Qualitative analysis of keratin in oral squamous cell carcinoma using alcian blue–periodic acid schiff, Ayoub-Shklar, and modified papanicolaou as special stains

Abstract

Background:

The morphological features of keratinisation can impact histopathological diagnosis. Routine haematoxylin and eosin (H&E) stain has its limitations in moderately differentiated cases of oral squamous cell carcinoma (OSCC).

Aims:

The aim of this study was to evaluate and compare the characteristics of staining and identification of keratin by routine H&E stain, Alcian Blue–Periodic Acid Schiff (AB-PAS), Ayoub-Shklar, and Modified Papanicolaou (PAP) stain in OSCC cases.

Material and Methods:

Overall, 30 cases of OSCC were included in experimental groups, and 20 cases of well-differentiated squamous cell carcinoma (WDSCC) and ten cases of moderately differentiated squamous cell carcinoma (MDSCC) and ten cases of normal keratinised oral mucosa (NKOM) were included in the control group. Histopathologicaly diagnosed cases of OSCC were selected, and tissue sections from each were prepared. Fisher’s exact test was used for the statistical analysis.

Results and Conclusions:

We observed that the staining intensity was greater in Ayoub-Shklar in WDSCC and in MDSCC followed by modified PAP and AB-PAS-stained sections and the staining specificity was greater in modified PAP in WDSCC and in MDSCC followed by AB-PAS and Ayoub-Shklar in stained sections. We conclude that special stains like AB-PAS, Ayoub-Shklar, and modified PAP stain can prove to be a simple solution to detect the presence and pattern of keratin histologically. The efficacy of distinct identification of keratin pearls and individual cell keratinisation in OSCC-modified PAP stain is better. It can be used as alternative stain to H&E for histological grading of OSCC based on keratinisation.

INTRODUCTION

More than 90% of head and neck cancers are squamous cell carcinoma, and this is particularly common in the developing world.[1] TNM classification is mainly used for the clinical prognostic evaluation for oral squamous cell carcinoma (OSCC), but this staging system is not sufficient for optimal prognostication and must be supplemented by other reliable methods. In 1927, a quantitative grading system was formed by Broder for OSCC. The biological activity of OSCC is descriptively categorised as highly, moderately, poorly, and very poorly (highly anaplastic) differentiated.[2] One of the features that influence histological grading has tendency of OSCC to form keratin. Keratins are structural proteins, found at their highest concentration and diversity in the keratinocytes of cutaneous as well as oral epithelium.[3]

Keratins are also used as marker proteins to differentiate between epithelial tumours and mesenchymal tumours, both histologically and immunohistochemically.[4]

Ayoub-Shklar stain is the one of the stains which are reliable histological markers to demonstrate the presence/absence and degree of keratinisation. In this stain, keratin appears distinct red in colour.[5] The combined technique of Alcian Blue (AB) and Periodic Acid Schiff’s (PAS) (AB-PAS) also improves the staining quality of keratins. Keratin is composed of insoluble proteins that in turn are made up of amino acids with high sulfhydryl and disulphide content and hence can be demonstrated by amino acid histochemical methods such as AB technique.[6] Papanicolaou (PAP) stain is commonly available in oral pathology laboratories used as a routinely staining technique. The main use of orange G6 in the PAP stain is to stain keratin. Superficial cells with high content of keratin stain yellow-orange hue and parabasal cells stain green to blue in colour.[7]

Hence, this study was designed to demonstrate the specificity and intensity of modified papanicolaou (Mod PAP) stain with phloxine - B in tissues where keratin was present. The specificity and intensity were compared with routine haematoxylin and eosin (H&E). Ayoub-Shklar stain and AB-PAS, the commonly used staining methods, were used, so as to device an optimal staining technique for keratin that would be specific, easy, and time-effective.[8] These special stains may highlight small foci of overt epithelial differentiation that sometimes is missed in routine H&E staining. Thus, this study was planned to evaluate and compare the distinct staining of keratin using the H&E stain, AB-PAS, Ayoub-Shklar, and Mod PAP.

MATERIALS AND METHOD

The present study was conducted in the Department of Oral and Maxillofacial Pathology and Oral Microbiology in our institute. Overall, 30 histopathologically diagnosed cases of OSCC [20 cases of well-differentiated squamous cell carcinoma (WDSCC) and ten cases of moderately differentiated squamous cell carcinoma (MDSCC)] and ten cases of normal keratinised tissues like gingival and buccal mucosa were retrieved from the departmental archives.

The only criterion selected for inclusion is that there should be enough tissue material in paraffin blocks. From each block, three serial sections were made of 4-micron thickness and stained by routine H&E stain, AB-PAS, Ayoub-Shklar, and Mod PAP stain for keratin.

The staining procedure was adopted from Ramulu et al.[8] for Mod PAP stain and for Ayoub-shklar stain, and AB-PAS stain procedure was taken from Rao et al.[3] followed for all the cases.

The criteria of clear and distinct levels of staining of keratin pearl and individual cell keratin were analysed through Score 1 (Poor), Score 2 (Satisfactory), Score 3 (Good), and Score 4 (Excellent) [Table 1]. The data were subjected to statistical analysis by Fisher’s exact test with the help of SPSS software version 16.

Table 1.

Criteria selected for evaluation of stains

Criteria	Score
Specificity of the stain	Poor
Staining intensity	Satisfactory
	Good
	Excellent

RESULTS

All four stains showed clear, distinct keratin [Figure 1].

Figure 1.

(a): H&E staining showing keratin pearls in WDSCC (10x), (b): Mod PAP staining showing keratin pearls in WDSCC (10x), (c): Ayoub-Shklar staining showing Keratin Pearls in WDSCC (10x), (d): AB-PAS staining showing keratin pearls in WDSCC (10x), (e): H&E staining showing individual cell keratinisation in MDSCC (10x), (f): Mod PAP staining showing individual cell keratinisation in MDSCC (10x), (g): Ayoub-Shklar staining showing individual cell keratinisation in MDSCC (10x), (h): AB-PAS staining showing individual cell keratinisation in MDSCC (10x), (i): H&E staining showing Individual keratin pearl (40x), (j): Modified PAP staining showing individual keratin pearl (40x), (k): Ayoub-Shklar staining showing individual keratin pearl (40x), (l): AB-PAS staining showing individual keratin pearl (40x), (m): H&E staining showing surface keratin (10x), (n): Mod PAP staining showing surface keratin (10x), (o): Ayoub-Shklar staining showing surface keratin (10x), (p): AB-PAS staining showing surface keratin (10x)

When staining intensity was analysed in WDSCC-stained sections, it was greater in Ayoub-Shklar, followed by Mod PAP and AB-PAS [Table 2]. Similarly, the staining intensity in MDSCC-stained sections was greater in Ayoub-Shklar, followed by Mod PAP and AB-PAS [Table 3]. The intensity in normal keratinised oral mucosa stained sections was greater in Ayoub-Shklar, followed by Mod PAP and AB-PAS [Table 4]. The P value was found to be statistically significant in each of these groups.

Table 2.

Distribution of special stains (intensity) according to different grades and test of association b/w grades and type of stains in group I (WDSCC)

STAINS	1	2	3	4	Total
	Poor	Satisfactory	Good	Excellent
AB-PAS	5 (100%)	10 (58.82%)	5 (21.74%)	0 (0%)	20
A-S	0 (0%)	2 (11.76%)	8 (34.78%)	10 (66.67%)	20
Mod. PAP	0 (0%)	5 (29.41%)	10 (43.48%)	5 (33.33%)	20
TOTAL	5	17	23	15	60
P (fisher’s exact test)	Fisher’s cal.=147.69 P=0.0007* (P<0.01) significant association				--------

*Shows a significant association b/w grades and type of stains at 0.01 level of significance. P<0.01

Table 3.

Distribution of special stains (intensity) according to different grades and test of association b/w grades and type of stains in group II (MDSCC)

STAINS	1	2	3	4	Total
	Poor	Satisfactory	Good	Excellent
AB-PAS	1 (100%)	5 (50%)	4 (30.77%)	0 (0%)	10
A-S	0	2 (20%)	4 (30.77%)	4 (66.67%)	10
Mod. PAP	0	3 (30%)	5 (38.47%)	2 (33.33%)	10
TOTAL	1	10	13	6	30
P (fisher’s exact test)	Fisher’s cal.=129.88 P=0.0004* (P<0.01) significant association				--------

*Shows a significant association b/w grades and type of stains at 0.01 level of significance. P<0.01

Table 4.

Distribution of special stains (intensity) according to different grades and test of association b/w grades and type of stains in group III (NKOM)

STAINS	1	2	3	4	Total
	Poor	Satisfactory	Good	Excellent
AB-PAS	8 (100%)	2 (22.22%)	0 (0%)	0	10
A-S	0	0 (0%)	6 (66.67%)	4 (100%)	10
Mod. PAP	0	7 (77.78%)	3 (33.33%)	0	10
TOTAL	8	9	9	4	30
P (fisher’s exact test)	Fisher’s cal.=159.36 P=0.0002* (P<0.01) significant association				------

*Shows a significant association b/w grades and type of stains at 0.01 level of significance. P<0.01

While evaluating staining specificity in WDSCC-stained sections, it was greater in Mod PAP followed by AB-PAS and Ayoub-Shklar [Table 5]. The staining specificity was greater in Mod PAP in MDSCC followed by AB-PAS and Ayoub-Shklar in stained sections [Table 6]. The staining specificity was greater in Ayoub-Shklar in normal keratinised oral mucosa followed by Mod PAP and AB-PAS in stained sections [Table 7]. The P value was again found to be statistically significant in each of these groups.

Table 5.

Distribution of special stains (specificity) according to different grades and test of association b/w grades and type of stains in group I (WDSCC)

STAINS	1	2	3	4	Total
	Poor	Satisfactory	Good	Excellent
AB-PAS	0	10 (45.45%)	7 (41.18%)	3 (60%)	20
A-S	16 (100%)	3 (13.64%)	1 (5.88%)	0 (0%)	20
Mod. PAP	0	9 (40.91%)	9 (52.94%)	2 (40%)	20
Total	16	22	17	5	60
P (fisher’s exact test)	Fisher’s cal.=118.74 P=0.0003* (P<0.01) significant association				-------

*Shows a significant association b/w grades and type of stains at 0.01 level of significance. P<0.01

Table 6.

Distribution of special stains (specificity) according to different grades and test of association b/w grades and type of stains in group II (MDSCC)

STAINS	1	2	3	4	Total
	Poor	Satisfactory	Good	Excellent
AB-PAS	0 (0%)	6 (60%)	3 (33.33%)	1 (33.33%)	10
A-S	8 (100%)	2 (20%)	0 (0%)	0 (0%)	10
Mod. PAP	0 (0%)	2 (20%)	6 (66.67%)	2 (66.67%)	10
Total	8	10	9	3	30
P (fisher’s exact test)	Fisher’s cal.=136.49 P=0.0004* (P<0.01) significant association				----

*Shows a significant association b/w grades and type of stains at 0.01 level of significance. P<0.01

Table 7.

Distribution of special stains (specificity) according to different grades and test of association b/w grades and type of stains in group III (NKOM)

STAINS	1	2	3	4	Total
	Poor	Satisfactory	Good	Excellent
AB-PAS	5 (100%)	5 (45.45%)	0 (0%)	0	10
A-S	0	0	6 (60%)	4 (100%)	10
Mod. PAP	0	6 (54.55%)	4 (40%)	0	10
Total	5	11	10	4	30
P (Fisher’s exact test)	Fisher’s cal.=109.43 P=0.0003* (P<0.01) significant association				----

*Shows a significant association b/w grades and type of stains at 0.01 level of significance. P<0.01

DISCUSSION

The identification and staining of individual cell keratin and keratin pearl is important in histopathological grading and diagnosis of OSCC. The quality and quantity of keratin synthesis reflect the differentiation level of normal and abnormal epithelial cells.[9]

As compared with H&E staining, not all the keratin pearls present in well-differentiated squamous cell carcinoma tissue sections could be stained positively by the special stains, the possible explanation being that the keratin pearls stained positively with special stains would have undergone the normal maturation process similar to physiologic keratin, whereas the unstained keratin pearls might have certain unknown biochemical differences in the maturation process, rendering them less reactive to the special stains.[10]

Collective analysis of both the parameters, staining intensity and specificity, revealed that it was greater in Ayoub-Shklar in normal keratinised oral mucosa, followed by Mod PAP and AB-PAS in stained sections.

The observations of the present study suggested that the Mod PAP stain stains keratin distinctly, yields nuclear detail, and uses readily available dyes compared to Ayoub-Shklar stain, which do not have a nuclear component. Although routine H-E stain is a gold standard, it is not specific for keratin, while Mod PAP is a polychrome stain which, apart from staining keratin magenta pink, also imparts different hues of colour to structures like nuclei, blue; collagen, green; bone, blue; and muscle and erythrocytes as red in colour. Hence, Mod PAP stain can be used as an alternative special stain to Ayoub-Shklar and AB-PAS stain in staining surface keratin distinctly.

Our results were congruent with the observations of the study conducted by Rao et al. (2015);[3] these authors concluded that A-S and AB-PAS demonstrated overall staining quality comparable to H&E, suggestive of their potential use as alternative stains for keratin.

Similarly, compared the staining specificity and staining intensity of special stains with respect to routine H&E stain. They revealed that Ayoub-Shklar stain good staining specificity and has good staining quality which helps in the identification of keratin. In another study done by Anthwal et al. 2020[11] in OSCC cases, the staining pattern and intensity were better in special stains like Mod PAP, Dane-Herman, and Ayoub-Shklar as compared to H&E stain. In their study, 20%, 40%, and 50% of the Ayoub-Shklar-stained sections of Odontogenic Keratocyst (OKC), Verrucous Vulgaris, and OSCC showed score 3, score 1, and score 2, respectively, where score 3 showed intense staining. The P value was found to be statistically significant.

In contrast, Bala et al. (2020),[12] in their study, assessed that H&E-stained sections showed a higher amount of keratin pearls than AB-PAS, Rapid PAP, and Gram stain. Among the special stains, the staining specificity and staining intensity of AB-PAS were found to be superior, followed by Rapid PAP and Gram stain. However, the overall staining specificity and staining intensity of H&E stain were found to be the best in comparison to all the special stains.

In a study by Caliaperoumal et al. (2017)[13] on OSCC cases, the authors concluded that the efficacy of distinct identification of keratin pearls and individual cell keratinisation in oral SCC by Modified PAP stain is better than modified Mallory’s stain and H&E stain. They further suggested that it can be used as an adjuvant in the case of OSCC. A study conducted by Preethi and Sivapathasundharam (2015)[14] on odontogenic keratocyst as well as well differentiated and moderately differentiated OSCC cases suggested that the PAP stain more clearly demonstrated key morphological features associated with keratinisation and it may be considered as a supplemental staining procedure, aiding in diagnosis. Results of the stained slides revealed a marked difference in staining with respect to colour balance, contrast, and intensity between HE and PAP stains.

CONCLUSION

The observations of the present study suggested that the Mod PAP stain stains keratin distinctly, yields nuclear detail, and uses readily available dyes compared to Ayoub-Shklar stain, which do not have a nuclear component. Although routine H-E stain is a gold standard, it is not specific for keratin, while Mod PAP is a polychrome stain which, apart from staining keratin magenta pink, also imparts different hues of colour to structures like nuclei, blue; collagen, green; bone, blue; and muscle and erythrocytes as red in colour. Hence, Mod PAP stain can be used as an alternative special stain to Ayoub-Shklar and AB-PAS stain in staining surface keratin distinctly.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.