Neural abnormalities of reward processing in adolescents with bipolar disorders: An ERP study

Adolescent onset is common in bipolar disorders (BDs) and is associated with a worse illness course in adulthood. A model of BDs suggests that a dysregulated behavioral approach system (BAS), a neural system that mobilizes reward-seeking behavior, is at the root of BDs. Normative adolescence is often accompanied by dynamic changes to neural structures underlying the BAS and related cognitive processes. It is possible that adolescent-onset BDs is associated with abnormal BAS neurodevelopment. Consistently, the present study is the first to compare specific BAS-relevant anticipatory and consummatory reward processes as indexed by event-related potentials (ERPs) in adolescents with BDs and typically developing peers. Using a sample of 43 adolescents with BDs and 56 without psychopathology, we analyzed N1 and P3 responses to anticipatory cues and feedback-related negativity (FRN) and P3 responses to feedback stimuli during a monetary incentive delay (MID) task. Hierarchical linear models examined relationships between ERP amplitudes and diagnostic group, MID condition, sex, and age. During anticipation phase, adolescent boys with BDs exhibited significantly larger N1 amplitudes in loss than even or gain trials. During feedback phase, compared to their healthy peers, adolescents with BDs had smaller FRN amplitudes across all conditions. Additional effects involving age, sex and trial type were observed. The findings indicate subtle, non-ubiquitous BAS-relevant neural abnormalities involving early attentional processes during reward anticipation and reward learning following feedback in adolescents with BDs. Adolescents with BDs did not show overall hypersensitive neural responses to monetary reward anticipation or feedback observed in adults with BDs.

Relationship status and perceived support in the social regulation of neural responses to threat

Strong social ties correspond with better health and well being, but the neural mechanisms linking social contact to health remain speculative. This study extends work on the social regulation of brain activity by supportive handholding in 110 participants (51 female) of diverse racial and socioeconomic origins. In addition to main effects of social regulation by handholding, we assessed the moderating effects of both perceived social support and relationship status (married, cohabiting, dating or platonic friends). Results suggest that, under threat of shock, handholding by familiar relational partners attenuates both subjective distress and activity in a network associated with salience, vigilance and regulatory self-control. Moreover, greater perceived social support corresponded with less brain activity in an extended network associated with similar processes, but only during partner handholding. In contrast, we did not observe any regulatory effects of handholding by strangers, and relationship status did not moderate the regulatory effects of partner handholding. These findings suggest that contact with a familiar relational partner is likely to attenuate subjective distress and a variety of neural responses associated with the presence of threat. This effect is likely enhanced by an individual's expectation of the availability of support from their wider social network.

MCT1 Modulates Cancer Cell Pyruvate Export and Growth of Tumors that Co-express MCT1 and MCT4

Monocarboxylate transporter 1 (MCT1) inhibition is thought to block tumor growth through disruption of lactate transport and glycolysis. Here, we show MCT1 inhibition impairs proliferation of glycolytic breast cancer cells co-expressing MCT1 and MCT4 via disruption of pyruvate rather than lactate export. MCT1 expression is elevated in glycolytic breast tumors, and high MCT1 expression predicts poor prognosis in breast and lung cancer patients. Acute MCT1 inhibition reduces pyruvate export but does not consistently alter lactate transport or glycolytic flux in breast cancer cells that co-express MCT1 and MCT4. Despite the lack of glycolysis impairment, MCT1 loss-of-function decreases breast cancer cell proliferation and blocks growth of mammary fat pad xenograft tumors. Our data suggest MCT1 expression is elevated in glycolytic cancers to promote pyruvate export that when inhibited, enhances oxidative metabolism and reduces proliferation. This study presents an alternative molecular consequence of MCT1 inhibitors, further supporting their use as anti-cancer therapeutics.

High expression of AGR2 in lung cancer is predictive of poor survival

Background

Anterior gradient 2 (AGR2) is a protein disulfide isomerase-like protein widely expressed in many normal tissues as well as cancers. In our study, non-neoplastic bronchial epithelial cells as well as non-small cell lung cancer (NSCLC) cells express AGR2 protein.

Methods

AGR2 expression was analyzed on lung tissue microarrays. Tumor staining was correlated with clinical outcomes.

Results

On a lung cancer tissue microarray using immunohistochemistry, expression levels in cancer showed generally decreasing intensities in order from adenocarcinomas with mucinous components, other adenocarcinomas, squamous carcinomas, to large cell carcinomas. The study cohort was comprised of 400 cases. As a group, there was a slight trend of lower expression with increasing tumor grade. AGR2 expression level was a significant predictor of overall survival in younger patients only. Patients under 65 with lower levels showed a significantly better survival for both men and women. Patients over 65, in contrast, showed no such trend.

Conclusions

Nearly all NSCLC tumors show AGR2 expression. Lung cancer expression of AGR2 has prognostic value for younger patients.

Ribonucleotide reductase subunit M2 predicts survival in subgroups of patients with non-small cell lung carcinoma: effects of gender and smoking status

Background

Ribonucleotide reductase catalyzes the conversion of ribonucleotide diphosphates to deoxyribonucleotide diphosphates. The functional enzyme consists of two subunits - one large (RRM1) and one small (RRM2 or RRM2b) subunit. Expression levels of each subunit have been implicated in prognostic outcomes in several different types of cancers.

Experimental Design

Immunohistochemistry for RRM1 and RRM2 was performed on a lung cancer tissue microarray (TMA) and analyzed. 326 patients from the microarray were included in this study.

Results

In non-small cell lung cancer (NSCLC), RRM2 expression was strongly predictive of disease-specific survival in women, non-smokers and former smokers who had quit at least 10 years prior to being diagnosed with lung cancer. Higher expression was associated with worse survival. This was not the case for men, current smokers and those who had stopped smoking for shorter periods of time. RRM1 was not predictive of survival outcomes in any subset of the patient group.

Conclusion

RRM2, but not RRM1, is a useful predictor of survival outcome in certain subsets of NSCLC patients.

A semi-parametric nonlinear model for event-related fMRI

Nonlinearity in evoked hemodynamic responses often presents in event-related fMRI studies. Volterra series, a higher-order extension of linear convolution, has been used in the literature to construct a nonlinear characterization of hemodynamic responses. Estimation of the Volterra kernel coefficients in these models is usually challenging due to the large number of parameters. We propose a new semi-parametric model based on Volterra series for the hemodynamic responses that greatly reduces the number of parameters and enables "information borrowing" among subjects. This model assumes that in the same brain region and under the same stimulus, the hemodynamic responses across subjects share a common but unknown functional shape that can differ in magnitude, latency and degree of interaction. We develop a computationally-efficient strategy based on splines to estimate the model parameters, and a hypothesis test on nonlinearity. The proposed method is compared with several existing methods via extensive simulations, and is applied to a real event-related fMRI study.

Increased default mode network activity in socially anxious individuals during reward processing

Background

Social anxiety has been associated with potentiated negative affect and, more recently, with diminished positive affect. It is unclear how these alterations in negative and positive affect are represented neurally in socially anxious individuals and, further, whether they generalize to non-social stimuli. To explore this, we used a monetary incentive paradigm to explore the association between social anxiety and both the anticipation and consumption of non-social incentives. Eighty-four individuals from a longitudinal community sample underwent functional magnetic resonance imaging (fMRI) while participating in a monetary incentive delay (MID) task. The MID task consisted of alternating cues indicating the potential to win or prevent losing varying amounts of money based on the speed of the participant’s response. We examined whether self-reported levels of social anxiety, averaged across approximately 7 years of data, moderated brain activity when contrasting gain or loss cues with neutral cues during the anticipation and outcome phases of incentive processing. Whole brain analyses and analyses restricted to the ventral striatum for the anticipation phase and the medial prefrontal cortex for the outcome phase were conducted.

Results

Social anxiety did not associate with differences in hit rates or reaction times when responding to cues. Further, socially anxious individuals did not exhibit decreased ventral striatum activity during anticipation of gains or decreased MPFC activity during the outcome of gain trials, contrary to expectations based on literature indicating blunted positive affect in social anxiety. Instead, social anxiety showed positive associations with extensive regions implicated in default mode network activity (for example, precuneus, posterior cingulate cortex, and parietal lobe) during anticipation and receipt of monetary gain. Social anxiety was further linked with decreased activity in the ventral striatum during anticipation of monetary loss.

Conclusions

Socially anxious individuals may increase default mode network activity during reward processing, suggesting high self-focused attention even in relation to potentially rewarding stimuli lacking explicit social connotations. Additionally, social anxiety may relate to decreased ventral striatum reactivity when anticipating potential losses.

Rationale and preclinical efficacy of a novel anti-EMP2 antibody for the treatment of invasive breast cancer

Despite significant advances in biology and medicine, the incidence and mortality due to breast cancer worldwide is still unacceptably high. Thus, there is an urgent need to discover new molecular targets. In this article, we show evidence for a novel target in human breast cancer, the tetraspan protein epithelial membrane protein-2 (EMP2). Using tissue tumor arrays, protein expression of EMP2 was measured and found to be minimal in normal mammary tissue, but it was upregulated in 63% of invasive breast cancer tumors and in 73% of triple-negative tumors tested. To test the hypothesis that EMP2 may be a suitable target for therapy, we constructed a fully human immunoglobulin G1 (IgG1) antibody specific for a conserved domain of human and murine EMP2. Treatment of breast cancer cells with the anti-EMP2 IgG1 significantly inhibited EMP2-mediated signaling, blocked FAK/Src signaling, inhibited invasion, and promoted apoptosis in vitro. In both human xenograft and syngeneic metastatic tumor monotherapy models, anti-EMP2 IgG1 retarded tumor growth without detectable systemic toxicity. This antitumor effect was, in part, attributable to a potent antibody-dependent cell-mediated cytotoxicity response as well as direct cytotoxicity induced by the monoclonal antibody. Together, these results identify EMP2 as a novel therapeutic target for invasive breast cancer.

The social regulation of threat-related attentional disengagement in highly anxious individuals

Social support may normalize stress reactivity among highly anxious individuals, yet little research has examined anxious reactions in social contexts. We examined the role of both state and trait anxiety in the link between social support and the neural response to threat. We employed an fMRI paradigm in which participants faced the threat of electric shock under three conditions: alone, holding a stranger's hand, and holding a friend's hand. We found significant interactions between trait anxiety and threat condition in regions including the hypothalamus, putamen, precentral gyrus, and precuneus. Analyses revealed that highly trait anxious individuals were less active in each of these brain regions while alone in the scanner-a pattern that suggests the attentional disengagement associated with the perception of high intensity threats. These findings support past research suggesting that individuals high in anxiety tend to have elevated neural responses to mild or moderate threats but paradoxically lower responses to high intensity threats, suggesting a curvilinear relationship between anxiety and threat responding. We hypothesized that for highly anxious individuals, shock cues would be perceived as highly threatening while alone in the scanner, possibly due to attentional disengagement, but this perception would be mitigated if they were holding someone's hand. The disengagement seen in highly anxious people under conditions of high perceived threat may thus be alleviated by social proximity. These results suggest a role for social support in regulating emotional responses in anxious individuals, which may aid in treatment outcomes.

Use-dependent dendritic regrowth is limited after unilateral controlled cortical impact to the forelimb sensorimotor cortex

Compensatory neural plasticity occurs in both hemispheres following unilateral cortical damage incurred by seizures, stroke, and focal lesions. Plasticity is thought to play a role in recovery of function, and is important for the utility of rehabilitation strategies. Such effects have not been well described in models of traumatic brain injury (TBI). We examined changes in immunoreactivity for neural structural and plasticity-relevant proteins in the area surrounding a controlled cortical impact (CCI) to the forelimb sensorimotor cortex (FL-SMC), and in the contralateral homotopic cortex over time (3-28 days). CCI resulted in considerable motor deficits in the forelimb contralateral to injury, and increased reliance on the ipsilateral forelimb. The density of dendritic processes, visualized with immunostaining for microtubule-associated protein-2 (MAP-2), were bilaterally decreased at all time points. Synaptophysin (SYN) immunoreactivity increased transiently in the injured hemisphere, but this reflected an atypical labeling pattern, and it was unchanged in the contralateral hemisphere compared to uninjured controls. The lack of compensatory neuronal structural plasticity in the contralateral homotopic cortex, despite behavioral asymmetries, is in contrast to previous findings in stroke models. In the cortex surrounding the injury (but not the contralateral cortex), decreases in dendrites were accompanied by neurodegeneration, as indicated by Fluoro-Jade B (FJB) staining, and increased expression of the growth-inhibitory protein Nogo-A. These studies indicate that, following unilateral CCI, the cortex undergoes neuronal structural degradation in both hemispheres out to 28 days post-injury, which may be indicative of compromised compensatory plasticity. This is likely to be an important consideration in designing therapeutic strategies aimed at enhancing plasticity following TBI.

Expression levels of estrogen receptor beta in conjunction with aromatase predict survival in non-small cell lung cancer

Estrogen signaling pathways may play a significant role in the pathogenesis of non-small cell lung cancers (NSCLC) as evidenced by the expression of aromatase and estrogen receptors (ERα and ERβ) in many of these tumors. Here we examine whether ERα and ERβ levels in conjunction with aromatase define patient groups with respect to survival outcomes and possible treatment regimens. Immunohistochemistry was performed on a high-density tissue microarray with resulting data and clinical information available for 377 patients. Patients were subdivided by gender, age and tumor histology, and survival data was determined using the Cox proportional hazards model and Kaplan-Meier curves. Neither ERα nor ERβ alone was predictor of survival in NSCLC. However, when coupled with aromatase expression, higher ERβ levels predicted worse survival in patients whose tumors expressed higher levels of aromatase. Although this finding was present in patients of both genders, it was especially pronounced in women ≥ 65 years old, where higher expression of both ERβ and aromatase indicated a markedly worse survival rate than that determined by aromatase alone. Expression of ERβ together with aromatase has predictive value for survival in different gender and age subgroups of NSCLC patients. This predictive value is stronger than each individual marker alone. Our results suggest treatment with aromatase inhibitors alone or combined with estrogen receptor modulators may be of benefit in some subpopulations of these patients.

Progesterone and estrogen receptor expression and activity in human non-small cell lung cancer

Lung cancer is the most common cause of cancer mortality in male and female patients in the US. Although it is clear that tobacco smoking is a major cause of lung cancer, about half of all women with lung cancer worldwide are never-smokers. Despite a declining smoking population, the incidence of non-small cell lung cancer (NSCLC), the predominant form of lung cancer, has reached epidemic proportions particularly in women. Emerging data suggest that factors other than tobacco, namely endogenous and exogenous female sex hormones, have a role in stimulating NSCLC progression. Aromatase, a key enzyme for estrogen biosynthesis, is expressed in NSCLC. Clinical data show that women with high levels of tumor aromatase (and high intratumoral estrogen) have worse survival than those with low aromatase. The present and previous studies also reveal significant expression and activity of estrogen receptors (ERα, ERβ) in both extranuclear and nuclear sites in most NSCLC. We now report further on the expression of progesterone receptor (PR) transcripts and protein in NSCLC. PR transcripts were significantly lower in cancerous as compared to non-malignant tissue. Using immunohistochemistry, expression of PR was observed in the nucleus and/or extranuclear compartments in the majority of human tumor specimens examined. Combinations of estrogen and progestins administered in vitro cooperate in promoting tumor secretion of vascular endothelial growth factor and, consequently, support tumor-associated angiogenesis. Further, dual treatment with estradiol and progestin increased the numbers of putative tumor stem/progenitor cells. Thus, ER- and/or PR-targeted therapies may offer new approaches to manage NSCLC.

Expression of phosphorylated raf kinase inhibitor protein (pRKIP) is a predictor of lung cancer survival

Background

Raf-1 kinase inhibitor protein (RKIP) has been reported to negatively regulate signal kinases of major survival pathways. RKIP activity is modulated in part by phosphorylation on Serine 153 by protein kinase C, which leads to dissociation of RKIP from Raf-1. RKIP expression is low in many human cancers and represents an indicator of poor prognosis and/or induction of metastasis. The prognostic power has typically been based on total RKIP expression and has not considered the significance of phospho-RKIP.

Methods

The present study examined the expression levels of both RKIP and phospho-RKIP in human lung cancer tissue microarray proteomics technology.

Results

Total RKIP and phospho-RKIP expression levels were similar in normal and cancerous tissues. phospho-RKIP levels slightly decreased in metastatic lesions. However, the expression levels of phospho-RKIP, in contrast to total RKIP, displayed significant predictive power for outcome with normal expression of phospho-RKIP predicting a more favorable survival compared to lower levels (P = 0.0118); this was even more pronounced in more senior individuals and in those with early stage lung cancer.

Conclusions

This study examines for the first time, the expression profile of RKIP and phospho-RKIP in lung cancer. Significantly, we found that phospho-RKIP was a predictive indicator of survival.

Protein expression based multimarker analysis of breast cancer samples

Background

Tissue microarray (TMA) data are commonly used to validate the prognostic accuracy of tumor markers. For example, breast cancer TMA data have led to the identification of several promising prognostic markers of survival time. Several studies have shown that TMA data can also be used to cluster patients into clinically distinct groups. Here we use breast cancer TMA data to cluster patients into distinct prognostic groups.

Methods

We apply weighted correlation network analysis (WGCNA) to TMA data consisting of 26 putative tumor biomarkers measured on 82 breast cancer patients. Based on this analysis we identify three groups of patients with low (5.4%), moderate (22%) and high (50%) mortality rates, respectively. We then develop a simple threshold rule using a subset of three markers (p53, Na-KATPase-β1, and TGF β receptor II) that can approximately define these mortality groups. We compare the results of this correlation network analysis with results from a standard Cox regression analysis.

Results

We find that the rule-based grouping variable (referred to as WGCNA*) is an independent predictor of survival time. While WGCNA* is based on protein measurements (TMA data), it validated in two independent Affymetrix microarray gene expression data (which measure mRNA abundance). We find that the WGCNA patient groups differed by 35% from mortality groups defined by a more conventional stepwise Cox regression analysis approach.

Conclusions

We show that correlation network methods, which are primarily used to analyze the relationships between gene products, are also useful for analyzing the relationships between patients and for defining distinct patient groups based on TMA data. We identify a rule based on three tumor markers for predicting breast cancer survival outcomes.

Abstract 903: Expression of estrogen receptor beta together with aromatase predicts survival in non-small cell lung cancer

Lung cancer continues to be the leading cause of cancer mortality. There is increasing evidence suggesting estrogens and estrogen signaling play a significant role in lung cancer pathophysiology. Estrogen receptors (ER) alpha and beta are expressed in normal lung and many non-small cell lung cancers. Previously we found that in women 65 years and older with non-small cell lung cancer (NSCLC), higher aromatase levels in tumor cells conferred a worse prognosis for survival. Here we examine whether ER alpha and ER beta levels in conjunction with aromatase can further define different patient groups with respect to survival outcomes and possible treatment regimens. Methods: Immunohistochemistry was performed on a high-density tissue microarray with marker data and clinical information available for 378 patients. Patients were subcategorized by gender, age, smoking status and tumor histology. Survival data was determined using the Cox proportional hazards model and Kaplan-Meier curves. Results: Unlike aromatase, neither ER alpha nor ER beta alone were predictors of survival in NSCLC patients or any subset of these patients examined. However, when coupled with aromatase expression, higher ER beta levels predicted poorer survival in those patients whose tumors expressed higher levels of aromatase. Although this survival difference was present in patients of both genders, it was more pronounced in women > 65 years old, where higher expression of both ER beta and aromatase showed a markedly worse survival rate than that determined by aromatase alone. Conclusion: Combined expression of ER beta with aromatase has predictive value for survival in different gender and age subgroups of NSCLC patients. This predictive value is stronger than each individual marker alone. These results suggest that possibly treatment with aromatase inhibitors alone or combined with estrogen receptor modulators may have benefit in some subpopulations of NSCLC patients.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 903. doi:10.1158/1538-7445.AM2011-903

Higher levels of GATA3 predict better survival in women with breast cancer

The GATA family members are zinc finger transcription factors involved in cell differentiation and proliferation. GATA3 in particular is necessary for mammary gland maturation, and its loss has been implicated in breast cancer development. Our goal was to validate the ability of GATA3 expression to predict survival in breast cancer patients. Protein expression of GATA3 was analyzed on a high-density tissue microarray consisting of 242 cases of breast cancer. We associated GATA3 expression with patient outcomes and clinicopathologic variables. Expression of GATA3 was significantly increased in breast cancer, in situ lesions, and hyperplastic tissue compared with normal breast tissue. GATA3 expression decreased with increasing tumor grade. Low GATA3 expression was a significant predictor of disease-related death in all patients, as well as in subgroups of estrogen receptor-positive or low-grade patients. In addition, low GATA3 expression correlated with increased tumor size and estrogen and progesterone receptor negativity. GATA3 is an important predictor of disease outcome in breast cancer patients. This finding has been validated in a diverse set of populations. Thus, GATA3 expression has utility as a prognostic indicator in breast cancer.

Differential expression of anterior gradient gene AGR2 in prostate cancer

Background

The protein AGR2 is a putative member of the protein disulfide isomerase family and was first identified as a homolog of the Xenopus laevis gene XAG-2. AGR2 has been implicated in a number of human cancers. In particular, AGR2 has previously been found to be one of several genes that encode secreted proteins showing increased expression in prostate cancer cells compared to normal prostatic epithelium.

Methods

Gene expression levels of AGR2 were examined in prostate cancer cells by microarray analysis. We further examined the relationship of AGR2 protein expression to histopathology and prostate cancer outcome on a population basis using tissue microarray technology.

Results

At the RNA and protein level, there was an increase in AGR2 expression in adenocarcinoma of the prostate compared to morphologically normal prostatic glandular epithelium. Using a tissue microarray, this enhanced AGR2 expression was seen as early as premalignant PIN lesions. Interestingly, within adenocarcinoma samples, there was a slight trend toward lower levels of AGR2 with increasing Gleason score. Consistent with this, relatively lower levels of AGR2 were highly predictive of disease recurrence in patients who had originally presented with high-stage primary prostate cancer (P = 0.009).

Conclusions

We have shown for the first time that despite an increase in AGR2 expression in prostate cancer compared to non-malignant cells, relatively lower levels of AGR2 are highly predictive of disease recurrence following radical prostatectomy.

Epithelial membrane protein-2 is a novel therapeutic target in ovarian cancer

PURPOSE

The tetraspan protein epithelial membrane protein-2 (EMP2) has been shown to regulate the surface display and signaling from select integrin pairs, and it was recently identified as a prognostic biomarker in human endometrial cancer. In this study, we assessed the role of EMP2 in human ovarian cancer.

EXPERIMENTAL DESIGN

We examined the expression of EMP2 within a population of women with ovarian cancer using tissue microarray assay technology. We evaluated the efficacy of EMP2-directed antibody therapy using a fully human recombinant bivalent antibody fragment (diabody) in vitro and ovarian cancer xenograft models in vivo.

RESULTS

EMP2 was found to be highly expressed in >70% of serous and endometrioid ovarian tumors compared with nonmalignant ovarian epithelium using a human ovarian cancer tissue microarray. Using anti-EMP2 diabody, we evaluated the in vitro response of nine human ovarian cancer cell lines with detectable EMP2 expression. Treatment of human ovarian cancer cell lines with anti-EMP2 diabodies induced cell death and retarded cell growth, and these response rates correlated with cellular EMP2 expression. We next assessed the effects of anti-EMP2 diabodies in mice bearing xenografts from the ovarian endometrioid carcinoma cell line OVCAR5. Anti-EMP2 diabodies significantly suppressed tumor growth and induced cell death in OVCAR5 xenografts.

CONCLUSIONS

These findings indicate that EMP2 is expressed in the majority of ovarian tumors and may be a feasible target in vivo.

Presence of a putative tumor-initiating progenitor cell population predicts poor prognosis in smokers with non-small cell lung cancer

Smoking is the most important known risk factor for the development of lung cancer. Tobacco exposure results in chronic inflammation, tissue injury, and repair. A recent hypothesis argues for a stem/progenitor cell involved in airway epithelial repair that may be a tumor-initiating cell in lung cancer and which may be associated with recurrence and metastasis. We used immunostaining, quantitative real-time PCR, Western blots, and lung cancer tissue microarrays to identify subpopulations of airway epithelial stem/progenitor cells under steady-state conditions, normal repair, aberrant repair with premalignant lesions and lung cancer, and their correlation with injury and prognosis. We identified a population of keratin 14 (K14)-expressing progenitor epithelial cells that was involved in repair after injury. Dysregulated repair resulted in the persistence of K14+ cells in the airway epithelium in potentially premalignant lesions. The presence of K14+ progenitor airway epithelial cells in NSCLC predicted a poor prognosis, and this predictive value was strongest in smokers, in which it also correlated with metastasis. This suggests that reparative K14+ progenitor cells may be tumor-initiating cells in this subgroup of smokers with NSCLC.