A novel immunohistochemical staining method of frozen sections improves the intraoperative diagnosis of Sclerosing Pneumocytoma and bronchiolar adenoma

Abstract

Background

Sclerosing Pneumocytoma (SP) and Bronchial Adenoma (BA) are rare benign tumors that are morphologically difficult to differentiate from lung adenocarcinoma during intraoperative diagnosis. Effective IHC on frozen section might be beneficial for their recognition. This study introduced a novel IHC method that could be applied to intraoperative frozen sections. Its performance was well demonstrated, and the improvement in pathological diagnosis and surgical decisions was analyzed.

Methods

Comparative cohorts of SP and BA were established from surgical samples. The influence of the novel method was analyzed using self-controlled samples. The diagnostic efficacy of the novel IHC on frozen sections was tested using selected markers. The impact of IHC using frozen sections on the pathological evaluations and surgical operations was analyzed.

Results

The new IHC protocol on frozen sections demonstrated superior performance. Compared to IHC on FFPE, the overall accuracy of the selected markers was satisfactory. It improved the diagnostic accuracy of junior pathologists, with the correct diagnosis rate increasing from 38.7% to 93.6% in SP and from 49.1% to 94.5% in BA. Additionally, the rate of lobectomy and segmentectomy decreased from 54.2% to 38.7% in the SP cohort. A unique group featuring single-layered bronchiolar adenomatoid lesions was described.

Conclusions

The optimized IHC protocol for frozen sections enhances intraoperative diagnostic accuracy, facilitating improved surgical decision-making for SP and BA, thereby enhancing the management of diseases that require accurate intraoperative diagnosis. The clinical application of the new IHC method and the special group of adenomatoid lesions warrant further investigations.

Background

Sclerosing Pneumocytoma (SP) and Bronchial Adenoma (BA) are rare benign tumors occurring in the airways or lung [1, 2]. In 2015, SP was clearly defined and subsequently reclassified as a subgroup of thoracic adenoma [3]. Although BA was recently recognized as a distinct adenoma in WHO guidelines in 2021, its diagnostic clarity remains limited [4].

However, the complex morphology of SP and BA continues to challenge pathologists. The SP consisted of bland stromal round cells and cuboidal surface cells. SP exhibits heterogeneous growth patterns that mimic other cell types, complicating diagnosis from frozen or small biopsy samples [3, 5]. BA comprises three cell types: basal cells, cuboidal luminal cells, and mucous cells. The typical architecture of BA was papillary and glandular, which presented luminal cells lined by continuous basal epithelium [1, 4]. Likewise, the morphological diagnosis of BA was even more complex. As in differential diagnosis, BA could be confused with a range of benign diseases and malignancies [6].

As benign tumors, both SP and BA are primarily treated by radical surgical excision [7, 8]. Till now, the surgical method has remained controversial in SP patients [9]. On the other hand, since an indolent clinical course without recurrence after surgery has been reported in BA, extended excision might be unnecessary [6, 10]. Misinterpretation of frozen-section morphology in SP and BA—critical for intraoperative decision-making—can readily result in inappropriate surgical choices. Meanwhile, Immunohistochemistry (IHC) has been widely applied to Formalin-Fixed and Paraffin-Embedded (FFPE) sections to distinguish these tricky lesions. In SP, both stromal and surface cells express Thyroid Transcription Factor-1 (TTF-1). Surface pneumocytic cells are Cytokeratin (CK)-positive, whereas stromal cells are CK-negative [11]. Comparatively, TTF-1 was extensively positive in luminal and basal cells in BA, while CK5/6 merely labeled the basal layer [12]. A proper combination of selected markers of IHC staining might significantly improve the intraoperative diagnosis of diseases with indeterminate morphology, such as SP and BA.

However, previous reports have indicated that standard IHC on surgical frozen sections yields suboptimal performance [13]. In order to minimize the staining duration and optimize the conjugation between the antibody and the antigen from frozen sections in immunohistochemical procedures, a brand-new practical IHC reagent kit was applied to surgical frozen sections. This novel frozen-section IHC protocol was effectively demonstrated and validated for SP and BA.

Materials and methods

Sample collection and processing

All the patients who underwent tumor resection between January 2021 and December 2024 at a tertiary hospital were included, and the samples were obtained from the thoracic surgery department of The Second Xiangya Hospital of Central South University. All the participants who underwent surgical treatment of “lung nodules” have signed a consent form after evaluation of the surgical indication, with the exclusion of contraindications. All the surgeries and operations were performed according to the standard operating procedures and regulations of The Second Xiangya Hospital of Central South University. The cases included those diagnosed as potential SP or BA by frozen sections or the cases that were misdiagnosed as SP or BA. Then, the cases that lacked clinical information, the frozen sections were not available, or biopsy specimens were excluded.

The criteria of pathologists’ diagnostic experience included seniority and clinical experience. The “senior pathologist” referred to those who were experienced in diagnosing frozen sections, with a cumulative frozen diagnosis of over 10,000 cases. The “junior pathologist” referred to those who were merely experienced in diagnosing FFPE sections, with less than a total of 100 frozen diagnoses. The SP cohort was composed of two groups diagnosed with and without IHC on frozen sections, so the surgical procedures were compared directly between these groups. Furthermore, available SP and BA frozen sections were gathered for re-evaluation by pathologists with varied titles before and after IHC staining. Then the diagnostic evaluations were pooled and summarized.

Intraoperative frozen section preparation

Fresh sampling was done by immediately freezing at -26℃ after surgical excision. The frozen sections were sliced on the cryostat (LEICA CM1950) at -21℃ in a thickness of 4–5 μm. The prepared serial sections would be instantly processed according to the following steps. One section from a frozen sample would be stained with Hematoxylin and Eosin (H&E) for the frozen diagnosis. Then, two other sections acquired from the same frozen sample would be treated with the frozen IHC staining kits (CK/TTF-1, CK5/6/TTF-1) for differential biomarkers. Meanwhile, residual tissue from the same tumor was collected and fixed in the normalized formalin for the following routine staining. Routine pathological diagnosis contained H&E and regular IHC stain on the formalin-fixed sections.

The IHC stain on frozen sections

Optimized primary antibodies were developed for the new frozen immunohistochemical technique, with a specially designed antibody diluent. The sensitivity of secondary antibody was enhanced, resulting in augmented conjugation between primary and secondary antibodies, and secondary antibodies and horseradish peroxidases. The protocol achieved ultra-fast diffusion and faster antibody binding kinetics for intraoperative frozen-section IHC workflows.

The IHC stain on the frozen sections was performed as follows: Slices were fixed in the supplied buffer and rinsed before being incubated with specific primary antibodies. Subsequently, the secondary antibody was incubated. Then, Diaminobenzidine (DAB) and hematoxylin were added for the chromogenic reaction. The entire procedure could be accomplished within 20 min. The primary antibodies for IHC staining on frozen sections included CK (AE1/AE3, Mouse), CK 5/6 (C6H1/C1C8, Mouse), and TTF-1(SPT24, Mouse). The dilutions of CK and CK5/6 primary antibodies were 1:50, and the dilutions of TTF-1 primary antibodies was 1:40. The secondary antibodies were provided by the CNT-PolyStacker™ anti-Mouse detection kit.

IHC stain of FFPE sections

The FFPE sections were prepared from the same tumor sample. Section-cutting and routine IHC stain were performed according to conventional procedures. The staining steps were performed according to previous publications [13, 14]. The primary antibodies for IHC staining on FFPE sections included CK (830F6E7, Mouse), CK 5/6 (258A7D1, Mouse), and TTF-1(847G1A6, Rabbit).

Data collection and analysis

All the diagnoses and pathological evaluations were recorded and analyzed by two individual pathologists for each cohort. The improvement of the diagnostic accuracy was demonstrated by Sankey diagram. The agreement between the classifications by IHC on the FFPE and frozen sections was tested by Cohen’s kappa of CK5/6 and TTF-1. The statistical analysis was performed using irr package in R.

Results

Performance of the IHC staining on the frozen sections

According to the established protocol, the time cost of the whole procedure could be limited to 20 min (Fig. 1). The representative staining performance of the IHC method on frozen sections was displayed (Fig. 2). As shown, the staining of different markers was easily distinguishable, which also clearly delineated the lesion margins. To verify the new methods’ efficacy, unique combinations of IHC markers were selected. Initially, 70 cases of potential SP and 65 cases of potential BA were screened. After manual check, a SP cohort of 55 cases and a BA cohort of 55 cases were established.

Fig. 1.

Steps of the IHC protocol on frozen sections

Fig. 2.

Performance of IHC staining on frozen sections. (a-c) SP case. (a) Representative morphology of SP (H&E, 100×). Papillae were composed of round cells (black arrows) and covered by cuboidal cells (red arrows). No fibrovascular stalk was observed. (b) CK was positive focally. (c) TTF-1 was positive extensively. H&E and IHC staining on frozen sections of two BA cases. (d-f) BA case. (d) Representative morphology of BA (H&E, 100×). Heterogeneous tubular structure and papillary tufts were observed. The papillae were lined by a bi-layered epithelium. Ciliated cells (black arrows) and mucinous cells (green arrows) comprised the lumen, layered with basal cells surrounding them (red arrows). (e) CK5/6 marked the basal cells and (f) TTF-1 expressed widely in the ciliated cells and basal cells, not in the mucinous cells. (g-l) Adenomatoid lesion. (g, j) Part of the typical basal epithelium was not identifiable in the glandular structures. Instead, a single layer of epithelial cells lined the lumen with mild atypia. (h, k) The CK5/6 expression was deficient in the single-structured basal layer. (i, l) TTF-1 was positive in luminal ciliated cells and negative in the mucinous cells of the single-structured basal layer

Since adenocarcinoma (ADC) was a major confusing lesion with the SP and BA, this study involved a cohort of 115 consecutive patients who underwent surgical procedures at the Second Xiangya Hospital during the period from March to May 2025. The positive and negative situations and expression foci of CK, CK5/6, and TTF-1, detected by IHC on frozen sections, were consistent with the results of IHC on FFPE sections. Yet, slightly weaker staining strengths were observed in individual biomarkers by frozen IHC (Fig. 3).

Fig. 3.

IHC staining on frozen and FFPE sections. Positivity of CK, CK5/6 and TTF-1 were consistent between frozen sections and FFPE sections in SP and BA. Immunoreactivity of CK, CK5/6 and TTF-1 in variable cells were analogous by different IHC methods

As shown in Table 1, we pooled and compared positive and negative cases from different kinds of tumors in CK, CK5/6 and TTF-1 detection. CK detection was only performed in SP cohort, all the cases were positive in CK staining. Performance of CK5/6 staining was summarized from ADC and BA cases. CK5/6 expression was observed in 55 BA cases and 115 ADC cases. There were 54 positive cases (51 BAs/3 ADCs) detected by IHC on frozen sections compared to 55 positive cases (52 BAs/3 ADCs) by IHC on FFPE sections. Meanwhile, 116 cases (4 BAs/112 ADCs) were detected negative by IHC on frozen sections compared to 115 negative cases (3 BAs/112 ADCs) by traditional IHC on FFPE. Additionally, staining of TTF-1 was employed in 31 SP cases, 55 BA cases and 115 ADC cases, and consistent by these two methods. The TTF-1 positive cases were 198 (31 SPs/53 BAs/114 ADCs) by IHC on frozen sections as well as on FFPE sections. Also, 3 cases (2 BAs/1 ADC) were consistently negative by IHC by both types of IHC. Compared to its staining on FFPE, the consistent positive rate of CK, CK5/6, and TTF-1 was 100%, 96.4%, and 100.0% respectively by IHC on the frozen sections. The cases from the “BA or probable BA” cohort and ADC cohort were pooled to analyze the staining performance of CK5/6 detection. The cases from the SP cohort, “BA or probable BA” cohort, and ADC cohort were pooled to analyze the staining performance of TTF-1 detection. The overall accuracy of TTF-1 detection were 100% which displayed complete consistent positivity between IHC on FFPE and frozen sections. The sensitivity, specificity, and overall accuracy of CK5/6 detection were 96.4%, 99.1%, and 98.2%. (Table 1). As summarized in Table 1, well consistency was observed between the biomarker detection by IHC on the FFPE and frozen sections. In the CK5/6 staining group, the kappa value was 0.96 and the p value was 0. In the TTF-1 staining group, the kappa value was 1 and the p value was 0.

Table 1.

Comparison of biomarker detection between IHC on FFPE and frozen sections

		IHC on FFPE
		CK5/6				TTF-1
		Pos	Neg			Pos	Neg
IHC on Frozen sections	Pos	53	1	p＜0.01	Pos	198	0	p＜0.01
	Neg	2	114		Neg	0	3

Improvement of SP diagnosis by the IHC staining on frozen sections

55 SP cases were screened out in total. Incidentally, 5 concomitant cases of ADC in situ or minimally invasive adenocarcinoma (MIA) were noticed among the 55 SP cases. Since their impact on surgeries was insignificant, the 5 cases have still been included for further analysis. 24 cases were diagnosed intraoperatively without IHC on frozen section, while IHC was applied on the frozen sections in 31 cases. To elucidate the assistance by IHC on frozen sections, the “primary diagnoses” and “final diagnoses” were summarized and compared by pathologists with different professional titles in the 31 cases. Before IHC on the frozen sections, the accurate rate of identification on the cellular morphology of frozen section was 38.7% by the junior pathologist, compared to 71.0% by the senior pathologist. Diagnosis has been improved with the assistance of CK and TTF-1 stains. After IHC on the frozen sections, only 2 cases were determined as “uncertain”. The accurate rate of junior pathologists increased to 93.6% on the frozen sections, equaling that of the senior pathologist (Fig. 4). In cases with intraoperative diagnoses referred to IHC on frozen sections, the proportion of cases that received lobectomy and segmentectomy decreased from 54.2% to 38.7%. The wedge resection proportion increased to 61.3% compared to merely 45.8% cases that received wedge resection in the group who were diagnosed according to H&E stain on frozen sections without IHC (Table 2).

Fig. 4.

Diagnostic evaluation by independent pathologists of frozen sections before and after IHC staining. Diagnostic evaluation by independent pathologists of frozen sections before and after IHC staining. (a) Diagnostic correction by IHC on frozen sections in SP cases. (b) Diagnostic correction by IHC on frozen sections in BA cases

Table 2.

Influence on the surgical operations in SP cohorts by IHC on frozen sections

	Without IHC	With IHC
	(n = 24)	(n = 31)
Wedge	11(45.8%)	19(61.3%)
Segmental	3(12.5%)	3(9.7%)
Lobectomy	10(41.7%)	9(29.0%)

Improvement of BA diagnosis by the IHC staining on frozen sections

Meanwhile, 55 cases with the initial frozen diagnosis of BA or potential BA were collected. All the cases received intraoperative pathological diagnosis using IHC staining. In the pathological re-evaluation, the “BA or probable BA” proportion was 63.6% in the senior pathologist group before IHC on frozen sections, compared to 49.1% by the junior pathologists. Even the senior pathologists reported 7 cases as “uncertain” and 12 cases as “malignant or probably malignant”. After confirmation by FFPE sections and traditional IHC, the 55 cases were verified as 43 BA cases, 3 ADC cases, and 9 cases as another subgroup with intermediate features. As a result, the accurate rate of diagnoses increased to 94.5%, and the pathologists’ assessments were compatible among separate groups (Fig. 4). Surgical operations were more extensive in the BA cohort, which might be attributed to its higher diagnostic uncertainty. With all the cases double-stained with CK5/6 and TTF-1 on both FFPE and frozen sections, 70.9% of the cases received lobectomy and segmentectomy, while 29.1% of the cases underwent wedge resection (Table 3).

Table 3.

Surgical operations in the “potential BA” cohort

	BA	BAL	ADC
	(n = 43)	(n = 9)	(n = 3)
Wedge	14(32.6%)	1(11.1%)	1(33.3%)
Segmental	19(44.2%)	6(66.7%)	2(66.7%)
Lobectomy	10(23.3%)	2(22.2%)	0

*BA Bronchial adenoma, BAL Bronchiolar Adenomatoid Lesions, ADC Adenocarcinoma

Histopathological features of the BA-like lesions

A typical group of “single-layered bronchiolar adenomatoid lesions” was demonstrated. Its intermediate features between BA and ADC were observed. This type of lesion was mainly composed of single-layered glandular structures with mild-to-moderate nuclear atypia. The expression of CK5/6 in adenomatoid lesions was focally positive in the cases, and the positivity of TTF-1 could be either extensive or focal (Fig. 2).

Discussion

In this study, a new protocol of IHC on frozen sections has been demonstrated. It applied the renovated antibodies to signify the target protein expression on frozen sections. Its performance was comparable to IHC on FFPE, which cost only 20 min. Its efficacy was verified in two rare and easily misdiagnosed airway diseases with specific markers. It provided efficient intraoperative assistance for surgeons, thus diminishing unnecessary iatrogenic injury. In the surgical samples, especially the intraoperative frozen sections, misdiagnosis or deferred diagnosis was noticeable in the diagnostic reports since the morphological characteristics were misleading. Generally, frozen sections would present swollen or distorted cellular shapes and structures. As shown in the data, the diagnostic accuracy was superior in the experienced pathologist cohort without IHC on frozen sections. Senior pathologists should be more perceptive of mimicking and rare diseases. However, the diagnostic pitfalls on frozen sections could not be avoided merely by the pathologists’ expertise.

IHC plays a vital role in recognizing and differentiating pulmonary tumors. Proper single or combined immunohistochemical staining significantly improved the diagnostic probability of pulmonary malignancies [15]. So, adequate IHC was recommended for recognition and differentiation of these tumors [16]. With the shifting strategy of screening and management, intraoperative diagnostic on frozen sections was important and more difficult in resected lung nodules [17]. Therefore, pathological diagnosis on intraoperative frozen sections has remained substantially challenging.

As rare or newly-introduced entities, SP and BA should be reminded as mimickers of malignant pulmonary tumors [18]. Disguised structures were observed early in the frozen sections of SP. The solid-dominant SP exhibited spreading small round cells lacking specific architectures that mimicked carcinoid lesions. The papillary-dominant SP was difficult to differentiate from adenocarcinoma. The appearance of stromal cells in the sclerosing subtype could be easily ignored on SP frozen Sects [19, 20]. As in BA, the rate of definite diagnosis on frozen sections could also be significantly low. A mere 3 cases out of 18 were correctly identified in the previous study [21]. Some of the cellular shapes were atypical without the featured pattern under the microscope, so misjudgment overwhelmed the provisional impressions [21]. These results were consistent with a subsequent study with an expanded sample size [12].

TTF-1 was positive in pneumocytes and stromal cells, while CK merely expressed in pneumocytes in SP [22]. Additionally, the CK5/6 staining demonstrated the basal epithelium in BA. On the contrary, the immunostaining of TTF-1 was reported to be heterogeneous in various compartments [23]. Generally, TTF-1 is expressed in ciliated columnar cells and basal cells rather than mucinous epithelium. A combined IHC stain of selected markers might be helpful to clarify the diseases. As varied expressions of different markers were manifested by IHC on FFPE, the qualified performance of IHC on frozen sections was expected. By applying the specific IHC stains, substantial consistency was demonstrated between the IHC stain on FFPE and frozen sections. The overall accuracy of CK, CK5/6, and TTF-1 detection was 100%, 98.2%, and 100% respectively (Table 1). The IHC-stained frozen sections displayed analogous positive and negative expressions of the same markers to the FFPE sections.

The main treatment of SP was surgery. Local excision was recommended for the small peripheral SP, while the lobectomy should be considered for potential metastasis. However, the percentage of SP showing malignant tendencies was at a relatively low level in a large-scale cohort analysis [9, 24]. On the other hand, surgery was the optimal treatment for BA. Limited procedures were favored as long as the resected margins were clean of tumor cells [10, 25]. Based on its benign classification, overdiagnosis or overtreatment should be avoided cautiously [26, 27].

Due to its limited adoption, only part of the SP samples has been stained by IHC on frozen sections, so the advantages brought to the intraoperative diagnosis and surgical procedures by IHC stain were analyzed. The accurate diagnostic rate by the junior pathologist increased by 54.9% after IHC on SP frozen sections (Fig. 4). Besides, the wedge resection rate of SP increased 15.5% with the assistance of intraoperative IHC on frozen sections (Table 2).

On the other hand, although the accurate diagnostic rate by the junior pathologist increased by 45.4% after IHC on frozen sections (Fig. 4), the surgical procedures of BA seemed to be more aggressive. The reasons for the surgeons’ choices could be attributed to the frequent concurrent carcinoma nodules along with the BA lesions, as well as the indecisive pathological evaluations. Generally, the IHC of representative markers on frozen sections bridged the gaps in diagnostic capability among differently experienced pathologists, thus bringing partial solutions to the surgeons’ worries. Confirmed intraoperative diagnoses helped to avoid reoperation and unnecessary injuries.

Incidentally, a special group of adenomatoid lesions was observed in this study. Representative monolayered cellular architectures without CK5/6 expression were identified in this group. Mild nuclear atypia and heterogeneously expressed TTF-1 were also observed. Besides, this particular structure could also be accompanied by the classic bi-layered cells in the same tumor. As a result, CK5/6 was focally positive throughout the whole section. Similar pathological lesions were perceived in previous studies [28, 29], yet confusion about the innate characteristics of the particular lesion has been proposed by other researchers [30, 31]. Due to the insufficient background knowledge and definition, we describe it as a “single-layered bronchiolar adenomatoid lesion” temporarily. Its indefinite features might indicate a special phase that represented the transformation from BA to ADC. In previous studies, bronchiolar adenomatoid lesions were mainly given indefinite diagnoses or misdiagnoses even with IHC. They might have been mistakenly categorized as ADC or BA before. Since their features might imply intermediate biological behavior between benign and malignant lesions, the morphological, molecular, and genetic features warrant further studies. Based on these considerations, the more aggressive surgical approaches described in the BA group might be attributed to the complexity and heterogeneity of the “BA group” per se.

The limitation of this study is the small sample size of the clinical cohorts. The retrospective design of the study might lead to inevitable selection bias. Since the new IHC on frozen sections was applied in 2020, and BA was introduced in 2021 as a new entity [4], the majority of BA cases included in this study were diagnosed with intraoperative IHC on frozen sections. Potentially, BA cases might be misdiagnosed or omitted due to the lack of awareness before 2021. In addition, only a several kinds of primary antibodies have been developed for the new IHC method, which limits its application to recognizing more subcategories of tumors. Finally, the application of this new IHC method prolonged the duration of intraoperative frozen section diagnosis, which led to delayed pathological turnaround time.

Conclusions

In summary, this study introduced and validated a newly developed protocol for IHC on frozen sections. This method has demonstrated superior performance by time-saving and efficacy augmentation of two rare and confusing airway/lung tumors. The method contributed to intraoperative diagnostic accuracy and led to modified surgical operations.

More cases or prospective trials should be considered to validate the method further. The situation when the reoperations might be attributed to the malignant lesions omitted on the intraoperative sections also needs attention in the following studies. Moreover, a unique type of adenomatoid lesions without a clear definition and classification was noticed. The exploration of its nature is expected in future investigations.

Authors’ contributions

Liu drafted and revised the manuscript, Luo and Liang conceptualized the idea and designed the study.

Funding

This study was supported by the following fundings: The National Natural Science Foundation of China (No:82400115), The National Natural Science Foundation of China (No:82200019), Natural Science Foundation of Hunan Province of China (No:2024JJ6583), Natural Science Foundation of Hunan Province of China (No:2025JJ50750), The National Key Clinical Specialty Construction Projects of China.

Data availability

The data used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Written informed consent was obtained from all individual participants included in the study. The ethical approval for this study was obtained from the Ethics Committee of the Second Xiangya Hospital (Approval No. KYZ20250043).

Consent for publication

All authors have read and approved the final manuscript for publication.

Competing interests

The authors declare no competing interests.