yolk sac integrity. (B1) Identify a well-vascularized area away from the embryo proper (star). (B2) Gently scrape the area with a cotton swab until slight bleeding occurs (indicated by an arrow). (B3) Place a silicone ring (arrowhead) at the site. (B4) Introduce the cell-Matrigel mixture to establish the tumor model.

Human pancreatic cancer cells (MIA PaCa-2, RRID: CVCL_0428) obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China) were harvested during the logarithmic growth phase, washed, and centrifuged for subsequent experiments. Then, the cells were resuspended in a mixture of Matrigel (5 mL/vial; ABW, Shanghai, China) and high-glucose Dulbecco’s modified Eagle medium (DMEM, 4.5 g/L; Gibco, Waltham, MA, USA) at a ratio of 2×10⁶ cells: 50 µL Matrigel: 50 µL culture medium, and stored at 4 ℃ for future implantation. Surviving embryos on ED7 were selected, and a sterile cotton swab was used to cautiously create an incision in the CAM vessels until visible bleeding occurred. The next step involved positioning a silicone ring (outer diameter: 7 mm, inner diameter: 5 mm) at the site of the breach, followed by implantation of 100 µL of the cell mixture within the ring (Figure 2, B1-B4). Once confirmed that there was no active bleeding from the CAM, the embryos were returned to an incubator for further incubation. The survival rate of embryos was recorded daily. On ED10, the silicone rings were removed to create more space for tumor growth.

Model grouping and mpMRI acquisition

At each examination time point (ED10, 12, and 14), eight randomly selected tumor-bearing embryos were scanned via a 3.0 T MRI system (uMR770, United Imaging, Shanghai, China) and subsequently sacrificed for pathological assessment. The remaining embryos were cultured until ED16 when all surviving models underwent mpMRI scanning, followed by pathological analysis.

The tumor-bearing embryos were positioned directly in the United Imaging wrist coil for scanning without requiring anesthesia and additional fixation. For contrast-enhanced MRI, 10 µL gadovist (7.5 mL/vial; Hengrui Medicine, Lianyungang, China) was applied dropwise to the CAM via micro-syringe, with imaging conducted 15 minutes later. The detailed MRI acquisition parameters are listed in Table 1.

Image analysis

The MR image data were imported into RadiAnt DICOM Viewer software (version 2023.1, RRID: SCR_022894; Medixant, Poznań, Poland). Tumor dimensions were measured on contrast-enhanced images, including the maximum diameter (a) and minimum diameter (b). Tumor volume (V) was calculated using the following formula: V = 0.5×a×b². On the uWS-MR workstation (R005, United Imaging, Shanghai, China), regions of interest (ROIs) were manually delineated on contrast-enhanced images. These ROIs were subsequently copied to the corresponding apparent diffusion coefficient (ADC) and T1 mapping sequences for quantitative measurements of ADC and T1 values.

Pathological examination

Following the MRI scans, euthanasia was performed on the chick embryos via decapitation. Nodules were then excised and precisely sectioned along their largest cross-sectional area in the coronal plane. The sections were immersed in a 4% formaldehyde solution (500 mL/vial; Nanjing SenBeiJia Biological Technology Co., Ltd., Nanjing, China) for 24 hours to achieve optimal tissue fixation. Hematoxylin and eosin (H&E) staining was conducted on the sections for histological analysis. After staining, the sections were scanned using a high-resolution digital pathology slide scanner (KF-PRO-120, KFBIO, Ningbo, China). The scanned images were analyzed using the K-Viewer software (version 1.7.1.1, KFBIO, Ningbo, China). Preliminary observations were made at low magnification (2.5×) to assess the overall lesion, followed by detailed examination of regions of interest at higher magnification (6×).

Statistical analysis

The statistical analysis was conducted using the software SPSS 20.0 (RRID: SCR_002865, IBM Corp., Armonk, NY, USA). Continuous variables were presented as mean ± standard deviation (mean ± SD). One-way analysis of variance (ANOVA) was employed to compare differences between groups, followed by least significant difference (LSD) tests for pairwise comparisons. A P value of <0.05 was considered statistically significant.

Results

Model establishment

Out of 50 chick embryos, three died on the day of transplantation due to yolk sac rupture. Two additional embryos succumbed to yolk sac damage the following day, characterized by yellow yolk leakage into the clear albumen. A total of 45 embryos survived until ED7 and were subsequently used for cancer cell implantation. However, four embryos perished on ED9, exhibiting signs of coagulation at the implantation site, indicating excessive bleeding. Five embryos succumbed on ED14, with an additional two expiring on ED16 before the examination (Figure 3). Macroscopic examination revealed intact yolk sacs in these latter cases; however, contraction was observed in their CAMs due to embryonic demise. On ED16, the last round of testing was carried out on the remaining 10 embryos.

Figure 3 Mortality of chicken embryos across embryonic days. The x-axis indicates the embryonic days, whereas the y-axis represents the mortality count. Mortality causes are categorized as yolk sac damage (blue), excessive bleeding (orange), euthanasia (gray), and unidentified reasons (yellow). ED, embryonic day.

mpMRI and pathological findings

At the first examination time point (ED10), MRI demonstrated isointensity on T1-weighted imaging (T1WI), hyperintensity on T2-weighted imaging (T2WI), and mild diffusion restriction changes in the implantation area. Additionally, slight marginal enhancement was observed around the nodule (Figure 4, A1-B2). T1 mapping was unable to observe the difference between the nodule and the surrounding tissue (Figure 4, B3). Gross specimens revealed a nodule located superior to the CAM at the implantation site (Figure 4, C1,C2). Pathological examination confirmed the presence of scattered individual cells within the unresolved Matrigel (Figure 4, D1,D2).

Figure 4 mpMRI (A1-B3) and histopathological features (C1-D2) of the ED10 model (n=8). (A1) T1WI shows a nodule (arrow) with isointense signal. (A2) T2WI reveals hyperintense signal within the nodule. (A3) Contrast-enhanced imaging displays slight marginal enhancement around the nodule (arrowhead). (B1,B2) DWI and the ADC map indicates mild diffusion restriction in the implantation area (arrowheads). (B3) T1 mapping shows no significant difference between the nodule (arrowhead) and the surrounding tissue. (C1,C2) Gross specimens exhibit a translucent, gel-like nodule (circle) located superior to the CAM (scale bars: 1 mm). (D1) H&E staining at low magnification (2.5×) demonstrates the nodule composed of abundant Matrigel (arrow) and tumor cells. (D2) High magnification (6×) H&E staining reveals individually dispersed tumor cells (arrowhead) within the Matrigel (scale bars: 1 mm). ADC, apparent diffusion coefficient; CAM, chorioallantoic membrane; DWI, diffusion-weighted imaging; ED, embryonic day; H&E, hematoxylin and eosin; mpMRI, multi-parametric MRI; MRI, magnetic resonance imaging; T1WI, T1-weighted imaging; T2WI, T2-weighted imaging.

For nodules at the second time point (ED12), T1WI exhibited slightly increased signal intensity along peripheral regions and consistent isointensity in central areas, whereas on T2WI, the nodule demonstrated a hyperintense core with an isointense peripheral rim. The diffusion-weighted imaging (DWI) revealed significant diffusion restriction at the nodule margins, and enhancement became more pronounced (Figure 5, A1-B2). T1 mapping could observe the color difference between the nodule and the surrounding tissue; however, center details could not be displayed (Figure 5, B3). In the gross specimen, the nodule was observed to be classically rounded and thick, and there were vascular branches around the tumor (Figure 5, C1,C2). H&E staining revealed high tumor cell density at the lesion margins, with abundant unabsorbed Matrigel persisting centrally (Figure 5, D1,D2).

Figure 5 mpMRI (A1-B3) and histopathological features (C1-D2) of the ED12 model (n=8). (A1) T1WI shows slightly increased signal intensity in peripheral regions (in contrast with ED10), with consistent isointensity in central areas (arrow). (A2) T2WI reveals a hyperintense core with an isointense peripheral rim. (A3) Contrast-enhanced imaging displays pronounced marginal enhancement (arrowhead). (B1,B2) DWI and ADC map demonstrates significant diffusion restriction at the nodule margins (arrowheads). (B3) T1 mapping clearly delineates the nodule (arrowhead) but fails to resolve detailed internal structures. (C1,C2) Gross specimens exhibit a translucent, elliptical nodule (circle) located both superior and inferior to the CAM, with clearly visible branching vasculature (arrowhead) (scale bars: 1 mm). (D1) H&E staining at low magnification (2.5×) shows abundant Matrigel in the center (arrow) and tumor cells at the periphery. (D2) High-magnification (6×) H&E staining reveals enlarged, irregular tumor cells (arrowhead) densely packed together (scale bars: 1 mm). ADC, apparent diffusion coefficient; CAM, chorioallantoic membrane; DWI, diffusion-weighted imaging; ED, embryonic day; H&E, hematoxylin and eosin; MRI, magnetic resonance imaging; mpMRI, multi-parametric MRI; T1WI, T1-weighted imaging; T2WI, T2-weighted imaging.

In the ED14 group, both T1WI and T2WI demonstrated nodule enlargement accompanied by a progressive signal shift, with central regions gradually replaced by peripheral signals. T1 mapping still failed to show any differences in the center of the nodule. Additionally, the area of restricted diffusion exhibited a significant increase compared to ED12 (Figure 6, A1-B3). The gross specimen observations were comparable to those recorded for ED12 (Figure 6, C1,C2). Histopathological examination revealed near-complete absorption of the Matrigel, with only residual traces remaining at the central region of the nodule (Figure 6, D1,D2).

Figure 6 mpMRI (A1-B3) and histopathological features (C1-D2) of the ED14 model (n=8). (A1) T1WI shows a larger nodule (arrow) compared with ED12 with high signal intensity peripherally and a smaller isointense central region. (A2) T2WI reveals a smaller hyperintense core surrounded by a broader isointense rim. (A3) Contrast-enhanced imaging displays pronounced marginal enhancement (arrowhead) with a non-enhanced central area. (B1,B2) DWI and ADC map demonstrate significant diffusion restriction throughout almost the entire nodule (arrowheads). (B3) T1 mapping clearly delineates the nodule (arrowhead) but fails to resolve detailed internal structures. (C1,C2) Gross specimens exhibit a translucent, elliptical nodule (circle) located both superior and inferior to the CAM, with clearly visible branching vasculature (arrowhead) (scale bars: 1 mm). (D1) H&E staining at low magnification (2.5×) shows near-complete absorption of Matrigel, with minimal residual Matrigel in the center (arrow). (D2) High-magnification (6×) H&E staining reveals the unabsorbed Matrigel (arrowhead) (scale bars: 1 mm). ADC, apparent diffusion coefficient; CAM, chorioallantoic membrane; DWI, diffusion-weighted imaging; ED, embryonic day; H&E, hematoxylin and eosin; MRI, magnetic resonance imaging; mpMRI, multi-parametric MRI; T1WI, T1-weighted imaging; T2WI, T2-weighted imaging.

However, on ED16, the nodule demonstrated slightly higher signal intensity on T1WI and isointense on T2WI images, disclosing uniform enhancement with areas of restricted diffusion; the T1 mapping nodule was similar in color to the previous one (Figure 7, A1-B3). Gross pathology observed an increase in tumor size from the previous one (Figure 7, C1,C2). Pathological analysis revealed a diffuse distribution of tumor cells with enlarged nuclei throughout the nodule (Figure 7, D1,D2).

Figure 7 mpMRI (A1-B3) and histopathological features (C1-D2) of the ED16 model (n=10). (A1) T1WI shows a nodule (arrow) with homogeneous slightly higher signal intensity. (A2) T2WI reveals uniform isointensity. (A3) Contrast-enhanced imaging displays marked enhancement throughout the lesion, including the central region (arrowhead). (B1,B2) DWI and ADC map demonstrate significant diffusion restriction across the entire nodule (arrowheads). (B3) T1 mapping shows signal characteristics (circle) similar to previous models with a persistent lack of detailed resolution. (C1,C2) Gross specimens exhibit a larger translucent, elliptical nodule (circle) located both superior and inferior to the CAM, with clearly visible branching vasculature (arrowhead) (scale bars: 1 mm). (D1) H&E staining at low magnification (2.5×) shows complete absorption of Matrigel, with tumor cells diffusely distributed throughout the lesion. (D2) High-magnification (6×) H&E staining confirms predominant tumor cell involvement in the central region of the lesion (scale bars: 1 mm). ADC, apparent diffusion coefficient; CAM, chorioallantoic membrane; DWI, diffusion-weighted imaging; ED, embryonic day; H&E, hematoxylin and eosin; MRI, magnetic resonance imaging; mpMRI, multi-parametric MRI; T1WI, T1-weighted imaging; T2WI, T2-weighted imaging.

Quantitative findings

Due to the small nodule volume and pathological findings on ED10, no statistical analysis was performed at this time point. For the remaining examination points—ED12 (n=8), ED14 (n=8), and ED16 (n=10)—the measured nodule volumes were 60.40±12.18, 87.44±17.45, and 108.13±5.45 mm³, respectively. ANOVA revealed statistically significant differences among all groups (F=33.87, P<0.001) (Table 2, lower section shows group comparison results).

The ADC values, derived from ADC maps, were (1.47±0.48)×10⁻³ mm²/s, (1.32±0.13)×10⁻³ mm²/s, and (0.96±0.07)×10⁻³ mm²/s for ED12, ED14, and ED16, respectively, with significant differences among the groups (F=8.21, P=0.002), whereas no significant difference was observed between ED12 and ED14 (P=0.29) (Table 2, lower section shows group comparison results).

The measured T1 values for each group were (1.64±0.31)×10³, (1.73±0.16)×10³, and (1.74±0.29)×10³ ms for ED12, ED14, and ED16, respectively, with no statistically significant differences among the groups (F=0.37, P=0.69).

Discussion

Ex ovo culture of chicken embryos presents a superior experimental platform, offering unparalleled visualization through complete CAM exposure and an unobstructed surgical field. This configuration significantly enhances surgical precision and facilitates comprehensive real-time observation throughout experimental procedures. A critical advantage of this methodology is its ability to precisely identify optimal tumor implantation sites, particularly distal to major vasculature, thus minimizing hemorrhage risks during graft placement. Furthermore, selecting implantation sites away from the embryo proper (notably limb and wing regions) eliminates motion artifacts during imaging. Our experimental results confirmed this technical refinement: only 8.89% (4/45) embryos were lost due to accidental large-vessel injuries during transplantation, with no observed motion artifacts in imaging.

Our preliminary experiments in this study indicated that the primary cause of mortality in ex ovo incubation of embryos was yolk sac rupture. To mitigate this risk, meticulous handling is imperative, and the incubation device must ensure that the early yolk sac maintains its spherical shape, thereby diminishing surface tension. Building upon the chamber proposed by Kim et al. (17), which is not readily available in our region, the chamber made by Dunn (18) poses replication challenges in our laboratory, particularly in achieving consistent and uniform conditions. We have designed and developed an innovative ex ovo culture chamber that combines the features of Kim et al.’s and Dunn’s. Compared to existing literature, our chamber provides sufficient space for the yolk sac, allowing it to maintain a near-spherical shape, which theoretically reduces the likelihood of rupture. In this experiment, only 10% of embryos (5/50) exhibited yolk sac rupture, validating the efficacy of our design. Additionally, it offers ample growth space for the CAM. Finally, our design makes it easier for others to replicate the chamber precisely (17-19) (Appendix 1).

MRI is one of the most suitable methods for evaluating CAM tumor models due to its high-resolution imaging of soft tissues, enabling precise monitoring of tumor morphological changes. Unlike computed tomography (CT), MRI involves no radiation exposure, and unlike ultrasound, which requires direct contact with the CAM, thereby increasing the risk of mechanical damage or contamination, MRI ensures safer longitudinal imaging. Moreover, functional MRI techniques can provide additional imaging-based biological information. Unlike other animal models, such as nude mice (20), imaging of CAM tumor models can be performed without additional anesthesia and immobilization. The distinction arises because tumor cells are implanted into the CAM rather than the embryo proper. During imaging sessions (typically lasting dozens of minutes), motion artifacts arising from CAM development—such as vascular network expansion and membrane growth—can be safely disregarded. Even in late-stage ex ovo incubation (typically beyond day 16), when the CAM gradually contracts and adheres to the embryo surface, motion interference can be minimized by placing the tumor graft away from highly mobile regions (e.g., the wings or legs). Additionally, the MRI scanning environment’s relatively low temperature (20–24 ℃) could further suppress spontaneous embryo movement (21). Based on the above theory, using a conventional 3.0 T MRI system, we not only successfully achieved morphological observation of lesions in the CAM tumor model at different growth stages, but also explored quantitative differences in lesions with functional imaging techniques.

On ED10, a small nodule was observed above the CAM. Given its small lesion volume and combined with gross histological and H&E staining findings, we speculated that the MRI signal at this stage primarily reflects the Matrigel component rather than the tumor cells. Therefore, the nodule at this time point was not further compared to those later. Notably, contrast-enhanced imaging revealed mild peripheral enhancement of the nodule, reflecting the onset of neovascularization at the lesion’s edge. This early vascularization implies that drug testing in CAM-based tumor models may be feasible before complete tumor mass development.

On ED12–14, differences in MRI signals emerged predominantly at the periphery of the nodules, characterized by bidirectional expansion of the signal range (both inward and outward). Histopathological examination revealed abundant large, pleomorphic tumor cells dispersed along the lesion margins, whereas the central region contained undigested Matrigel. Over time, this central Matrigel component gradually diminished as tumor cells replaced it. We hypothesized that on T1WI and T2WI, the peripheral signals primarily reflect tumor region, whereas the central signals represent residual Matrigel. Corresponding DWI shows restricted diffusion in tumor cell-dense peripheral areas. These observations suggest that the CAM tumor model exhibits a unique growth pattern: no substantial tumor mass forms at early stages (ED10, 3 days post-implantation). As growth continues, tumor cells progressively accumulate at the lesion periphery and grow bidirectionally (outward and inward). However, ADC measurements showed no statistically significant differences in nodule parameters between the ED12 and ED14 groups. This phenomenon may be attributed to persistent unrestricted diffusion areas in the tumor center during this period, which could mask localized changes in cellular density and confound overall ADC values.

By ED16, the nodules exhibited near-complete diffusion restriction, corroborated histopathologically by dense tumor cell infiltration throughout the lesion. Notably, the ADC values at this stage were significantly lower than those observed in the ED12 and ED14 groups, reflecting the transition from a Matrigel center to a uniformly cellularized tumor mass. These findings highlight the utility of conventional MRI not only for tumor detection and volumetric assessment but also for quantitative differentiation of nodular heterogeneity across developmental stages. As demonstrated in this study, imaging investigations are most appropriately conducted on ED16 when Matrigel absorption is complete; thus, its confounding effects on signal interpretation are minimized.

MRI T1 mapping is a quantitative imaging technique that measures tissues’ longitudinal relaxation time, providing valuable insights into their microenvironment and pathology. T1 mapping holds promising applications in tumor assessment, offering improved diagnostic accuracy and treatment monitoring. The variations in T1 values can facilitate the distinction of tumor malignancy and serve as a metric for assessing treatment efficacy (22,23). However, no statistically significant differences in T1 values were observed across different time points in this study, which may be attributed to the focus on a single tumor cell line at various time points, without comparisons across different cell lines or inclusion of contrast-enhanced-T1 mapping-based evaluations.

This study has several limitations that should be acknowledged. First, continuous dynamic scanning data at varying time points were not acquired for individual samples, which may limit the temporal resolution of our observations. Second, comparisons incorporating multiple tumor cell lines in the CAM tumor model, particularly in relation to functional MRI quantitative parameters, were not conducted, potentially restricting the broader applicability of our findings. Additionally, the experimental protocols described and the observed imaging manifestations of nodule growth may not be generalizable to other tumor types, as the study focused on a specific subset of conditions.

Conclusions

The present study has demonstrated the merits of utilizing the CAM tumor model, using pancreatic cancer as an example, highlighting its ease of establishment and high tumor take rate as significant features. Furthermore, preliminary insights have been provided into the feasibility of employing MRI imaging, particularly for multi-parameter quantitative analysis of tumors.

Acknowledgments

We express our sincere gratitude to our colleague for generously providing laboratory space, which greatly facilitated our research endeavors. Special acknowledgments are owed to Jing Xia, the wife of Guodong Feng, for her meticulous proofreading and editing of the manuscript, resulting in a significant enhancement of its clarity and linguistic quality. Furthermore, we extend our appreciation to Jiajun Wu for his invaluable assistance with tumor cell culture and Xiaoli Xie for her expertise in histological analysis.

Footnote

Reporting Checklist: The authors have completed the ARRIVE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2024-2612/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2024-2612/dss

Funding: The research was financially supported by the Nature Science Foundation of Shanghai (grant No. 19ZR1446200 to B.S.), the Shanghai Municipal Health Commission (grant No. 202140325 to B.S.), the Self-Funded Science and Technology Innovation Project of Foshan (grant No. 2220001005664 to Y.J.L.), and the Minhang Hospital New Seedling Program (grant No. 2024MHXM03 to G.F.).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2024-2612/coif). G.F. receives funding from the Minhang Hospital New Seedling Program (grant No. 2024MHXM03). B.S. receives funding from Nature Science Foundation of Shanghai (grant No. 19ZR1446200) and Shanghai Municipal Health Commission (grant No. 202140325). Y.J.L. receives funding from the Self-Funded Science and Technology Innovation Project of Foshan (grant No. 2220001005664). The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the People’s Republic of China National Standard (GB/T 35892-2018), which categorizes chick embryos as non-animal, thereby exempting ethical approval.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Hubrecht RC, Carter E. The 3Rs and Humane Experimental Technique: Implementing Change. Animals (Basel) 2019;9:754. [Crossref] [PubMed]
2. Ponte G, Roumbedakis K, Galligioni V, Dickel L, Bellanger C, et al. General and species-specific recommendations for minimal requirements for the use of cephalopods in scientific research. Lab Anim 2023;57:26-39. [Crossref] [PubMed]
3. Smith JA, Andrews PLR, Hawkins P, Louhimies S, Ponte G, Dickel L. Cephalopod research and EU Directive 2010/63/EU: Requirements, impacts and ethical review. Journal of Experimental Marine Biology and Ecology 2013;447:31-45.
4. Garcia P, Wang Y, Viallet J, Macek Jilkova Z. The Chicken Embryo Model: A Novel and Relevant Model for Immune-Based Studies. Front Immunol 2021;12:791081. [Crossref] [PubMed]
5. Ko KH, Lee IK, Kim G, Gu MJ, Kim HY, Park BC, Park TS, Han SH, Yun CH. Changes in bursal B cells in chicken during embryonic development and early life after hatching. Sci Rep 2018;8:16905. [Crossref] [PubMed]
6. Rous P. A sarcoma of the fowl transmissible by an agent separable from the tumor cells. J Exp Med 1911;13:397-411. [Crossref] [PubMed]
7. Murphy J. Transplantability of malignant tumors to the embryos of a foreign species. JAMA 1912;LIX:874-5.
8. Smith LM, Greenwood HE, Tyrrell WE, Edwards RS, de Santis V, Baark F, Firth G, Tanc M, Terry SYA, Herrmann A, Southworth R, Witney TH. The chicken chorioallantoic membrane as a low-cost, high-throughput model for cancer imaging. Npj Imaging 2023;1:1. [Crossref] [PubMed]
9. Charbonneau M, Harper K, Brochu-Gaudreau K, Perreault A, Roy LO, Lucien F, Tian S, Fortin D, Dubois CM. The development of a rapid patient-derived xenograft model to predict chemotherapeutic drug sensitivity/resistance in malignant glial tumors. Neuro Oncol 2023;25:1605-16. [Crossref] [PubMed]
10. Chen L, Wang M, Feng Y, Gao L, Yu J, Geng L, Xie Y, Coudyzer W, Li Y, Ni Y. Multi-parametric investigations on the effects of vascular disrupting agents based on a platform of chorioallantoic membrane of chick embryos. Quant Imaging Med Surg 2024;14:1729-46. [Crossref] [PubMed]
11. Lewis MT, Caldas C. The Power and Promise of Patient-Derived Xenografts of Human Breast Cancer. Cold Spring Harb Perspect Med 2024;14:a041329. [Crossref] [PubMed]
12. Ndreshkjana B, Çapci A, Klein V, Chanvorachote P, Muenzner JK, Huebner K, Steinmann S, Erlenbach-Wuensch K, Geppert CI, Agaimy A, Ballout F, El-Baba C, Gali-Muhtasib H, Roehe AV, Hartmann A, Tsogoeva SB, Schneider-Stock R. Combination of 5-fluorouracil and thymoquinone targets stem cell gene signature in colorectal cancer cells. Cell Death Dis 2019;10:379. [Crossref] [PubMed]
13. Swadi R, Mather G, Pizer BL, Losty PD, See V, Moss D. Optimising the chick chorioallantoic membrane xenograft model of neuroblastoma for drug delivery. BMC Cancer 2018;18:28. [Crossref] [PubMed]
14. Herrmann A, Taylor A, Murray P, Poptani H, Sée V. Magnetic Resonance Imaging for Characterization of a Chick Embryo Model of Cancer Cell Metastases. Mol Imaging 2018;17:1536012118809585. [Crossref] [PubMed]
15. Wang L, Yan J. XinyuWang, Xu Y, Pan D, Chen C, Shao Y, Song X, Qi K, Yang M, Tu J. Evaluation of chicken chorioallantoic membrane model for tumor imaging and drug development: Promising findings. Animal Model Exp Med 2025;8:287-94. [Crossref] [PubMed]
16. Winter G, Koch ABF, Löffler J, Lindén M, Solbach C, Abaei A, Li H, Glatting G, Beer AJ, Rasche V. Multi-Modal PET and MR Imaging in the Hen’s Egg Test-Chorioallantoic Membrane (HET-CAM) Model for Initial in Vivo Testing of Target-Specific Radioligands. Cancers (Basel) 2020;12:1248. [Crossref] [PubMed]
17. Kim Y, Williams KC, Gavin CT, Jardine E, Chambers AF, Leong HS. Quantification of cancer cell extravasation in vivo. Nat Protoc 2016;11:937-48. [Crossref] [PubMed]
18. Dunn BE, Boone MA. Growth of the chick embryo in vitro. Poult Sci 1976;55:1067-71. [Crossref] [PubMed]
19. Huang W, Itayama M, Arai F, Furukawa KS, Ushida T, Kawahara T. An angiogenesis platform using a cubic artificial eggshell with patterned blood vessels on chicken chorioallantoic membrane. PLoS One 2017;12:e0175595. [Crossref] [PubMed]
20. Chaker SC, Manzanera Esteve IV, Yan L, Hung YC, James AJ, Saad M, Thayer WP. In-Vivo MRI in Rodents: A Protocol for Optimal Animal Positioning. Ann Plast Surg 2024;93:S116-8. [Crossref] [PubMed]
21. Noiva RM, Menezes AC, Peleteiro MC. Influence of temperature and humidity manipulation on chicken embryonic development. BMC Vet Res 2014;10:234. [Crossref] [PubMed]
22. Lim JZ, Gokul A, Misra SL, Pan X, Charlton A, McGhee CNJ. An optimized 3T MRI scan protocol to assess iris melanoma with subsequent histopathological verification - A prospective study. Asia Pac J Ophthalmol (Phila) 2024;13:100047. [Crossref] [PubMed]
23. Payne MM, Mali I, Shrestha TB, Basel MT, Timmerman S, Pyle M, Sebek J, Prakash P, Bossmann SH. T(1)-mapping characterization of two tumor types. Biophys Rep (N Y) 2024;4:100157. [Crossref] [PubMed]