Prevalence of Thyroid Transcription Factor-1 (TTF-1)-Negative Small Cell Carcinoma and Napsin A Positivity in Small Cell Carcinoma in a Cross-Sectional Study of Lung Core Biopsies

Background The prevalence of thyroid transcription factor-1 (TTF-1) and napsin A expression are poorly characterized in lung core biopsies of small cell carcinoma. Locally, the TTF-1 clone is 8G7G3/1 (Agilent/Dako), and the napsin A clone is IP64 (Leica Biosystems). Methods All in-house lung core biopsy reports for cases accessioned at a regional laboratory from January 2011 to December 2020 were retrieved and analyzed using a validated hierarchical free-text string matching algorithm (HFTSMA) to establish the diagnosis. TTF-1 and napsin A were manually coded with the assistance of a logical text parsing tool. All TTF-1-negative small cell lung carcinoma (SCLC) cases had a full report review by pathologists. Results The cohort had 5,867 lung core biopsies, and 232 cases were confirmed as small cell carcinoma on pathologist review. TTF-1 immunostain results were available in 173 SCLC cases, and 16 cases of TTF-1-negative SCLC were confirmed on full report review. These 16 cases had at least one positive neuroendocrine (NE) marker and positive keratin staining; cases with mixed histology or positive CK5/6 staining were excluded. Ki-67 was done in 10/16 cases; the average Ki-67 was 75%. Napsin A was negative in 50/51 small cell carcinomas, and 0/3 TTF-1-negative SCLC had napsin A positivity. Conclusions Standardized immunostain reporting would simplify such analyses. Based on the cohort, approximately 9% (16/173) of SCLC is TTF-1 negative. Napsin A positivity in suspected small cell carcinoma should prompt consideration of an alternate diagnosis or explanation.


Introduction
Small cell lung carcinoma (SCLC) is relatively less common; it represents approximately 15% of lung cancers [1,2]. It has a strong association with cigarette smoking, and most cases are diagnosed at an advanced stage. The diagnosis is usually a constellation of histo/cytomorphology along with immunohistochemical workup. The typical immunohistochemical workup includes neuroendocrine (NE) markers, thyroid transcription factor-1 (TTF-1), and proliferation marker (Ki-67). SCLC is typically TTF-1 positive; however, it has only been studied in modest-size cohorts [3,4]. The classic study by Folpe et al., using the clone 8G7G3/1 (Agilent/Dako), showed that TTF-1 was positive in 20 of 21 SCLCs [5]. A study by Misch et al. using the SP141 TTF-1 clone (Ventana Systems) found 38 of 221 SCLC patients had TTF-1-negative tumors and that the TTF-1 status did not predict survival [3]. A case-control study by Iida et al., with 11 TTF-1-negative SCLCs and 24 TTF-1-positive SCLCs using the SPT24 TTF-1 clone (Leica Biosystems), suggested that TTF-1negative SCLCs express NE markers less frequently and, like Misch et al., do not have a significantly different survival between groups [4].
Napsin A is a commonly used immunostain in the workup of suspected primary lung adenocarcinoma. It is typically positive in lung adenocarcinoma [11]; however, the literature on napsin A expression in lung NE neoplasms is limited. A series of 37 SCLC surgical resection cases were all negative for napsin A [12]; likewise, a series of 36 cytology cases of SCLC were all napsin A negative. [13]. A larger review of napsin A staining suggested positivity of 0%-17% in (lung) small cell carcinoma [11]. Napsin A expression has been examined in a series of 112 large-cell NE carcinomas; in that context, napsin A is positive in 15% of cases [12]. In medium-size pathology practices without sub-specialization and a modest volume of lung biopsies, small cell carcinoma may be a diagnosis that is seen by the individual pathologist once or twice a year. In such environments, where the working diagnosis is small cell carcinoma but the TTF-1 immunostain is negative, an external consultation may be sought due to uncertainty about the tumor sub-type. Prior (unpublished) work showed that individual pathologists in our thoracic referral center diagnose only two to three small cell carcinomas on lung core biopsies per year.

Objective
The primary objective of this study was to estimate the number of TTF-1-negative SCLCs in a large lung core biopsy cohort. The secondary objective was to assess the prevalence of napsin A staining in all small cell carcinomas.

Materials And Methods
Research ethics board approval was obtained to retrieve lung pathology reports; Hamilton Integrated Research Ethics Board (HiREB) issued approval 3811. The TTF-1 clone in use is 8G7G3/1 (Agilent/Dako), and the napsin A clone is IP64 (Leica Biosystems). All in-house lung core biopsy reports for cases accessioned at a regional laboratory from January 2011 to December 2020 were retrieved using complex search criteria (to exclude open lung biopsies and resections) and analyzed using a validated hierarchical free-text string matching algorithm (HFTSMA) to establish the diagnosis. The diagnostic codes used to classify cases and the hierarchy are defined in Appendix A.
A list of immunostains (see Appendix B) was coded by a logical text parsing tool (LTPT) for all lung biopsies. The LTPT separated sentences and determined the immunostain interpretation based on its location in the sentence in relation to phrases for "negative" and "positive." The LTPT classifications were reviewed (and corrected if necessary) by a pathologist reviewing the report text of all (possible) small cell carcinomas identified by the HFTSMA. All possible TTF-1-negative small cell carcinoma cases had a full report review by two pathologists (AN and MB).
The inclusion criteria for cases in this study were: (1) cases reported as "small cell carcinoma," (2) TTF-1negative cases with at least one reported positive NE marker, and (3) reported positive keratin staining (CAM5.2 or AE1/AE3 or CK7). Cases with mixed histology or positive CK5/6 staining were excluded.

Results
The cohort had 5,867 lung core biopsies that came from 4,973 patients. The HFTSMA categorized 5,725 cases (98%). A total of 254 cases were identified as possible small cell carcinoma by the HFTSMA, and 232 cases were classified as small cell carcinoma on a pathologist review of the reports. Cases were excluded if the diagnosis in the report was not unambiguously small cell carcinoma or suggested a mixture of histologic types (15 cases). TTF-1 immunostain results were available in 173 of the 232 cases of confirmed small cell carcinoma. TTF-1 negative cases were excluded if no keratin positivity was reported (seven cases).

TTF-1-negative small cell carcinomas
Sixteen cases of SCLC were confirmed in the full report review. These 16 had at least one positive NE marker (chromogranin A, synaptophysin, CD56) and positive keratin staining with CAM5.2, CK7, or AE1/AE3; cases with mixed histology or positive CK5/6 staining were excluded. CAM5.2 was positive in 13 of 13 cases. CK7 was positive in seven of nine cases. Keratin staining was described in the report as dots in three cases. All cases (16/16) were CD56 positive. Six cases had two NE markers positive, and three cases had all three NE markers positive. Seven were stained for CK20 and were negative. The results of the case are shown in Table  1.

Napsin A staining
Napsin A staining status by the LTPT with pathologist review demonstrated that only 1/51 case was napsin A positive. The napsin A positive case was TTF-1/CD56/synaptophysin (diffuse and strong) positive; chromogranin A was focal and weak; p63 and CK5/6 were negative; Ki67 proliferation index was 90%, and the napsin staining was described as "very rare cells showing cytoplasmic staining with napsin A." All TTF-1negative small cell carcinomas with reported napsin A staining were napsin A negative (0/3) ( Table 1).

Discussion
The findings suggest that TTF-1-negative small cell carcinoma of the lung is an infrequent occurrence. In our environment, it is seen approximately two times a year on lung core biopsy. In environments where lung core biopsies are relatively less frequent, finding that a presumed small cell carcinoma is TTF-1 negative may be disconcerting and may prompt an external review.
In relation to Iida et al., the cohort in this study had more NE marker positivity; this difference may be explained by the different TTF-1 clones used (SPT24 versus 8G7G3/1) [4]. It should be noted that the International Association of Lung Cancer Study (IASLC) has criteria for TTF-1 positivity [14]. It is presumed that these were applied in the routine in-house practice; however, this was not assessed. Similarly, napsin A staining was presumed to be assessed based on positive internal controls (cytoplasmic expression in type II pneumocytes and alveolar macrophages).
Lung tumors with morphology suggestive of small cell carcinoma that is negative for TTF-1 should prompt consideration of a wider differential diagnosis. The differential diagnosis of SCLC includes other primary NE tumors (typical carcinoid, atypical carcinoid, and large cell NE carcinoma), basaloid squamous cell carcinoma, combined adenocarcinoma, small cell carcinoma, small round cell sarcomas both in the Ewing sarcoma family (e.g., Ewing sarcoma) and recently described morphologically similar tumors lacking EWSR1 gene rearrangement (e.g., CIC-DUX4-rearranged and BCOR-CCNB3-rearranged tumors), Merkel cell carcinoma, lymphomas, thoracic SMARCA4-deficient undifferentiated tumors (SMARCA4-UT) with small round cell morphology, NUT carcinoma, and metastasis from small cell carcinoma of HPV-related sites (cervix, head, and neck) [15][16][17]. Ki-67 immunostaining is often very helpful in differentiating NE tumor types, especially in small biopsies [18]. In the context of negative TTF-1 staining, keratin positivity, NE marker positivity, and clinical history are important to make an accurate diagnosis. It should be noted that many tumors have a small cell variant, so there should be a low threshold for consultation with an expert [1].
Napsin A negativity in small cell carcinoma was seen in 98% of cases (50/51). Napsin A was negative in all three TTF-1-negative small cell carcinomas, where a napsin A result was available. The one case that was napsin A positive was reviewed by a fellowship-trained lung pathologist and had an immunoprofile compatible with small cell carcinoma except for the reported napsin A staining. Napsin A positivity, in the context of suspected small cell carcinoma, appears to be very uncommon. We believe napsin A positivity should prompt strong consideration of alternative diagnoses or an alternative explanation, such as napsin A staining in non-tumor cells that are in the background. One consideration is mixed non-small cell and small cell carcinoma.

Conclusions
Useful information can be extracted from free-text pathology reports using text processing programs; however, auditing is required to ensure accurate classification. Standardized immunostain reporting would simplify such analyses. Based on the available data in the cohort, napsin A positivity with clone IP64 in small cell carcinoma is very rare and should prompt consideration of alternate diagnoses. Approximately 9% (16/173) of SCLC is TTF-1 negative with the clone 8G7G3/1.

Additional Information
Disclosures organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.