Home About us Editorial board Ahead of print Current issue Search Archives Submit article Instructions Subscribe Contacts Login 

 Table of Contents  
Year : 2017  |  Volume : 13  |  Issue : 2  |  Page : 213-217

The pitfalls in cytology diagnosis of poorly differentiated neuroendocrine carcinoma of lung and their treatment response

1 Department of Pathology, Kasturba Medical College, Manipal University, Mangalore, Karnataka, India
2 Department of Pathology, Kasturba Medical College, Mangalore, Karnataka, India
3 Department of Radiotherapy, Kasturba Medical College, Mangalore, Karnataka, India

Date of Web Publication23-Jun-2017

Correspondence Address:
Debarshi Saha
Department of Pathology, Kasturba Medical College, Manipal University, Light House Hill Road, Mangalore - 575 001, Karnataka
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/0973-1482.192761

Rights and Permissions
 > Abstract 

Context: Lung is the most common site of small cell carcinoma (SCLC) – a poorly differentiated neuroendocrine carcinoma (PDNEC). SCLC comprises 15–20% of the invasive cancers of the lung.
Aim: This study was conducted to appraise the accuracy and pitfalls of the diagnosis of PDNEC on cytology along with treatment responses if available.
Settings and Design: Retrospective study for 2 years yielded 21 cases on cytology.
Subjects and Methods: Slides of fine-needle aspiration of lymph nodes, the tumor, bronchial brush, and bronchoalveolar lavage specimens were used. The histological correlation was obtained as were treatment responses.
Results: Eighteen SCLCs were confirmed on review. Of these, 13 initial reports were concordant and five, discordant. The rest three cases which initially reported as SCLC were found to be negative (2) and combined SCLC (1). One SCLC with concordant initial and reviewed diagnoses failed to confirm on histopathology. The patients, all heavy smokers, were predominantly males in the seventh to eighth decade age group. The sensitivity and specificity of reviewed diagnoses were better than that of the original. The difference between histopathology and cytology diagnoses (reviewed and original) was statistically insignificant. All patients were categorized as “extensive stage” by positron emission tomography-computerized tomography, and five were treated with etoposide and cisplatin with/without radiotherapy.
Conclusion: Age group (61–70) and gender (males) distribution were statistically significant. Intermediate variants of SCLC may be misdiagnosed as adenocarcinoma. Similarly, combined SCLC may be missed on cytology if the observer does not sustain a high index of suspicion. Unequivocal cytology diagnosis opposed to negative histopathology report demands repeat biopsy.

Keywords: Fine-needle aspiration, National Comprehensive Cancer Network, neuroendocrine carcinoma, response evaluation criteria in solid tumors, small cell lung carcinoma

How to cite this article:
Saha D, Kumar A, Banerjee S, Nirupama M, Sridevi H B, Garg P, Lobo FD. The pitfalls in cytology diagnosis of poorly differentiated neuroendocrine carcinoma of lung and their treatment response. J Can Res Ther 2017;13:213-7

How to cite this URL:
Saha D, Kumar A, Banerjee S, Nirupama M, Sridevi H B, Garg P, Lobo FD. The pitfalls in cytology diagnosis of poorly differentiated neuroendocrine carcinoma of lung and their treatment response. J Can Res Ther [serial online] 2017 [cited 2022 Nov 29];13:213-7. Available from: https://www.cancerjournal.net/text.asp?2017/13/2/213/192761

 > Introduction Top

Neuroendocrine carcinomas (NECs) comprise 25% of primary neoplasms of the lung, of which 23% are poorly differentiated (PDNEC) and the rest 2%, well differentiated (WDNEC).[1] Of the invasive lung cancers, small cell lung carcinoma (SCLC) contributes 15–20%, large cell NEC (LCNEC) approximately 3%, typical carcinoid (TC) ≤2%, and atypical carcinoid (AC) ≤0.2% of the cases.

The lung is the source of 95% of SCLC at inception.[1],[2] Ergo, metastatic SCLC found anywhere in the body should instigate an elementary search for the primary in the lung.[1],[3] Approximately 30% of WDNEC originate in the lung which is second only to the gastrointestinal tract (sans liver and pancreas).[1],[4] Lung NECs currently can be classified as:

  1. Well differentiated (G1) – TC – low grade
  2. G2 – AC – intermediate grade
  3. Poorly differentiated (G3) – SCLC, LCNEC, and combined SCLC (SCLC + non-SCLC) – high grade.[5]

Their widely diversified etiology, therapeutic paradigms, and thereby prognostic ramifications demand an accurate pathologic diagnosis be reached in time. The existing treatment options for SCLC patients provide merely a 5-year survival rate of 2%.[3],[6]

This study was conducted to evaluate the accuracy of the diagnosis of NEC in cytologic (fine-needle aspiration cytology [FNAC], bronchial brush [BB], and bronchoalveolar lavage [BAL]) samples, apprise and reevaluate errors incurred as preparation artefacts and reestablish the need for cytological and histological correlation in either direction. Furthermore, the treatment response of these patients was studied wherever records were available.

 > Subjects and Methods Top

Study design

Totally, 28 cases of lung PDNEC were reported on cytology between January 2014 and August 2015 (1 year and 8 months). Before this period, the smears were faded and, thus, unsuitable for the study. Of these, seven were excluded due to the unavailability of either medical records, slides or histopathological correlation. Cases reported as SCLC, PDNEC either on cytology or their corresponding histopathology sections were considered for the study. Other lung carcinomas were ignored in the current study. The study was started pending clearance by the Institutional Ethics Committee.

Cytology methods and sampling

The sample distribution is shown in [Table 1]. The three FNAC standalone cases consisted of aspirates from the supraclavicular node. Each node was sampled thrice with a 22-gauge needle mounted on a 10 ml syringe held in Cameco syringe pistol. Three to five passes were executed that usually resulted in enough material to be smeared on five to eight slides. The initial slides were immediately immersed in 100% methanol (wet fixed), and the latter slides were left to be air-dried. The wet fixed slides were stained by the standard papanicolaou (Pap) method and then air-dried by May–Grunwald–Giemsa (MGG) stain. All these cases were correlated by computerized tomography (CT)-guided biopsy from the corresponding lung mass. None of these patients suffered from any other primary malignancy after extensive search chiefly in the upper aerodigestive tract, breast, mediastinum, stomach, vertebrae, and prostate.
Table 1: The sample distribution

Click here to view

The BB samples were collected by a minute brush during video bronchoscopy, immediately smeared on glass slides and thereby fixed in 100% methanol. The slides were transferred to the cytology laboratory to be stained by Pap method.

The BAL samples were received in the standard saline solution. They were then centrifuged by the standard method of 1500 rpm for 10 min. After extracting the test tubes, the supernatant was poured off and the sediment then smeared on four slides. The first two slides were fixed in 100% methanol to be stained by Pap. The other two air-dried smears are stained by hematoxylin and eosin (H and E) and Ziehl–Neelsen method each.

In our study, those FNAC samples received along with BB + BAL samples were mostly obtained from both CT-guided and video bronchoscopy-assisted biopsies, in which both FNA samples and biopsy samples for histopathology were acquired. The cytology impressions realized from whichever source was then correlated with the histopathology slides of the same case. Both the cytology and histopathology review processes were blinded to eliminate bias. The author (Debarshi Saha) who reviewed them was not prescient about the original reports. Immunohistochemistry (IHC) was not asked for any of the cases as it is considered superfluous and histopathology alone is sufficient for diagnosis.[7]

Treatment methods

The complete treatment records of five patients were available for study; others had refused treatment, possibly due to high costs incurred in a nonsubsidized private setup like ours. All these patients were clinically staged according to the Veterans' Administration (VA) scheme for SCLC, validated by positron emission tomography (PET)-CT scans.[8]

According to new response evaluation criteria in solid tumors (RECISTs) criteria 2009, “baseline sum diameters” of target lesions were calculated at the commencement of the treatment. After three cycles of chemotherapy (ChT), PET-CT was repeated to assess disease response. The response of the target lesions to treatment is classified into four groups:[9]

  1. Complete response – Target lesions invisible on radiology
  2. Partial response (PR) – Reduction of at least 30% of the “baseline sum diameters” of target lesions
  3. Progressive disease (PD) – Increment of at least 20% of the “baseline sum diameters” of target lesions
  4. Stable disease – Neither enough reduction nor enough escalation in the size of the target lesions for a “PR” or “PD” eligibility, respectively.

Statistical analysis

Statistical analysis was done on SPSS version 20 software, SPSS Inc., Chicago, IL, USA. Chi-square with Fisher's exact test was used for comparison between old and reviewed cytological diagnoses. McNemar test was used for comparison between cytological and histopathological diagnoses. P<0.05 was considered statistically significant.

 > Results and Observations Top

The age and gender distribution of all the patients is shown in [Table 2]. All the SCLC cases were male in their seventh to eighth decade of life except four, who were in their sixth and one female patient in her seventh. The 21 cases reviewed revealed 18 SCLCs, one “combined SCLC” (SCLC + squamous cell carcinoma) and two, “negative for cancer.” Of these 18 SCLC cases, 13 were concordant with the initial report (SCLC) and the rest five, discordant. These five cases were initially reported as PDNEC – three cases, adenocarcinoma – one case, and atypical/suspicious for malignancy – one case. The two “negative for cancer” and one combined SCLC cases were initially reported as SCLC. In addition, one case reported SCLC both initially as well as on review (concordant) did not show any malignancy on histopathology possibly due to inadequate biopsy sampling. The cytology was from BB and the biopsy was accomplished by bronchoscopy. Later, the diagnosis was confirmed by CT-guided biopsy from the alleged mass in the lung. The review scenario is presented in [Figure 1].
Table 2: The age distribution of the patient population

Click here to view
Figure 1: Review scenario of cytology diagnoses

Click here to view

All the 21 cases were confirmed on histopathology.

All patients without exception had smoked heavily in their life.

Comparison between previous and reviewed cytological diagnoses was insignificant (P = 0.549).

Comparison of cytological diagnoses (previous and reviewed) with histopathological diagnoses was insignificant (P = 1.00).

Comparison between previous cytological diagnoses and reviewed diagnoses revealed improvement in sensitivity, specificity, positive predictive value, and negative predictive value as shown in [Table 3].
Table 3: Comparison of original and reviewed diagnoses with respect to sensitivity, specificity, positive predictive value, and negative predictive value

Click here to view

The smears of small cell carcinoma were predominantly cellular, had hyperchromatic cells with very scant cytoplasm in discohesive clusters, and exhibited molding of the nuclei and prominent streaking of nuclear material in their slides. Necrotic and karyorrhectic debris was generally identifiable in their slides [Figure 2]a and [Figure 2]b.
Figure 2: (a and b) Usual appearance of small cell lung carcinoma. (c and d) The intermediate appearance of small cell lung carcinoma

Click here to view

The case diagnosed as adenocarcinoma initially, and SCLC on review had larger cells (>3 lymphocytes), in general, with a moderate amount of cytoplasm with a very low overall cellularity. Nuclear hyperchromasia was lacking, but nuclear molding was conspicuous as was occasional streaking, features ignored initially. This, probably, was the “intermediate” look imparted to SCLC [Figure 2]c and [Figure 2]d.

All the five patients were conferred “extensive stage” according to the VA scheme. They were treated with etoposide and cisplatin and also radiotherapy (RT) if the patient survived the six cycles of ChT. After three cycles of ChT, three patients showed a partial response on PET-CT scan and two, PD. The two PD and additionally two PR patients could not survive the next three cycles, and only one was alive, undergoing RT at the time of the study.

 > Discussion Top

The morphology of small cell carcinoma on histopathology requires the cell size be generally <3 small lymphocyte diameters [1],[7] with a caveat that large cells (>6–7 small lymphocyte diameters) may represent a sizeable proportion, even with an occasional giant cell in their midst.[10] On cytology (FNAC/BB/BAL), the cell size is accepted as 1.5 times small lymphocyte diameter. The molding, hyperchromasia, fine powdery or salt-and-pepper chromatin, and Azzopardi effect are common to both cytology and histopathology with an added feature of streaking artefact possibly due to vigorous smearing technique existent only on cytology.[11],[12]

Two cases originally reported as SCLC, on review, were opined “negative for malignancy.” Their corresponding smears featured immature metaplastic cells, karyorrhectic debris, and apoptotic bodies. The latter two possibly had been mistaken for tumor diathesis initially. The absence of striking irregular hyperchromatic nuclei, nuclear streaking, and signature molding was conspicuous.[12] Immature metaplastic cells are a known pitfall for SCLC. Their higher nucleocytoplasmic ratio, relative hyperchromasia, and tendency to mold may confound the cytologist. Clues to accurate diagnosis are the uniformity of the cells, tendency to remain cohesive, relatively regular nuclear borders, discernible cytoplasm, smaller nuclear size, and an overall monolayered cobblestoned sheet.[12] A detailed clinical history with radiological correlation could also serve as pointers to the diagnosis.

One case of adenocarcinoma previously reported, on review, was sparsely cellular in the Pap-stained smears. The cells appeared tad outsized with larger nuclei enveloped in apparent light staining cytoplasm. The MGG-stained smears revealed just few cells in small clusters. A thorough scanning of all slides exposed definite molding and intense hyperchromasia in few cells, sparse streaking, and karyorrhectic debris – explicit attributes of SCLC.

Many review articles allude the “intermediate” appearance of SCLC as realized above to alert the pathologist to the possibility of diagnosing erroneously a non-SCLC in cases of SCLC. Here, routine H and E staining may not be sufficient to recognize an otherwise obvious SCLC; IHC may have to be resorted to.[1],[7] Thus, Travis lauds the superiority of cytology over histopathology to resolve confidently such ambiguities.[7]

Three cases previously reported as PDNEC, on review, revealed unmistakable features of SCLC. In our opinion, SCLC cases should be reported as such. The other varieties of PDNEC, namely, LCNEC and combined SCLC (SCLC with non-SCLC component, e.g., squamous cell carcinoma or adenocarcinoma) may beget a surgical role in their treatment.[8] Parenthetically, SCLC cases are advised lobectomy with mediastinal lymph node dissection in the T1–T2, N0 stage if pathologic mediastinal staging is negative.[8] Therefore, the likelihood of a different treatment modality in PDNEC cases other than SCLC should urge the pathologist to strive to characterize the neoplasm rather than steering the treating physician toward an indecision created out of his/her own inadequacy by resorting to such generalization as PDNEC.

One case with an initial and reviewed diagnosis of SCLC was reported “negative for malignancy” on histopathology. A consensus opinion taken on the cytology smears still remained SCLC. CT scan clearly showed a mass lesion adjacent to the main bronchus in the right upper lobe. Thus, we concurred the event to be inadequate sampling at biopsy. A repeat biopsy outside our institution was positive for SCLC. We concluded a negative histopathology report opposing a diagnosis of frank SCLC on cytology should have a note appended implying the possibility of inadequate sampling.

All the five patients treated at our setup were staged according to the VA scheme though the current National Comprehensive Cancer Network (version 1.2016) guidelines encourage clinicians to follow the tumor node metastasis staging since it provides to identify patients in limited stage (T1–T2 N0) disease amenable to surgical possibilities.[8] These patients were in the extensive stage of disease. Both the SCLC and the combined (SCLC with admixed non-SCLC) cases in the extensive stage are advised cisplatin and etoposide for four to six cycles [8] as was planned for all the five cases in our study. A response assessment according to the RECIST criteria was performed after three cycles of ChT by PET-CT. Of the two PD cases, one died just after assessment and RT was initiated for the other. Although his performance score was poor, he survived the six cycles of ChT but died soon after. Of the three remaining PR cases, RT was initiated for only one and withheld for the others arguably due to cost constraints. The survivor after six cycles of ChT was the one whom RT was administered and is still alive after a year and a half. The others perished, one at the fifth cycle of ChT and the other after the sixth cycle. A repeat examination of the survivor's smears and biopsy histopathology slides revealed minimal necrosis; the cytology slides showed excessive streaking. This observation of longer survival in a patient with minimal necrosis at the outset may be a coincidence. The oncologist, however, argues that the RT led to a better outcome in this patient.

The reviewer's diagnoses in our study were a trifle better than the original impression of the cytology slides. The reviewer had studied both books and articles on this topic before going over the slides. Therefore, a better level of expertise caused the reviewer to reach an exact diagnosis more often. “Prior knowledge augments diagnostic acumen,” particularly in pathology has been attested by a study by Nivala et al.[13]

 > Conclusion Top

  1. SCLC was significantly found in the male gender in the age group 61-70 years.
  2. Any one cytological feature is insufficient for diagnosis, a constellation of many features should lead to the final impression.
  3. A consensus and unequivocal diagnosis of SCLC made on cytology obviates need for histopathological confirmation as far as commencement of treatment is concerned (significant difference nonexistent between histopathology and cytology diagnoses, P=1.00).
  4. The 'Intermediate' type of SCLC may be misdiagnosed as adenocarcinoma. Similarly, 'Combined SCLC' may be missed on cytology if the observer does not sustain a high index of suspicion.
  5. Biopsy sampling should be adequate. Before delivering a negative histopathological report, a pathologist must review the cytology slides to ensure that a tumor does exist.
  6. Current treatment modality is largely inadequate to control the relentless progression of the disease resulting unfortunately in very early death.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.

 > References Top

Rekhtman N. Neuroendocrine tumors of the lung: An update. Arch Pathol Lab Med 2010;134:1628-38.  Back to cited text no. 1
Walenkamp AM, Sonke GS, Sleijfer DT. Clinical and therapeutic aspects of extrapulmonary small cell carcinoma. Cancer Treat Rev 2009;35:228-36.  Back to cited text no. 2
Fisseler-Eckhoff A, Demes M. Neuroendocrine tumors of the lung. Cancers (Basel) 2012;4:777-98.  Back to cited text no. 3
Gustafsson BI, Kidd M, Modlin IM. Neuroendocrine tumors of the diffuse neuroendocrine system. Curr Opin Oncol 2008;20:1-12.  Back to cited text no. 4
Travis WD, Brambilla E, Muller-Hermelink HK, Harris CC, editors. World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of the Lung, Pleura, Thymus and Heart. Lyon: IARC Press; 2004. p. 19-25.  Back to cited text no. 5
Skuladottin H, Hirsch FR, Hansen HH, Olsen JH. Pulmonary neuroendocrine tumors: Incidence and prognosis of histological subtypes. A population-based study in Denmark. Lung Cancer 2002;37:127-35.  Back to cited text no. 6
Travis WD. Update on small cell carcinoma and its differentiation from squamous cell carcinoma and other non-small cell carcinomas. Mod Pathol 2012;25 Suppl 1:S18-30.  Back to cited text no. 7
National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology (NCCN Guidelines). Small Cell Lung Cancer. Version 1. 2016. p. SCL 1-6, B, C. (Downloaded from NCCN. Org). Available at: https://www.nccn.org/professionals/physician_gls/pdf/sclc.pdf. [Last accessed on 2015 Nov 15].  Back to cited text no. 8
Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R, et al. New response evaluation criteria in solid tumours: Revised RECIST guideline (version 1.1). Eur J Cancer 2009;45:228-47.  Back to cited text no. 9
Bégin P, Sahai S, Wang NS. Giant cell formation in small cell carcinoma of the lung. Cancer 1983;52:1875-9.  Back to cited text no. 10
Travis WD. Advances in neuroendocrine lung tumors. Ann Oncol 2010;21 Suppl 7:vii65-71.  Back to cited text no. 11
Johnston WW, Elson CE. Respiratory tract. In: Bibbo M, Wilbur DC, editors. Comprehensive Cytopathology. 3rd ed. Philadelphia (PA): Saunders Elsevier; 2008. p. 342-6.  Back to cited text no. 12
Nivala M, Lehtinen E, Helle L, Kronqvist P, Paranko J, Säljö R. Histological knowledge as a predictor of medical students' performance in diagnostic pathology. Anat Sci Educ 2013;6:361-7.  Back to cited text no. 13


  [Figure 1], [Figure 2]

  [Table 1], [Table 2], [Table 3]

This article has been cited by
1 Cytologic diagnosis of neuroendocrine neoplasms including carcinoid tumours- A retrospective study
Rary P Mony, Lakshmi K, Renu Sukumaran, Nileena Nayak
Indian Journal of Pathology and Oncology. 2022; 9(2): 123
[Pubmed] | [DOI]
2 Neuroendocrine carcinomas: Cytological mimics and diagnostic dilemmas
Saumya Gupta, Parikshaa Gupta, Manish Rohilla, Nalini Gupta, Radhika Srinivasan, Arvind Rajwanshi, Uma N. Saikia, Pranab Dey
Diagnostic Cytopathology. 2020; 48(5): 440
[Pubmed] | [DOI]


Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

  >Abstract>Introduction>Subjects and Methods>Results and Obse...>Discussion>Conclusion>Article Figures>Article Tables
  In this article

 Article Access Statistics
    PDF Downloaded167    
    Comments [Add]    
    Cited by others 2    

Recommend this journal