Alterations in the Expression of p53, KLF4, and p21 in ...

Alterations in the Expression of p53, KLF4, and p21 in Neuroendocrine Lung Tumors ´ Jose´ Manuel Naranjo Gomez, MD, PhD; Jose Fernando Val Bernal, MD, PhD; Pilar Garc´ıa Arranz, MD; ´ ´ Fernańdez, MD; Jose Javier Gomez Servando Lazuen Roman, MD, PhD

Context.—Neuroendocrine lung neoplasms are a heterogeneous group of tumors with different clinical behavior and prognosis. Objectives.—To evaluate the expression of p53, KLF4, and p21 in neuroendocrine lung neoplasms and to analyze the influence that expression has on the prognosis of those tumors. Design.—All neuroendocrine lung neoplasms (N ¼ 109) resected in our institution were reviewed, with the collection of histologic slides and paraffin blocks of 47 typical carcinoids (43%), 9 atypical carcinoids (8%), 35 large cell neuroendocrine carcinomas (32%), and 18 small cell lung carcinomas (17%), as well as 10 tumorlets (100%). Four tissue microarrays were performed. Followup was assessed in all cases (119 of 119; 100%). Results.—p53 protein immunostaining results were negative in both the tumorlets and typical carcinoids and

were overexpressed in 11% (1 of 9) of the atypical carcinoids and in 68% (36 of 53) of the carcinomas. KLF4 results were positive in all tumorlets (10 of 10; 100%), 32% (15 of 47) of the typical carcinoids, 44% (4 of 9) of the atypical carcinoids, and 62% (33 of 53) of the carcinomas. p21 expression did not differ among the groups. The lack of KLF4 and p21 expression was associated with an accumulation of aggressive features in typical carcinoids (P ¼ .04 and P ¼ .004, respectively, Fisher exact test). Conclusions.—p53, KLF4, and p21 showed altered expression patterns in pulmonary neuroendocrine neoplasms. Lack of KLF4 and p21 expression was associated with accumulation of aggressive features in typical carcinoids. (Arch Pathol Lab Med. 2014;138:936–942; doi: 10.5858/ arpa.2013-0119-OA)

A

proteins p53, KLF4, and p21. When DNA has been damaged, the p53 gene is activated and exerts a potent regulatory effect that leads to cell cycle arrest or apoptosis. Indeed, mutations in this gene are among the most frequent genetic lesions detected in cancer.5–6 p53 induces the expression of several genes, such as the cyclin-dependent kinase inhibitor p21 (also called WAF1 or CIP1), which leads to growth arrest in G1.7–9 On the other hand, mutations in p21 are extremely rare.10 The gut-enriched Kruppel-like ¨ factor (GKLF, KLF4) is a transcription factor with a high level of expression in gastrointestinal epithelial cells, the skin, and the thymus.11 KLF4 has an inhibitory effect on cellular growth in most cells,12 and it is an important component of the p53 tumor suppressor pathway.13,14 In fact, p53 activation produces an increase in KLF4 synthesis. p53 interacts physically with KLF4, allowing access to the proximal p21 promoter, at the SP1-1 site. p21 is induced and activated by that pathway, producing an immediate arrest in cell cycle progression, blocking G1/S phase transition. Because we believe that the expression of proteins that control the cell cycle differs between low- to intermediategrade and high-grade neuroendocrine tumors, we evaluated the status of p53, KLF4, and p21 in these neoplasms, as well as their relationship with patient prognosis and survival.

bout 20% of lung neoplasms show neuroendocrine characteristics on pathologic evaluation.1–3 Those neoplasms have several common morphologic and biochemical characteristics, such as the presence of dense cytoplasmic granules and the ability to synthesize neuropeptides. This heterogeneous group of tumors has been the object of considerable debate, leading to multiple classifications, the last being in 2004.4 This spectrum of entities includes low-grade typical carcinoids (TCs) and intermediate-grade atypical carcinoids (ACs) as well as the highgrade categories of large cell neuroendocrine carcinomas (LCNECs) and small cell lung carcinomas (SCLCs). Moreover, limited, small-size tumors, known as tumorlets, are considered benign preneoplastic neuroendocrine cell proliferations. DNA repair is a constant cell process that is essential to cell survival and protects the genome from damage and mutations. That process is mediated by the central cell cycle

Accepted for publication September 17, 2013. ´ From the Departments of Thoracic Surgery (Dr Naranjo Gomez) and Pathology (Drs Val Bernal, Garc´ıa Arranz, Lazue´ n Fernańdez, ´ and Gomez Roman), Hospital Universitario Marque´ s de Valdecilla, Santander, Spain. The authors have no relevant financial interest in the products or companies described in this article. ´ Reprints: José Manuel Naranjo Gomez, MD, PhD, Department of Thoracic Surgery, Hospital Universitario Marqués de Valdecilla, Avda de Valdecilla, Santander, Cantabria 39012 Spain, (e-mail: [email protected]). 936 Arch Pathol Lab Med—Vol 138, July 2014

MATERIALS AND METHODS We studied 109 neuroendocrine cell neoplasms from surgical specimens, 10 tumorlets found incidentally in lung tissue resected for other reasons, and 4 normal lung samples selected from the ´ p53, KLF4, and p21 in Neuroendocrine Lung Tumors—Naranjo Gomez et al

surgical pathology archives of the Hospital Universitario Marques de Valdecilla (Santander, Spain). The surgical specimens were obtained from patients who had undergone surgery with intention to cure from 1988 to 2005. All the histologic slides were reviewed, and the initial diagnosis was confirmed by 2 pathologists (P.G.A. and J.J.G.R.). All the paraffin-embedded, donor tissue blocks were subjected to manual microdissection and sampled with a 0.6-mm punch using a Beecher tissue microarray instrument (Beecher Instruments Inc, Sun Prairie, Wisconsin). Two samples of each specimen were prepared to avoid neoplastic heterogeneity. The 10 tumorlets (100%) and 47 TC (43%), 9 AC (8%), 35 LCNEC (32%), and 18 peripheral SCLC (17%) tissues were placed in 4 tissue microarrays. Paraffin tissuearray blocks containing arrayed core cylinders underwent routine hematoxylin-eosin staining (Leica Biosystems, Wetzlar, Germany) and immunohistochemical staining for the KLF4 protein (monoclonal antibody dilution, 1:50, Santa Cruz Biotechnology, Inc, Santa Cruz, California), the p21 protein (monoclonal antibody dilution, 1:100, Dako, Glostrup, Denmark), and the p53 protein (DO7 clone dilution, 1:100, Dako). To block endogenous peroxidase activity, the sections were treated with 0.3% hydrogen peroxide for 30 minutes. Antigen recovery was performed by boiling the sections for 90 seconds in 10mM sodium citrate buffer, at pH 6.0, using a pressure cooker. For the immunohistochemical study, 5-lm sections were prepared on precoated slides and stained following the labeled-polymer method EnVisionþ (Dako) using a Techmate 550 automated immunostainer (BioTek, Santa Barbara, California). Diaminobenzidine (Dako) was used as a chromogen. The slides were dehydrated after counterstaining with Mayer hematoxylin (Merck, Darmstadt, Germany). As controls, histologic sections of each case were processed without the addition of a primary antibody for each antigen retrieval method along with positive/ negative, multitissue control sections. The expression of these proteins was examined by the 2 pathologists mentioned above, without any information about clinicopathologic features or prognosis. Immunostaining was evaluated as follows: the percentage of positive cells, regardless of the intensity, was scored semiquantitatively: negative (), 0% to 9%; weakly positive (þ), 10% to 25%; moderately positive (þþ), 26% to 50%; and strongly positive (þþþ), 51% to 100%. Thus, the samples were classified according to positive status into negative () and positive (þ, þþ, þþþ) expression. Negative staining was considered as the absence of mutations in p53 (p53wt) and lack of expression in the case of KLF4 and p21. On the other hand, positive results were interpreted as a mutation in p53 and protein overexpression in KLF4 and p21. Neoplastic nuclei that could not be evaluated were excluded from the analysis. The clinical reports of the patients were reviewed, and the sociodemographic variables, pathologic staging, resection technique, adjuvant treatment, and follow-up data were collected. The SPSS 8.0 program (IBM, Armonk, New York) was used for statistical analysis, and the Fisher exact test, Kaplan-Meier, and log rank survival tests were applied. The Ethical Committee at our center approved this study.

RESULTS Clinical Data For the 47 TC tissues (women, 28 [60%]; men, 19 [40%]), the mean (SD6) age of the patients at presentation was 49 (17) years. The general approach was thoracotomy: 2 (4%) underwent pneumonectomy, 38 (81%) had lobectomies, and 7 (15%) had wedge resection (these were peripheral tumors with relapse in none of the cases). None of these patients was on adjuvant treatment. For the 9 AC tissues (women, 5 [56%]; men, 4 [44%]), the mean (SD) age at presentation was 51 (20) years. Three patients (33%) underwent pneumonectomy, and 6 (67%) Arch Pathol Lab Med—Vol 138, July 2014

had lobectomies. None of these patients was on adjuvant treatment. For the 35 LCNEC tissues (women, 1 [3%]; men, 34 [97%]), the mean (SD) age at presentation was 64 (9) years. Three patients (9%) underwent pneumonectomy, and 32 (91%) had lobectomies. Patients with N1 and N2 disease received cisplatin-based adjuvant treatment. For the 18 peripheral SCLC cases (all men), the mean (SD) age at presentation was 62 (10) years. Five patients (28%) underwent pneumonectomy, and 13 (72%) lobectomy. All (18 of 18; 100%) were on cisplatin-based adjuvant treatment and prophylactic cranial irradiation. Moreover, all the patients with N2 disease underwent mediastinal irradiation. Immunohistochemistry Normal lung samples were negative for the 3 immunohistochemical markers. All the tumorlets (10 of 10; 100%) were p53 and KLF4þ. p21 results were positive in 6 cases (60%) (Figure 1, A). The TCs were negative for p53 immunostaining (p53wt); KLF4 was positive in 15 of the 47 cases (32%). p21 immunostaining showed no differences in the percentage of cases that were immunoreactive in these 2 categories (Figure 1, B). We observed one AC case (1 of 9; 11%) positive for p53. The AC response to KLF4 and p21 was similar to that of the TC tissues (Figure 1, C). We observed 28 of 35 positive LCNEC cases (80%) for p53, 23 of 35 positive cases (66%) for KLF4, and 33 of 35 positive cases (94%) for p21 (Figure 1, D). Of the 18 SCLC cases, 11 (61%) were positive for p53, all (18 of 18; 100%) were positive for KLF4, and 16 of 18 (89%) were positive for p21 (Figure 1, E). Some examples are shown in Figure 2, A through P. Survival No patient (0 of 10; 0%) with a tumorlet died. The 5- and 10-year survival rates for the patients with TCs were 98% (46 of 47) and 89% (42 of 47), respectively, whereas, the survival rate at 5 years was 44% (4 of 9) for AC, 37% (13 of 35) for LCNEC, and 22% (4 of 18) for SCLC. The disease stages are shown in Table 1. The mean follow-up was 103 months. Prognosis Correlated With Immunohistochemistry According to the positivity or negativity of the immunostaining of the proteins studied, all the TCs with aggressive features (8 of 8; 100%) were negative for KLF4, and most (5 of 8; 63%) were for p21 as well. In the TC cases, a central localization (specifically, in the main bronchus) was considered an aggressive feature, as was having N1 or N2 disease, and the presentation of distant metastasis on follow-up.15 Out of 32 KLF4 patients, we found 8 aggressive cases (25%): 2 of the 8 patients (25%) with liver metastases on follow-up, 1 (13%) presenting with pulmonary recurrence, 2 patients (25%) with tumors located in the main bronchus, 1 (12%) with liver and brain metastases on follow-up, and 2 patients (25%) with mediastinal or hilar lymph node metastases (1 with N2 disease, and the other with N1 disease). None of the 15 KLF4þ TCs (0%) showed aggressive features (P ¼ .04, Fisher exact test; Table 2). We tried to demonstrate a correlation between higher TNM stage and lack of KLF4 expression. However, for the T descriptor, we did not find any differences between the T1 and the T2 group in relation to the expression of KLF4. A

´ p53, KLF4, and p21 in Neuroendocrine Lung Tumors—Naranjo Gomez et al 937

Figure 1. Immunohistochemical results. A, Tumorlet. All the tumorlets were p53 and KLF4þ. B, Typical carcinoid (TC). The TCs were negative for p53 immunostaining (p53wt); KLF4 was positive in 15 of 47 cases (32%). p21 immunostaining showed no differences in the percentage of cases that were immunoreactive in these two categories. C, Atypical carcinoid (AC). We observed one positive case for p53. The response of KLF4 and p21 was similar to that of the TCs in number of cases. D, Large cell neuroendocrine carcinoma (LCNEC). E, Small cell lung carcinoma (SCLC). Most LCNEC and SCLC cases showed p53, KLF4, and p21 immunoreactivity. Positive cases are in black; negative cases are in grey. Cases in which neither positivity nor negativity could be determined are in white.

correlation was only achieved if we considered the cases involving the main bronchus as a subset of the T2, regardless of the size, and added those cases to those cases with nodal or distant metastases. Five- and 10-year survival rates were 100% (15 of 15) in KLF4þ patients, and 97% (31 of 32) and 84% (27 of 32), respectively, in KLF4 patients (log rank, 1.21, P ¼ .27; Figure 3, A). The patients who died (6 of 47; 13%) did so because of disease recurrence. Interestingly, although 3 of 38 p21þ patients (8%) developed aggressive features, in 9 p21 cases, 4 patients (44%) presented with those features (P ¼ .004, Fisher exact test; Table 3). Again, no correlation was obtained between TNM and the lack of p21 expression. The survival curves did not differ. In the multivariate analysis, we could not determine whether negativity in the expression of KLF4 was an independent predictor of poor prognosis. For the AC cases, 5 of the 9 patients (56%) were KLF4, and 4 of the 5 (80%) died from tumor recurrence or metastasis. The surviving patient (1 of 5; 20%) remained well at 71 months after surgery. Four KLF4þ patients (44%) were identified, 2 of the 4 (50%) were alive and well, and 1 (25%) presented with liver metastases 4 years after surgery, was on treatment, and remained alive and well 4 years later. The other KLF4þ patient (25%) died from tumor recurrence after 28 months, and that tumor was also positive for p53 (P ¼ .20, Fisher exact test; Table 4). The 5-year survival was 75% (3 of 4) for KLF4þ patients and 20% (1 of 5) for KLF4 patients (log rank, 2.70, P ¼ .10; Figure 3, B). No differences in either survival or mortality were found for p21. In relation to the high-grade neuroendocrine carcinomas, we analyzed the mortality and survival rates for all the 938 Arch Pathol Lab Med—Vol 138, July 2014

stages in surgically resected cases, and no differences were found in p53, KLF4, or p21 status. COMMENT The spectrum of the neuroendocrine pulmonary tumors ranges from incidental benign tumorlets to highly malignant small cell carcinomas. This histology-based classification has demonstrated a significant effect on patient survival, with great differences among low-, intermediate-, and highgrade tumors. Typical pulmonary carcinoids are neoplasms that present a 10-year survival rate close to 90%, if properly treated with anatomic resection and mediastinal lymph node dissection.16–19 However, with respect to ACs, the median 10-year survival drops to 25% to 69%,18–21 and was separately found to be 21% at 5 years for LCNEC18 and 9% at 5 years for SCLC.18 Our series shows an analogous survival rate. From a molecular point of view, these prognostic differences remain to be clarified, and genetic profiling, as well as the molecular pathways implied in neoplastic development, has not, to date, been well described. We analyzed the influence of the p53, KLF4, and p21 proteins on neuroendocrine lung neoplasm prognosis and patient survival. These three proteins are pivotal in cell cycle control mechanisms, being related to cell cycle arrest. Pulmonary tumorlets are self-limited neuroendocrine cell proliferations that are mainly diagnosed as incidental findings. None of the 10 cases described in this study expressed the p53 biomarker. However, all showed KLF4 overexpression, a well-known sign of DNA damage, suggesting the involvement of cell cycle arrest in this kind

´ p53, KLF4, and p21 in Neuroendocrine Lung Tumors—Naranjo Gomez et al

Figure 2. A, Microarray, panoramic view. B, KLF4þ tumorlet. C, KLF4þ typical carcinoid. D, KLF4 typical carcinoid. E, p53 atypical carcinoid. F, p53þ atypical carcinoid. G, p21 large cell neuroendocrine carcinoma (LCNEC). H, p21þ LCNEC. I, p53 small cell lung carcinoma (SCLC). J, p53þ SCLC. K, KLF4 atypical carcinoid. L, KLF4þ atypical carcinoid. M, p21 typical carcinoid. N, p21þ typical carcinoid. O, KLF4 LCNEC. P, KLF4þ LCNEC. Positive cases show brown nuclei staining, whereas negative cases show blue nuclei staining (original magnification 325 [A]; KLF4, original magnifications 3100 [B, C, K, L, and P] and 350 [D and O]; p53, original magnifications 350 [E and I] and 3100 [F and J]; p21, original magnifications 350 [G and M] and 3100 [H and N]).

of lesion. Thus, maintenance of KLF4 expression indicates continued tumorlet cell cycle control impeding cell cycle progression and aggressive behavior. We observed that the TCs that did not express either KLF4 or p21 showed significant accumulations of aggressive Arch Pathol Lab Med—Vol 138, July 2014

features. Survival was also shorter in the KLF4 cases. In the TC group, 2 different subgroups were described: those expressing KLF4 and maintaining control over the cell cycle, and those who did not, with subsequent loss of control, allowing the cell cycle to progress and proliferate. This effect


Table 1.

Pathologic Stages of the Neuroendocrine Lung Tumors Studied

Stage

Table 2.

Frequency, No. (%)

TC T1N0 T1N2 T2N0 T2N1 Total AC T1N0 T1N1 T2N0 T2N1 Total LCNEC T1N0 T2N0 T3N0 T2N1 T1N2 T2N2 T3N2 Total SCLC T1N0 T2N0 T1N1 T2N1 T1N2 T2N2 T3N2 T4N2 Total

30 1 15 1 47

(64) (2) (32) (2) (100)

2 1 3 3 9

(22) (11) (33) (33) (100)

6 19 3 1 1 4 1 35

(17) (54) (9) (3) (3) (11) (3) (100)

3 4 1 3 1 4 1 1 18

(17) (22) (6) (17) (6) (22) (6) (6) (100)

Abbreviations: AC, atypical carcinoid; LCNEC, large cell neuroendocrine carcinoma; SCLC, small cell lung carcinoma; TC, typical carcinoid.

was evident in the significant accumulation of aggressive features that we demonstrated to be associated with the lack of expression in these neoplasms. For survival, the increase in difference over time was of note, especially after 90

KLF4þ KLF4 Total a

Typical Carcinoid Prognosis According to KLF4 Status,a n ¼ 47

Good Prognosis, No. (%)

Poor Prognosis, No. (%)

Total, No. (%)

15 (32) 24 (51) 39 (83)

0 (0) 8 (17) 8 (17)

15 (32) 32 (68) 47 (100)

P ¼ .04 by Fisher exact test.

months. Previously, that difference has been small or null because the malignancy of these neoplasms was low. Therefore, all the patients evolved well following the presentation of the disease, regardless of KLF4 status. However, 3 KLF4 patients died of disease recurrence several years after disease onset. Although the differences in survival lacked statistical significance, the few deaths occurring in this study may be an influence on the absence of statistical power to demonstrate the lower survival in KLF4-negative cases. A larger series of cases would be necessary to determine whether it is an independent predictor of poor prognosis. For ACs, we also observed the accumulation of aggressive features in KLF4 cases. In our series, one of the patients who died with recurrent disease was p53þ and KLF4þ. Too few AC cases were analyzed to obtain valid conclusions, but this is an uncommon tumor, making the study of a large series difficult. Nevertheless, despite the lack of statistical significance, our data suggest the potential risk of recurrence or metastasis associated with KLF4 and p53 expression in these neoplasms. For high-grade neuroendocrine carcinomas, we did not find any influence of p53, KLF4, or p21 expression on patient outcome. Moreover, the distribution of those proteins was not homogeneous or indicative of any specific type of carcinoma. These cells undergo many genetic alterations, thereby making it difficult to evaluate the clinical consequences of a particular mutation in a single protein.

Figure 3. Survival. A, Typical carcinoid survival according to KLF4 status. B, Atypical carcinoid survival according to KLF4 status.

940 Arch Pathol Lab Med—Vol 138, July 2014


Table 3.



Total, No. (%)

35 (74) 4 (9) 39 (83)

3 (6) 5 (11) 8 (17)

38 (81) 9 (19) 47 (100)

p21þ p21 Total a

Typical Carcinoid Prognosis According to p21 Status,a n ¼ 47

P ¼ .04 by Fisher exact test.

be necessary to determine whether the same occurs in these neoplasms. The expression of KLF4 seems to show a dual effect on neuroendocrine lung neoplasms, with a protective effect in low- and intermediate-grade tumors (tumorlets and TC), which disappears in carcinomas. These findings are important for determining prognostic factors related to neuroendocrine tumors and support their possible use to stratify the TC, and perhaps the AC, cases most likely to present a more-aggressive clinical course. References

Our results suggest that TC and AC are very different neoplasms from LCNEC and SCLC, regarding immunohistochemistry and survival, and that the latter are not particularly different from each other in relation to their expression of these markers. Interestingly, in the literature, there is debate regarding whether LCNEC and SCLC should be lumped together as high-grade neuroendocrine carcinomas or should remain distinct.22,23 As mentioned above, p21 mutations are extremely rare.10 In contrast, p53 mutations are among the most frequent genetic lesions in cancer5–6 and are important in the prognosis of lung cancer.24–30 The lack of KLF4 expression has previously been reported in other types of tumors, especially in large bowel and gastric adenocarcinomas,31–33 and it has also been associated with an aggressive evolution in gastric cancer.33 Likewise, suppression of growth and metastasis in human pancreatic cancer cells by KLF4 expression has been described,34 suggesting that it may behave as a suppressor gene. Conversely, and in contrast to the consideration of KLF4 as a tumor suppressor protein, 3 laboratories have identified it as a potential oncogene.35–37 To date, KLF4 expression has not been studied in lung tumors, to our knowledge. In this study, we describe several changes in KLF4 expression within a neoplastic spectrum, ranging from overexpression in preneoplastic benign lesions, such as tumorlets, to a significant lack of expression in typical and atypical carcinoids, to a high overexpression rate in neuroendocrine lung carcinomas. Similarly, KLF4 overexpression has been reported in oropharyngeal epidermoid carcinoma,35 skin squamous cell carcinomas,38 and intraductal breast carcinomas,35 related to an aggressive course.39 In conclusion, the expression of the proteins controlling relevant processes in the cell cycle is altered in neuroendocrine lung neoplasms, with the lack of KLF4 or p21 expression being associated with a statistically significant accumulation in cases with aggressive features in TC. Although statistical significance was not achieved in the log rank analysis, our data suggest that the lack of KLF4 expression has an important effect on survival in this group of neoplasms. A larger series of cases would be necessary to determine whether it is an independent predictor of poor prognosis. For ACs, further studies with more cases would Table 4.

KLF4þ KLF4 Total a b

Atypical Carcinoid Prognosis According to KLF4 Status,a n ¼ 9



Total, No. (%)

3 (33) 1 (11) 4 (44)

1b (11) 4 (44) 5 (56)

4 (44) 5 (56) 9 (100)

P ¼ .04 by Fisher exact test. The KLF4þ case with poor prognosis was also p53þ.

Arch Pathol Lab Med—Vol 138, July 2014

1. Quaedvlieg PF, Visser O, Lamers CB, Janssen-Heijen ML, Taal BG. Epidemiology and survival in patients with carcinoid disease in The Netherlands: an epidemiological study with 2391 patients. Ann Oncol. 2001;12(9):1295– 1300. 2. Modlin IM, Lye KD, Kidd M. A 5-decade analysis of 13,715 carcinoid tumors. Cancer. 2003;97(4):934–959. 3. Hemminki K, Li X. Incidence trends and risk factors of carcinoid tumors: a nationwide epidemiologic study from Sweden. Cancer. 2001;92(8):2204–2210. 4. Travis WD. The 2004 WHO classification of neuroendocrine lung tumors. In: Travis WD, Brambilla E, Muller-Hermelink H, Harris CC, eds. Pathology and ¨ Genetics of Tumors of the Lung, Pleura, Thymus and Heart. 3rd ed. Lyon, France: IARC Press; 2004:59–63. World Health Organization Classification of Tumors; vol 10. 5. Levine AJ. p53, the cellular gatekeeper for growth and division. Cell. 1997; 88(3):323–331. 6. Nigro JM, Baker SJ, Preisinger AC, et al. Mutations in the p53 gene occur in diverse tumor types. Nature. 1989;342(6250):705–708. 7. Yoon HS, Chen X, Yang VW. Kruppel-like factor mediates p53-dependent ¨ G1/S cell cycle arrest in response to DNA damage. J Biol Chem. 2003;278(4): 2101–2105. 8. Ogryzko VV, Wong P, Howard BH. WAF1 retards S-phase progression primarily by inhibition of cyclin-dependent kinases. Mol Cell Biol. 1997;17(8): 4877–4882. 9. Radhakrishnan SK, Feliciano CS, Najmabadi F, et al. Constitutive expression of E2F-1 leads to p21-dependent cell cycle arrest in S phase of the cell cycle. Oncogene. 2004;23(23):4173–4176. 10. Gartel AL, Radhakrishnan SK. Lost in transcription: p21 repression, mechanisms, and consequences. Cancer Res. 2005;65(10):3980–3985. 11. Zhao W, Hisamuddin IM, Nandan MO, Babbin BA, Lamb NE, Yang VW. Identification of Kruppel-like factor 4 as a potential tumor suppressor gene in ¨ colorectal cancer. Oncogene. 2004;23(2):395–402. 12. Rowland BD, Peeper DS. KLF4, p21 and context-dependent opposing forces in cancer. Nat Rev Cancer. 2006;6(1):11–23. 13. Zhang W, Geiman DE, Shields JM, et al. The gut-enriched Kruppel-like ¨ factor (Kruppel-like factor 4) mediates the transactivation effect of p53 on the ¨ WAF1/Cip1 promoter. J Biol Chem. 2000;275(24):18391–18398. p21 14. Chen X, Jhons DC, Geiman DE, et al. Kruppel-like factor 4 (gut-enriched ¨ Kruppel-like factor) inhibits cell proliferation by blocking G1/S progression of the ¨ cell cycle. J Biol Chem. 2001;276(32):30423–30428. 15. Harpole DH Jr, Feldman JM, Buchanan S, Young WG, Wolfe WG. Bronchial carcinoid tumors: a retrospective analysis of 126 patients. Ann Thorac Surg. 1992;54(1):50–54. 16. Cooper WA, Thourani VH, Gal AA, Lee RB, Mansour KA, Miller JI. The surgical spectrum of pulmonary neuroendocrine neoplasms. Chest. 2001;119(1): 14–18. 17. Galbis JM, Mafé JJ, Baschwitz B, Ortega E, Aranda FI, Rodriguez-Paniagua ´ JM. Tratamiento quirurgico ´ de los tumores neuroendocrinos de localizacion broncopulmonar. Cir Esp. 2003;74(2):201–205. 18. Travis WD, Rush W, Flieder DB, et al. Survival analysis of 200 pulmonary neuroendocrine tumors with clarification of criteria for atypical carcinoid and its separation from typical carcinoid. Am J Surg Pathol. 1998;22(8):934–944. 19. Thomas CF Jr, Tazelaar HD, Jett JR. Typical and atypical pulmonary carcinoids: outcome in patients presenting with regional lymph node involvement. Chest. 2001;119(4):1143–1150. 20. Beasley MB, Thunnissen FB, Brambilla E, et al. Pulmonary atypical carcinoid: predictors of survival in 106 cases. Hum Pathol. 2000;31(10):1255– 1265. 21. Grote TH, Macom WR, Davis B, Greco FA, Johnson DH. Atypical carcinoid of the lung: a distinct clinicopathologic entity. Chest. 1988;93(2):370– 375. ´ ´ 22. Naranjo Gomez JM, Gomez Romań JJ. Behaviour and survival of highgrade neuroendocrine carcinomas of the lung. Respir Med. 2010;104(12):1929– 1936. 23. Asamura H, Kameya T, Matsuno Y. Neuroendocrine neoplasms of the lung: a prognostic spectrum. J Clin Oncol. 2006;24(1):70–76. 24. Suzuki H, Takahashi T, Kuroishi T, et al. p53 mutations in non-small cell lung cancer in Japan: association between mutations and smoking. Cancer Res. 1992;52(3):734–736. 25. Quinlan DC, Davidson AG, Summers CL, Warden HE, Doshi HM. Accumulation of p53 protein correlates with a poor prognosis in human lung cancer. Cancer Res. 1992;52(17):4828–4831.


26. Mitsudomi T, Oyama T, Kusano T, Osaki T, Nakanishi R, Shirakusa T. Mutations of the p53 gene as a predictor of poor prognosis in patients with nonsmall cell lung cancer. J Natl Cancer Inst. 1993;85(24):2018–2023. 27. Steels E, Paesmans M, Gerghmans T, et al. Role of p53 as a prognostic factor for survival in lung cancer: a systematic review of the literature with a meta-analysis. Eur Respir J. 2001;18(4):705–719. 28. Ahrendt SA, Hu Y, Buta M, et al. p53 mutations and survival in stage I nonsmall-cell lung cancer: results of a prospective study. J Natl Cancer Inst. 2003; 95(13):961–970. 29. Mitsudomi T, Steinberg SM, Nau MM, et al: p53 gene mutations in nonsmall cell lung cancer cell lines and their correlation with the presence of ras mutations and clinical features. Oncogene. 1992;7(1):171–180. 30. Brambilla E, Grazzeri S, Mooro D, et al: Immunohistochemical study of p53 in human lung carcinomas. Am J Pathol. 1993;143(1):199–210. 31. Katz JP, Perreault N, Goldstein BG, et al. Loss of Klf4 in mice causes altered proliferation and differentiation and precancerous changes in the adult stomach. Gastroenterology. 2005;128(4):935–945. 32. Wei D, Kanai M, Huang S, et al. Emerging role of KLF4 in human gastrointestinal cancer. Carcinogenesis. 2006;27(1):23–31.

942 Arch Pathol Lab Med—Vol 138, July 2014

33. Wei D, Gong W, Kanai M, et al. Drastic down-regulation of Kruppel-like ¨ factor 4 expression is critical in human gastric cancer development and progression. Cancer Res. 2005;65(7):2746–2754. 34. Wei D, Kanai M, Jia Z, Le X, Xie K. Kruppel-like factor 4 induces p27Kip1 ¨ expression in and suppresses the growth and metastasis of human pancreatic cancer cells. Cancer Res. 2008;68(12):4631–4639. 35. Foster KW, Ren S, Louro ID, et al. Oncogene expression cloning by retroviral transduction of adenovirus E1A-immortalized rat kidney RK3E cells: transformation of a host with epithelial features by c-MYC and the zinc finger protein GKLF. Cell Growth Differ. 1999;10(6):423–434. 36. Suzuki T, Shen H, Akagi K, et al. New genes involved in cancer identified by retroviral tagging. Nat Genet. 2002;32(1):166–174. 37. Rowland BD, Bernards R, Peeper DS. The KLF4 tumour suppressor is a transcriptional repressor of p53 that acts as a context-dependent oncogene. Nat Cell Biol. 2005;7(11):1074–1082. 38. Chen YJ, Wu CY, Chang CC, et al. Nuclear Kruppel-like factor 4 expression ¨ is associated with human skin squamous cell carcinoma progression and metastasis. Cancer Biol Ther. 2008;7(5):777–782. 39. Pandya AY, Talley LI, Frost AR, et al. Nuclear localization of KLF4 is associated with an aggressive phenotype in early-stage breast cancer. Clin Cancer Res. 2004;10(8):2709–2719.
