A method for scoring 4-nitroquinoline 1-oxide-induced murine esophageal squamous neoplasia

Abstract

A mouse model for esophageal squamous cell carcinoma (ESCC) is induced by oral administration of the carcinogen 4-nitroquinoline 1-oxide (4-NQO). There is not an objective method for determining histopathologic severity of disease in this model. We aim to create a clearly defined and easily applied scoring system that can quantify the severity of 4-NQO-induced murine ESCC. Fifteen wild-type C57BL/6J mice were treated with 4-NQO for 8 (n = 8) or 16 (n = 7) weeks, while the rest (n = 9) were treated with vehicle, as 8 weeks of 4-NQO typically results in dysplasia and 16 weeks in carcinoma. We identified histologic abnormalities of the esophagus in this model and developed metrics to grade severity of dysplasia, papillomas, and invasion. Scores were then calculated using quantitative digitized image analysis for measuring depth and extent of each feature within the entire sample. Each feature was also assigned a weight based on its relation to cancer severity. Histology scores were significantly different in the three groups, suggesting that this method can discriminate dysplasia from carcinoma. This model can be applied to any mouse treated with 4-NQO.

Keywords

digestive tract disease severity metrics esophageal squamous cell cancer mouse model

Esophageal carcinomas are the sixth most common cause of cancer-related deaths worldwide and are the eighth-most common cancer type worldwide.¹⁵ The histologic subtype esophageal squamous cell carcinoma (ESCC) accounts for most esophageal cancers.^2,20,22,29 In 2020, there were 604,100 new cases of esophageal cancer in humans in 185 countries, and of these 512,500 cases (85%) were ESCC.¹⁵ Moreover, ESCC has a high mortality rate and a 5-year survival rate of less than 20%.^2,21,29

Esophageal squamous cell carcinoma is the malignant transformation of normal esophageal stratified squamous epithelium. The cause is multifactorial, likely due to a combination of genetic and environmental risk factors, including tobacco and alcohol abuse.^{8,16,17,20,21,26} In the western hemisphere, it is typically not diagnosed until it is at an advanced stage, when it can be difficult to treat.^20,21 Thus, increased understanding of ESCC pathogenesis is needed to improve management of the disease.

One mouse model employed to study the development of ESCC is induced with 4-nitroquinoline 1-oxide (4-NQO),^24,29 since it mimics the development of ESCC in humans.²⁹ 4-Nitroquinoline 1-oxide contains similar carcinogens to those present in tobacco and causes DNA damage.^19,27,28 Although several publications have described how 4-NQO results in the development of ESCC using various timelines, there is currently not an objective method for assessing the severity of disease in mice.

Most groups employing 4-NQO use subjective scoring systems to broadly characterize abnormalities in esophageal epithelial architecture. These categories are not always specifically defined and can vary between studies.^{4
–7,10
–14,18,23
–25,30,31} For example, one study classified 4-NQO-induced esophageal cancer into four categories: normal, hyperplastic, dysplastic (low grade or high grade), and carcinoma,¹⁸ while another study grouped mice in six categories: normal, inflammatory, hyperplasia, dysplasia, carcinoma in situ, and invasive carcinoma.³⁰ Most studies have shown that esophageal squamous neoplasia induced by 4-NQO progresses from hyperplasia to dysplasia to carcinoma over time.^4,30

Hyperplasia is typically defined as a thickened epithelium with prominent surface keratinization with or without elongated rete pegs.^10,14,24 Dysplasia is characterized by a loss of polarity in epithelial cells, nuclear pleomorphism, nuclear hyperchromasia, abnormal single cell keratinization (dyskeratosis), and increased or abnormal mitotic figures.^4,10,14,24 Papilloma is defined as a mature acanthotic squamous epithelium arranged along branched fibrovascular cores with an exophytic architectural pattern.^3,11,14,24 Carcinoma in situ is used to describe lesions with changes involving the entire thickness of epithelium, and invasive carcinoma is a lesion with invasion into the subepithelial tissues.^10,14,24 Although these general descriptions of papilloma, hyperplasia, dysplasia, and carcinoma are informative, they do not allow for objective comparisons between mice that may have varying degrees of severity within a category. While some reports further classify lesions within these categories as mild, moderate, or severe, they often do not explain which features define those categories.^4,11,23,29 Currently, there is no quantitative scoring system available for the 4-NQO model. Thus, the aim of this manuscript is to create a more defined and granular scoring system to assess disease severity in the 4-NQO mouse model of ESCC. We have identified common features in mice treated with 4-NQO associated with ESCC neoplasia and developed a semiquantitative scoring system based on the measurements of these features, their extent, and weighted severity.

Materials & Methods

Mice

Eight- to 12-week-old C57BL/6J mice were used for this study. These mice were maintained in controlled conditions with a consistent diet and kept in normal light/dark cycle, temperature, and humidity. All experiments were approved by Institutional Animal Care and Use Committee (IACUC) at Vanderbilt University Medical Center (Protocol V1900036-01) and Tennessee Valley Health System IACUC (Protocol 1405914).

4-NQO Treatment

4-Nitroquinoline 1-oxide (Acros Organics Catalog # 203792500) was dissolved in propylene glycol (Fisher Scientific Catalog # P355-4) vehicle as a stock solution at a dosage of 50 mg/mL and stored at 4°C. A new stock solution was made weekly. The 4-NQO stock solution was then diluted in drinking water to a dosage of 100 µg/mL consistent with other 4-NQO-induced ESCC reports and was changed weekly.^{5,4,10
–12,18,23
–25,30} Mice were continuously allowed access to the drinking water over the duration of the experiment.

Mice were divided into 3 groups: (1) 8 weeks 4-NQO, (2) 16 weeks 4-NQO, and (3) control. These time points were chosen because previous studies have shown that treating mice with 4-NQO for 8 weeks followed by 8 weeks vehicle results in dysplasia,^5,4,14,29,30 while treating mice with 4-NQO for 16 weeks followed by 12 weeks vehicle results in carcinoma (Fig. 1).^4,10,13,24 The control group was given propylene glycol (vehicle) in the drinking water over the course of 28 weeks. All three experimental groups were coordinated to begin and end at the same time. The 8-week 4-NQO treatment group was given only vehicle for the first 12 weeks of the experiment, then received 4-NQO treatment from weeks 12 to 20, and then given only vehicle for the last 8 weeks of the experiment. The 16 week 4-NQO treatment group was given 4-NQO from weeks 1 to 16 and then given only vehicle for the last 12 weeks of the experiment. All mice were sacrificed at 28 weeks from the beginning of the experiment.

Figure 1.

4-nitroquinilone 1-oxide (4-NQO) treatment timelines.

In our experiment, we used a total of 24 mice (n = 9 [5 male, 4 female] control, n = 8, 8 week 4-NQO [4 male, 4 female], and n = 7 [4 male, 3 female] 16 weeks 4-NQO). Sacrifice was done by placing mice in a sealed container, where isoflurane was inhaled until respiration ceased and death occurred. Immediately after death, the esophagi of all mice were dissected and then Swiss-rolled. To do this, the inside of the esophagus was washed with phosphate-buffered saline (PBS) using a gavage needle, opened longitudinally, and then rolled with the distal esophagus in the center and proximal esophagus on the outside of the roll. Then, this Swiss-roll was fixed in 10% buffered formalin for 24 hours, transferred to 70% ethanol, routinely processed, and embedded in paraffin. Precision cut 5-µm sections of each sample containing a cross-section of the entire esophagus was cut and stained with hematoxylin and eosin (H&E) for histopathological analysis. Paraffin-embedding, sectioning, and staining was done by Vanderbilt Tissue Pathology Shared Resource (TPSR).

Slide Upload and QuPath Software

QuPath version 0.3.2, a free open-source software,¹ was used to analyze whole-slide images of esophageal histology. QuPath allows for the analysis of high-resolution, whole-slide images and has a compatible ImageJ plugin that was used to capture images in .tiff format. Slides were digitally scanned by the Vanderbilt Digital Histology Shared Resource (DHSR) core. The DHSR uses an Aperio AT2 and Leica SCN400 Slide Scanner for high-resolution bright field scanning. After scanning, the images are uploaded digitally into a web-based digital slide-viewing environment (Digital Slide Archive, DSA) for rapid retrieval. From there, the digital slide was downloaded and the image file (.scn format) was uploaded into QuPath, which we used to identify and measure each category in our scoring system. Specifically, the line annotation tool was used to make straight line measurements, while the polyline tool was used to make line measurements that were curved. Accurate measurements were generated based on scaling data provided from the .scn file of the image downloaded from the DSA.

Mice

We used a total of 24 mice (n = 9 [5 male, 4 female] control, n = 8, 8 week 4-NQO [4 male, 4 female], and n = 7 [4 male, 3 female] 16 weeks 4-NQO). One mouse from the 16-week 4-NQO treatment died (1/7, 14%) during the 28-week experiment at the 27th week of the experiment and was excluded from the study because its histology could not be compared to other mice in the experiment at the same time point. This rate is comparable to other studies.^10,24 All other mice were sacrificed on the same day, at the 28-week timepoint of the experiment. Supplemental Fig. S1 shows the weight curves of the mice throughout the experiment (a) and number of esophageal tumors seen grossly (b).

Histological Analysis

While reviewing the slides, we (MKW, an expert GI pathologist, and YC) first identified 3 categories for defining preneoplastic and neoplastic lesions in the esophagus: dysplasia, papilloma, and invasion. Dysplasia was further characterized by rete pegs and nuclear histologic abnormality (NHA). Each category is weighted to reflect the severity of the feature in relation to cancer development with invasion carrying the greatest weight and dysplasia having the least weight. This is because dysplasia is less severe and develops prior to carcinoma.⁹ Supplemental Table S1 shows definitions of each histological category and a summary of how the score is calculated.

Dysplasia

Nuclear histologic abnormality

Nuclear histologic abnormality was indicated by the presence of abnormal cells in the epithelium (Fig. 2). A score of 0 indicates no abnormality is present (Fig. 2a). A score of 25 indicates mild nuclear changes characterized by increased nuclear to cytoplasmic ratio, nuclear hyperchromasia, and nuclear pleomorphism (Fig. 2b). Less than 50% of the epithelium demonstrates these abnormal cells. A score of 50 (Fig. 2c) indicates severe dysplasia with more pronounced and extensive nuclear changes. Abnormal cells occupy greater than 50% of the epithelial thickness and show loss of polarity. To determine extent of NHA, all areas with the highest score were measured using the polyline tool and then added together. Next, the total esophageal length was measured, using the polyline tool to trace along the entire esophagus in the roll (Fig. 3a). The total length of the esophagus with NHA was divided by the total esophagus length to determine the proportion in which this NHA score was present. Finally, the proportion of the affected esophagus was multiplied by the score and by 5 so that its contribution to the total score was not significantly less than the other features. For example, suppose the most severe part of the esophagus has a score of 50, and the total measurement where there is NHA with a score of 50 is 300 µm. If the total esophagus length is 15,000 µm, the formula for determining NHA is 50*(3000/15,000)*5 = 50.

Figure 2.

Assessment of nuclear histologic abnormalities in 4-nitroquinilone 1-oxide treated mouse esophagus. Hematoxylin and eosin. (a) Representative image of nuclear histological abnormality with a score of 0. Basal cells have no histologic abnormalities or deformations in the nuclei. (b) Representative image of nuclear histological abnormality with a score of 25. Basal cell nuclei showing slight variation in size, shape, and spacing. Less than 50% of the epithelium consists of abnormal cells. (c) Representative image of nuclear histological abnormality with a score of 50. There is enlarged nuclear to cytoplasmic ratio in basal cells. Cell shape and nuclei size are highly irregular between cells. Greater than 50% of the epithelium consists of abnormal cells.

Figure 3.

Measurement of murine esophageal histopathologic features: rete pegs, papillomas, and invasion. Hematoxylin and eosin. (a) Representative image for measurement of total esophageal length. Polyline tool is used to trace along the length of the entire esophagus. (b) Representative image for measurement of rete pegs. Vertical yellow lines demonstrate that rete peg depth is measured from the end of the basal layer to the end of the elongated downward projection into the lamina propria. Horizontal yellow lines demonstrate the length of rete peg. (c) Representative image for measurement of papillomas. Vertical yellow line shows that papilloma height is the measurement of the entire epithelial projection into the lumen measured along its longest dimension. Horizontal yellow line is the length of the papilloma. (d) Representative image for measurement of invasive lesions. The vertical yellow line represents the invasion depth, which is the vertical distance from the epithelial layer to the end of the neoplastic epithelium that has extended into the submucosa along its longest dimension. The horizontal yellow line measures the length of the invasive area along the esophagus.

Rete pegs

Rete pegs were epithelial projections that penetrate the connective tissue underneath the epithelium but do not invade beyond the basement membrane (Fig. 3b). To quantify the severity of rete pegs in a sample, the length of the each rete peg throughout the esophagus was measured in micrometers using the line annotation tool in QuPath and then measurements were added together (horizontal yellow lines in Fig. 3b), to determine the total esophageal length affected by rete pegs. Next, the depth of all rete pegs was measured from the basal cell layer downward into the cusp of its underlying epithelial extension into the lamina propria (vertical yellow lines in Fig. 3b). The depth of all rete pegs was averaged, which was then multiplied by the proportion of the esophagus in which pegs were present (length of affected esophagus/total esophageal length). Finally, this number was multiplied by 5. For example, suppose there were 3 rete pegs in a sample which had depth measurements of 30, 40, and 50 µm. The length of each peg was 60 µm. The entire length of the esophagus is 14,000 µm. Five is the multiplier for rete pegs. Thus, the rete peg score in this case would be 40*(180/14,000)*5 = 2.57. This method takes into account the entire esophagus and does not arbitrarily select 5 rete pegs as representative.

Papilloma score

A papilloma was defined as a mature acanthotic squamous epithelium arranged along branched fibrovascular cores with an exophytic architectural pattern (Fig. 3c).³ This had the second most weight in the total score, as it is more severe than dysplasia but less severe than invasion. To quantify the papilloma score, both the height of the papilloma protruding into the lumen (vertical line in Fig. 3c) and the length of the affected esophagus (horizontal line in Fig. 3c) were measured. All heights of papillomas were measured and averaged, while all papilloma lengths were summed and divided by the total length of the esophagus. Then, the average height was multiplied by the sum length of the papillomas divided by the total esophageal length. Finally, this was multiplied by 10. For example, suppose there are two papillomas with heights of 300 µm and lengths of 150 and 400 µm, and the esophagus length is 15,000 µm. In this case, the papilloma score would be 300*(550/15,000)*10 = 110.

Invasion score

Invasion was defined as neoplastic epithelium that has penetrated the basement membrane and extended into the submucosa (Fig. 3d). It was measured by the vertical distance from the basal region of the epithelial layer to the end of the neoplastic epithelium that extended into the muscularis mucosa along its greatest depth. Invasion had the greatest weight in the total score because it portends the greatest severity of disease. This was scored by measuring both the length (horizontal line in Fig. 3d) and depth of invasion (vertical line in Fig. 3d) in all areas where it is present in micrometers using the line annotation tool in QuPath (yellow line in Fig. 3d). Then, the average depth of invasion was multiplied by extent of invasion (i.e. sum of the lengths where there is invasion divided by total esophageal length). Finally, this score was multiplied by 20 so that it had greater weight than the other features. For example, suppose there were 2 areas of invasion, with depths of 150 and 100 µm. If each had a lengths along the esophagus of 400 and 500 µm, and the total esophageal length was 16,000 µm, then the score would be 125*(900/16,000)*20 = 140.6.

Statistical analysis

Comparison of tumor number in the 3 groups was done using Kruskal-Wallis test. Comparison of control group histology score with 8 week 4-NQO and 16 week 4-NQO was also done using Kruskal-Wallis test. Comparison of 8 week 4-NQO and 16 week 4-NQO histology scores was done using Mann-Whitney test. ROUT (Q = 1%) was done on all 3 groups to identify outliers, and outliers were excluded.

Results

Comparison of Three Groups

All scores for each mouse are reported in Table 1. The control group had significantly lower total scores as compared with 8 week 4-NQO (3.64 ± 3.36 vs. 76.55 ± 38.6, P < .05) and 16 week 4-NQO (3.64 ± 3.36 vs. 660.9 ± 615.1, P < .001) groups. The 8 week 4-NQO group had significantly lower scores than the 16 week 4-NQO group (76.55 ± 38.6 vs. 660.9 ± 615.1, P < .05). One mouse in the 8 week 4-NQO group was excluded as an outlier by the Robust Regression and Outlier removal (ROUT) test.

Table 1.

Total scores for all mice.

Condition	DysplasiaNuclear Histologic Abnormality	DysplasiaRete Pegs	Papilloma	Invasive Lesion	Total Score
Control	0	0	0	0	0
Control	0	1.69	0	0	1.69
Control	0	1.82	0	0	1.82
Control	0	8.64	0	0	8.64
Control	0	3.21	0	0	3.21
Control	0	7.46	0	0	7.46
Control	0	2.40	0	0	2.40
Control	0	7.56	0	0	7.56
Control	0	0	0	0	0
Control total score (mean ± SD)					3.64 ± 3.36
8 week 4-NQO	7.23	22.62	23.84	41.11	94.81
8 week 4-NQO	30.63	21.52	0	0	52.15
8 week 4-NQO^a	62.36	96.03	222.71	0	381.10
8 week 4-NQO	10.65	17.57	10.43	0	38.66
8 week 4-NQO	29.81	31.32	36.02	55.62	152.79
8 week 4-NQO	31.25	17.61	0	0	48.86
8 week 4-NQO	21.53	52.88	0	0	74.41
8 week 4-NQO	22.24	10.32	0	41.63	74.19
8 week 4-NQO total score (mean ± SD)					76.55 ± 38.6^b
16 week 4-NQO	28.07	39.13	0	0	67.21
16 week 4-NQO	56.47	34.80	336.07	642.95	1164.04
16 week 4-NQO	70.46	35.29	0	101.86	207.61
16 week 4-NQO	107.40	33.83	99.89	427.34	668.44
16 week 4-NQO	66.25	68.11	0	1479.19	1613.55
16 week 4-NQO	20.96	31.65	54.15	138.08	244.84
16 week 4-NQO total score (mean ± SD)					660.9 ± 615.1^c,d

Abbreviation: 4-NQO, 4-nitroquinoline 1-oxide.

This mouse was an outlier based on ROUT (Q = 1%) test.

8 week 4-NQO vs Control, P < .05.

16 week 4-NQO vs Control, P < .001.

16 week 4-NQO vs 8 week 4-NQO, P < .05.

Histological Scoring Examples

In this section, we will give one example of the scoring rubric from the 8-week dysplasia group and one from the 16-week cancer group.

Example 1

The first example of the scoring system application is from the 8-week 4-NQO treatment group (Fig. 4). The total esophageal length was measured using the polyline tool, and it was 13,577.3 µm. For NHA, the greatest severity in the sample had a score of 25, due to mild nuclear changes including abnormal shape and size but without severe changes to nuclear characteristics to qualify for a score of 50 (Fig. 4b). The total length of all areas with NHA score 25 was 1157.3 µm. The total score for NHA was therefore 25*(1157.3/13577.3) × 5 = 10.65. The length and depth of all rete pegs were measured (one example shown in Fig. 4c). The average depth (56.45 µm), and the total width of all rete pegs was 845.2 µm. Thus, the rete peg score was 56.45*(845.2/13577.3)*5 = 17.57. The total dysplasia score was thus 10.65 + 17.57 = 28.22. The sample had 1 papilloma (Fig. 4d) with a height of 92.94 µm. The length of this papilloma was 152.45 µm. Thus, the papilloma score was 92.94*(152.45/13,577.3)*10 = 10.44. There was no evidence of invasive lesions present so total score for invasion was 0. The total score is 10.65 + 17.57 + 10.44 = 38.66. See Table 2 for a summary of the results.

Table 2.

Example score card for a mouse esophagus in the 8-week dysplasia group.

Total Esophagus Length (µm)	13,577.3
Dysplasia	Score (0–50)	Total Length	Multiplier	Total Score
Nuclear Histologic Abnormality	25	1157.3	×5	10.65
Rete Pegs	Average Depth (µm) 56.45	Total Length 845.2	Multiplier ×5	Total Score 17.57
Papilloma	Average Height (µm)	Total Length	Multiplier	Total Score
Papilloma #1	92.94	152.46
Papilloma #2
Papilloma #3
Papilloma #4
Papilloma #5
Average/Total	92.94	152.46	×10	10.44
Invasion	Average Depth (µm)	Total Length	Multiplier	Total Score
Invasion Field 1:
Invasion Field 2:
Invasion Field 3:
Average	0	0.0	×20	0
			Total Score	38.66

Figure 4.

Scoring example from 8-week dysplasia group. Hematoxylin and eosin. (a) Swiss roll of a mouse esophagus from 8-week dysplasia group. Letters denote where images for b, c, and d were acquired. (b) This image is an example of nuclear histologic abnormality with a score of 25. The yellow line demonstrates the length of the area of nuclear histologic abnormality. Basal cell nuclei show variation in size, shape and spacing. Less than 50% of the epithelium consists of abnormal cells. (c) Representative image of rete peg. The vertical yellow line on the image indicates the depth of this rete peg, and the horizontal yellow line indicates the length. (d) This is an image of the only papilloma in this image. The vertical yellow line shows the height or degree of protrusion into the lumen, while the horizontal yellow line indicates the length of the papilloma along the esophagus.

Example 2

The example is from the 16-week 4-NQO treatment group (Fig. 5). The total esophageal length was 22,642.25 µm. For NHA, the sample had severe nuclear changes occupying greater than 50% of epithelial thickness in some areas as well as a loss of polarity; thus, the score was 50 (Fig. 5b). The total length of NHA with score of 50 was 5114 µm. The total score for NHA was thus 50*(5114/22,642.25)*5 = 56.47. The length and depth of all rete pegs was then measured (Fig. 5c). The average depth was 64.28 µm, and the total length was 2451.3 µm. Thus, the total rete pegs score was 64.28*(2451.3/22,642.25)*5 = 34.79. The total dysplasia score was 64.28 + 34.79 = 99.07. The sample had 2 papillomas (one example in Fig. 5d). The average height of the 2 papillomas was 574.38 µm, and the total length was 1323.76 µm. Thus, the total papilloma score was 574.38*(1326.76/22,642.25)*10 = 336.07. There were three areas of invasion. The average depth was 238.65 µm, and the total length of these invasive areas was 3050.03 µm. Thus, the invasion score was 238.65*(3050.03/22,642.25)*20 = 642.95. The total score is thus 64.28 + 34.79 + 336.07 + 642.95 = 1164.04. See Table 3 for a summary of the results for this sample. For a blank table to fill in scores, see Table 4.

Table 3.

Example score card for a mouse esophagus in the 16-week cancer group.

Total Esophagus Length (µm)	22,642.25
Dysplasia	Score (0–50)	Total Length	Multiplier	Total Score
Nuclear Histologic Abnormality	50	5114	×5	56.47
Rete Pegs	Average Depth (µm) 64.28	Total Length 2451.3	Multiplier ×5	Total Score 34.79
Papilloma	Average Height (µm)	Total Length	Multiplier	Total Score
Papilloma #1	527.71	668.69
Papilloma #2	621.95	655.07
Papilloma #3
Papilloma #4
Papilloma #5
Average/total	574.83	1323.76	×10	336.07
Invasion	Average Depth (µm)	Total Length	Multiplier	Total Score
Invasion Field 1:	276.23	1904.8
Invasion Field 2:	214.64	970.23
Invasion Field 3:	225.08	175
Average	238.65	3050.03	×20	642.95
			Total Score	1164.04

Table 4.

Empty table for recording the histologic score of a mouse esophagus treated with 4-NQO.

Total Esophagus Length (µm)
Dysplasia	Score (0–50)	Total Length	Multiplier	Total Score
Nuclear Histologic Abnormality			×5
Rete Pegs	Average Depth (µm)	Total Length	Multiplier ×5	Total Score
Papilloma	Average Height (µm)	Total Length	Multiplier	Total Score
Papilloma #1
Papilloma #2
Papilloma #3
Papilloma #4
Papilloma #5
Average/Total			×10
Invasion	Average Depth (µm)	Total Length	Multiplier	Total Score
Invasion Field 1:
Invasion Field 2:
Invasion Field 3:
Average			×20
			Total Score

Figure 5.

Scoring example from 16-week cancer group. (a) Swiss roll of a mouse esophagus from 16-week cancer group. Letters denote where images for b, c, and d were acquired. (b) Image is an example of nuclear histologic abnormality with a score of 50. The yellow line demonstrates the length of the area of nuclear histologic abnormality. There is enlarged nuclear to cytoplasmic ratio in basal cells. Cell shape and nuclei size are highly irregular between cells. Greater than 50% of the epithelium consists of abnormal cells. (c) Representative image of rete peg. The vertical yellow line on the image indicates the depth of this rete peg, and the horizontal yellow line indicates the length. (d) Image of one papilloma in this example. The vertical yellow line shows the height or degree of protrusion into the lumen, while the horizontal yellow line indicates the length of the papilloma along the esophagus. (e) This is an image of one field of invasion from this mouse. The vertical yellow line depicts depth measurement, while the horizontal yellow line depicts length of the invasive area.

For the total number of each feature for all mice, see Supplemental Table S2.

Discussion

While 4-NQO is the most common model of murine ESCC, currently there is no objective histologic grading system to quantify the severity of dysplasia or cancer. In this report, we reviewed H&E stains of esophagi from mice that were treated with 4-NQO and identified metrics for assessing disease severity: nuclear histological abnormality, rete pegs, papillomas, and invasion. We then determined how to quantify each metric so that mice can be compared as objectively as possible. We have created a scoring system for the 4-NQO model and have provided specific methods for quantifying these features. We have also demonstrated that our method can differentiate mice treated with 4-NQO for 8 weeks versus 16 weeks (Table 1). A blank table which can be printed and used for quantification is provided (Table 4).

The major strength of our study is that this method is easily applicable to any mouse that has been treated with 4-NQO, as long as one uses software which allows for accurate measurement. Furthermore, our system provides a way to compare mice quantitatively by categorical features and by total score. Our system also weights categories differently based on its relation to cancer severity. Another strength is that we determined that mice treated with 4-NQO for 16 weeks had significantly higher scores than those treated with 4-NQO for 8 weeks, demonstrating that our metrics were applicable to both time points and reflected the increased severity in mice treated with 4-NQO for a longer time.

One limitation of our study is that there still may be variations in interpretation of some histologic features, especially in assigning scores for nuclear histological abnormality. The major reason histologic nuclear abnormalities are not quantitative is that measuring the shape of each nucleus in an esophagus is not practical. Moreover, detecting nuclear hyperchromasia and/or nuclear/cytoplasmic ratio with QuPath is difficult and not always accurate. However, our overall rationale was to be as quantitative as possible and take into account the whole esophagus as best as possible. A second limitation may arise from how a sample is prepared and presented for histological analysis. The entire esophagus is not examined on each cut, so it is possible that there could be sampling error, especially with regard to the number of gross tumors. For example, when sections are cut from paraffin-embedded blocks, sometimes papillomas are missed and other times only a piece of a papilloma is visualized on H&E. Due to this, one should always evaluate the number of tumors grossly and present that data separately, which we have done in Supplementary Fig. S1. Although we did not do it for this manuscript, the height, length, and width of tumors can be measured grossly as well and also be presented separately. Despite these limitations, however, this is the most quantitative and objective method for quantifying severity of disease with the 4-NQO model. It is easily applicable and is useful in comparing mice.

Supplemental Material

sj-pdf-1-vet-10.1177_03009858231151381 – Supplemental material for A method for scoring 4-nitroquinoline 1-oxide-induced murine esophageal squamous neoplasia

Supplemental material, sj-pdf-1-vet-10.1177_03009858231151381 for A method for scoring 4-nitroquinoline 1-oxide-induced murine esophageal squamous neoplasia by Zaryab Aziz, Mary Kay Washington, Justin Jacobse and Yash Choksi in Veterinary Pathology

Footnotes

Author Contributions

Zaryab Aziz: Conduct of data analysis and manuscript.

Mary Kay Washington: Conduct of data analysis and manuscript.

Justin Jacobse: Conduct of data analysis.

Yash Choksi: Study concept, design, and manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Tissue Pathology Shared Sources at VUMC (NCI/NIH Cancer Center Support Grant no. 5P30CA68485-19), VICC SPORE in GI Cancer (grant no. P50CA236733 to MKW) and Molecular and Cellular Basis of Digestive Diseases (grant no. P30DK058404 to MKW). It was also supported by Academy Ter Meulen Grant from the Royal Netherlands Academy of Arts and Sciences to JJ and Department of Veterans Affairs (Career Development Award-2 IK2BX004648 to YAC).

ORCID iD

Yash Choksi

Supplemental Material for this article is available online.

References

Bankhead

Loughrey

Fernandez

, et al QuPath: open source software for digital pathology image analysis. Sci Rep. 2017;7:16878.

Bray

Ferlay

Soerjomataram

, et al Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68:394–424.

Gordon

Goldbum

. Esophagus. In: Mills

, ed. Sternberg’s Diagnostic Surgical Pathology. 6th ed. Philadelphia, PA: Wolters Kluwer; 2015:1375–1409.

Hasina

Martin

Kasza

, et al ABT-510 is an effective chemopreventive agent in the mouse 4-nitroquinoline 1-oxide model of oral carcinogenesis. Cancer Prev Res (Phila). 2009;2(4):385–393.

Hasina

Mollberg

Kawada

, et al Critical role for the receptor tyrosine kinase EPHB4 in esophageal cancers [published correction appears in Cancer Res. 2013;73(14):4594]. Cancerres. 2013;73(1):184–194.

Rivera

Diossy

, et al BRCA1/Trp53 heterozygosity and replication stress drive esophageal cancer development in a mouse model. Proc Natl Acad Sci U S A. 2021;118(41):e2108421118.

Horiuchi

Taguchi

Hirose

, et al Cellular Nrf2 levels determine cell fate during chemical carcinogenesis in the esophageal epithelium. Mol Cell Biol. 2021;41:e00536-20.

Islami

Kamangar

Nasrollahzadeh

, et al Socio-economic status and oesophageal cancer: results from a population-based case-control study in a high-risk area. Int J Epidemiol. 2009;38(4):978–988.

Leininger

Jokinen

Dangler

, et al Oral cavity, esophagus, and stomach. In: Maronpot

, ed. Pathology of the Mouse. 1st ed. Sant Louis, MO: Cache River Press; 1999:38–42.

10.

Chang

AYW

, et al Angiotensin II type I receptor (AT1R) is an independent prognosticator of esophageal squamous cell carcinoma and promotes cells proliferation via mTOR activation. Oncotarget. 2016;7(41):67150–67165.

11.

Dong

Yang

, et al Mouse tumor-bearing models as preclinical study platforms for oral squamous cell carcinoma. Front Oncol. 2020;10:212.

12.

Lin

Dai

, et al Mcl-1 inhibitor suppresses tumor growth of esophageal squamous cell carcinoma in a mouse model. Oncotarget. 2017;8(70):114457–114462.

13.

Matrka

Cimperman

Haas

, et al Dek overexpression in murine epithelia increases overt esophageal squamous cell carcinoma incidence. PLoS Genet. 2018;14(3):e1007227.

14.

Miyamoto

Yasui

Kim

, et al A novel rasH2 mouse carcinogenesis model that is highly susceptible to 4-NQO-induced tongue and esophageal carcinogenesis is useful for preclinical chemoprevention studies. Carcinogenesis. 2008;29(2):418–426.

15.

Morgan

Soerjomataram

Rumgay

, et al The global landscape of esophageal squamous cell carcinoma and esophageal adenocarcinoma incidence and mortality in 2020 and projections to 2040: new estimates from GLOBOCAN 2020. Gastroenterology. 2022;163:649–658.e2.

16.

Morita

Saeki

Mori

, et al Risk factors for esophageal cancer and the multiple occurrence of carcinoma in the upper aerodigestive tract. Surgery. 2002;131(suppl 1):S1–S6.

17.

Napier

Scheerer

Misra

Esophageal cancer: a review of epidemiology, pathogenesis, staging workup and treatment modalities. World J Gastrointest Oncol. 2014;6(5):112–120.

18.

Osei-Sarfo

Urvalek

Tang

, et al Initiation of esophageal squamous cell carcinoma (ESCC) in a murine 4-nitroquinoline-1-oxide and alcohol carcinogenesis model. Oncotarget. 2015;6(8):6040–6052.

19.

Panigrahi

Walker

IG.

The N2-guanine adduct, but not the C8-guanine or N6-adenine adducts formed by 4-nitroquinoline 1-oxide, blocks the 3’-5’ exonuclease action of T4 DNA polymerase. Biochemistry. 1990;29:2122–2126.

20.

Reichenbach

Murray

Saxena

, et al Clinical and translational advances in esophageal squamous cell carcinoma. Adv Cancer Res. 2019;144:95–135.

21.

Rustgi

El-Serag

HB.

Esophageal carcinoma. N Engl J Med. 2014;371(26):2499–2509.

22.

Siegel

Miller

Jemal

Cancer statistics, 2018. CA Cancer J Clin. 2018;68:7–30.

23.

Tang

Albert

Scognamiglio

, et al A DNA methyltransferase inhibitor and all-trans retinoic acid reduce oral cavity carcinogenesis induced by the carcinogen 4-nitroquinolone 1-oxide. Cancer Prev Res. 2009;12:1100–1110.

24.

Tang

Knudsen

Bemis

, et al Oral cavity and esophageal carcinogenesis modeled in carcinogen-treated mice. Clin Cancer Res. 2004;10:301–313.

25.

Tétreault

M-P

. Esophageal cancer: insights from mouse models. Cancer Growth Metastasis. 2015;8(suppl 1):37–46.

26.

The Cancer Genome Atlas Research Network. Integrated genomic characterization of oesophageal carcinoma. Nature. 2017;541(7636):169–175.

27.

Thomas

Husain

Chaney

, et al Sequence effect on incision by (A)BC exonuclease of 4NQO adducts and UV photoproducts. Nucleic Acids Res. 1991;19:365–370.

28.

Venkat

Shami

Davis

, et al Relative genotoxic activities of pesticides evaluated by a modified SOS microplate assay. Environ Mol Mutagen. 1995;25:67–76.

29.

Yang

Guan

Men

, et al Comparable molecular alterations in 4-nitroquinoline 1-oxide-induced oral and esophageal cancer in mice and in human esophageal cancer, associated with poor prognosis of patients. In Vivo. 2013;27:473–484.

30.

Yao

Cui

Fan

, et al Single-cell transcriptomic analysis in a mouse model deciphers cell transition states in the multistep development of esophageal cancer. Nat Commun. 2020;11:3715.

31.

Zheng

Zhaoyu

A sex-dimorphic mouse model of esophageal squamous cell carcinoma. Am J Cancer Res. 1992;9(2):429–433.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.31 MB