Persistent tracheostomy after primary chemoradiation for advanced laryngeal or hypopharyngeal cancer

Transcription

1 ORIGINAL ARTICLE Persistent tracheostomy after primary chemoradiation for advanced laryngeal or hypopharyngeal cancer Paul A. Tennant, MD, * Elizabeth Cash, PhD, Jeffrey M. Bumpous, MD, Kevin L. Potts, MD Division of Otolaryngology, Department of Surgery, James Graham Brown Cancer Center, University of Louisville, Louisville, Kentucky. Accepted 10 September 2013 Published online 18 December 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI /hed ABSTRACT: Background. Despite the demonstrated survival equivalence between chemoradiation and ablative surgery as primary treatment for advanced laryngeal and hypopharyngeal cancers, a subset of patients who undergo organ-preservation therapy have persistent tracheostomy requirement after completion of treatment. Methods. Patients who received primary chemoradiation for advanced laryngeal or hypopharyngeal cancer in a 3-year interval were identified. Rate of persistent posttreatment tracheostomy requirement was evaluated. The 12-month overall mortality rate was compared between patients who did and did not receive a tracheostomy before treatment. Results. In 60 patients identified for this study, T3/T4 status and hemilarynx fixation at the time of presentation were associated with persistent tracheostomy requirement 6 and 12 months posttreatment (p 5.022; p <.001; and p 5.032; p , respectively). Twelvemonth mortality was higher in T3/T4 patients who received pretreatment tracheostomy (p 5.034). Conclusion. Patients with advanced laryngeal or hypopharyngeal cancer who require tracheostomy before treatment have low rates of decannulation and higher short-term mortality than those who do not require tracheostomy before organ-preservation therapy. VC 2013 Wiley Periodicals, Inc. Head Neck 36: , 2014 KEY WORDS: advanced laryngeal and hypopharyngeal cancer, chemoradiation, laryngeal organ preservation, tracheostomy, persistent tracheostomy requirement INTRODUCTION Advanced laryngeal and hypopharyngeal cancers have challenged oncologists because of the functional implications of the disease and therapies traditionally used in the treatment for these malignancies. The ideal treatment regimen for this head and neck cancer subset not only achieves high rates of survival but also restores speech, swallowing, and respiration to the premorbid state. The Veterans Affairs Laryngeal Cancer Study Group 1 findings published in 1991 heralded a shift in emphasis toward organ preservation via chemoradiation for treatment of advanced laryngeal cancer. More contemporary studies have corroborated the survival equivalence between primary surgical extirpation with adjuvant radiotherapy and/ or chemotherapy and concurrent chemoradiation with surgery reserved for the salvage situation. 2 4 Unfortunately, the morbidity of chemoradiation for treatment of advanced laryngeal and hypopharyngeal cancer has sometimes undermined the intent of these organ-sparing protocols. Although a majority of patients *Corresponding author: P. A. Tennant, University of Louisville, Division of Otolaryngology, James Graham Brown Cancer Center, 3rd Floor, 529 South Jackson Street, Louisville, KY paulalberttennant@gmail.com This work was presented at the Combined Otolaryngology Spring Meetings, San Diego, California, April 18 22, achieve locoregional control with primary chemoradiation, 5,6 many experience significant long-term toxicities. 7 9 When primary chemoradiation regimens achieve oncologic cure but result in poor functional outcome, quality of life is adversely affected. 10 The frequent presence of late complications from organ-preserving treatment emphasizes the need for rigorous investigation to establish which pretreatment factors are associated with poor posttreatment laryngeal function. If predictive factors indicating a poor prospect for functional organ preservation even in the setting of oncologic cure can be determined, patients can be counseled in a prospective manner. Of the potential negative sequelae associated with primary chemoradiation treatment, speech and swallowing dysfunction have been the most extensively studied. Although these parameters obviously remain important considerations in assessing adequacy of laryngeal function, the persistent need for tracheostomy also represents a treatment complication from which patients experience significant morbidity. The presence of longterm tracheostomy has important financial, speech, and swallowing implications, which adversely affect quality of life. 11 Given the substantial morbidity associated with longterm tracheostomy in patients who undergo primary chemoradiation therapy for treatment of advanced laryngeal and hypopharyngeal cancers with organ-preserving intent, persistent tracheostomy requirement is viewed as a 1628 HEAD & NECK DOI /HED NOVEMBER 2014

2 PERSISTENT TRACHEOSTOMY AFTER PRIMARY CHEMORADIATION TABLE 1. Patient sample characteristics. Characteristic Frequency Percentage Mean (SD) Range Age, y (10.34) Sex Male Female Smoking status, (30.08) pack-years Primary site Supraglottis Glottis Hypopharynx Stage III IVA IVB suboptimal outcome. This analysis was performed to identify patient-related and disease-related factors that predict persistent tracheostomy requirement in patients who otherwise remain clinically disease-free. It is anticipated that an enhanced comprehension of these variables will further refine the selection criteria for organ preservation strategies and consequently reduce posttreatment morbidity. This study seeks to clarify the impact on overall survival for patients who remain tracheostomy dependent after laryngeal preservation therapy despite clinical evidence of oncologic cure. MATERIALS AND METHODS After approval was obtained from the University of Louisville Institutional Review Board, a clinical database was searched to identify all patients diagnosed with stage III, IVA, or IVB laryngeal or hypopharyngeal squamous cell carcinoma who were treated with definitive primary chemoradiation between 2006 and 2008 at this institution. All malignancies were histologically proven; no patient included in this study had received previous treatment for any diagnosis of malignancy. Each patient included in this study was presented to a multidisciplinary head and neck oncology tumor board, and individual treatment recommendations were prospectively developed in this setting. Excluded from this study were patients who had evidence of distant metastasis (M1; stage IVC) at the time of presentation or who were subsequently found to have biopsyproven recurrent or persistent malignant disease within 12 months of the primary chemoradiation completion date. Also excluded from consideration were patients who underwent surgical therapy for nonfunctional larynx or chondroradionecrosis within the 12-month posttreatment time frame in order to control for factors that would otherwise serve as independent indications for salvage total laryngectomy regardless of tracheostomy status. For patients who met criteria, demographic information was obtained, including age, sex, and smoking history. Initial staging was also documented according to the standard TNM classification based upon American Joint Committee on Cancer 2002 criteria. The site of primary tumor and mobility of the hemilarynx based on fiber-optic laryngoscopy was detailed for each patient as well as tracheostomy status at the 0-, 6-, and 12-month time intervals after completion of primary chemoradiation treatment. Sample characteristics are summarized in Table 1. Radiation therapy was performed solely at the University of Louisville James Graham Brown Cancer Center according to standardized protocols with intensitymodulated radiation therapy or 3D conformal radiation therapy to a dose of 70 to 72 Gray (Gy) to the primary site and clinically positive neck disease and 50 Gy to any clinically negative cervical lymph node basins at risk for regional metastasis. Chemotherapy regimens were platinum-based and delivered weekly during radiation therapy to standard area-under-the-curve dosing. Statistical analysis was performed to determine whether significant differences existed between patients who did and did not require tracheostomy before initiation of treatment. A 1-way analysis of variance was used to compare these groups with regard to age, sex, smoking history, stage, subsite, and vocal fold mobility at presentation. We wished to efficiently compare patients who presented with less primary site disease burden (T1 and T2) with those with more advanced disease (T3 or T4) at the primary site. Thus, patients with T1 or T2 primary tumors, and separately, patients with T3 or T4 malignancies were compared using Fisher s exact tests, chi-square tests, or analysis of variance models, where appropriate. Specific notation was made for any patient who received a tracheostomy before treatment but was later decannulated within the 12-month posttreatment time period; patients who were not tracheostomy dependent initially but subsequently required a tracheostomy for airway compromise were also documented. Utilizing the study sample, mortality rate at 12 months posttreatment was calculated for patients whose tracheostomy status was known at the 0-month and 12-month time intervals or at the time of death. Of this subset, patients with T3/T4 disease were further selected and compared in order to control for T classification at the time of presentation. Logistic regression models adjusted for age at presentation were used to estimate survival status. RESULTS Between 2006 and 2008, 60 patients (48 men) were identified who underwent primary chemoradiation for stage III or IV laryngeal or hypopharyngeal squamous cell carcinoma and remained clinically disease-free during the initial 12-month follow-up period. Demographic and TABLE 2. Site distribution and frequency of true vocal cord fixation among patients with advanced laryngeal and hypopharyngeal cancer. No. of patients T classification N0 N1 N2 N3 Total TVC fixation T T T T Abbreviation: TVC, true vocal cord. HEAD & NECK DOI /HED NOVEMBER

3 TENNANT ET AL. FIGURE 1. Diagram demonstrating the tracheostomy status of patients over the course of the study observation period. clinical information are presented in Tables 1 and 2. Progression of tracheostomy status over the follow-up period is presented in Figure 1. Tests confirmed there were no significant differences when comparing T1 and T2 patients, and when comparing T3 and T4 patients, on the variables of age, sex, smoking status, and subsite (all p values were >.247). Therefore, patients were grouped into low (T1 and T2) and high (T3 and T4) primary disease burden cohorts. Tests also demonstrated that there were no sex differences noted on any study variable. The overall rate of tracheostomy before primary chemoradiation was 34% (18 of 53 patients), and did not vary significantly by subsite. Fourteen of 41 men and 4 of 12 women required tracheostomy before treatment. Of patients with a supraglottic primary site, 37% required tracheostomy, whereas the rate of tracheostomy was 25% for glottic tumors and 36% for hypopharyngeal tumors. Seventeen percent of patients (2 of 12) with T1 or T2 primary site disease burden received pretreatment tracheostomy; in contrast, 39% of T3/T4 patients (16 of 41) had a tracheostomy before initiation of primary chemoradiation. Although the rate of tracheostomy before initiation of primary chemoradiation in patients with T3/T4 disease was not statistically different than in the T1/T2 cohort (p 5.154), T1/T2 patients were less likely to require tracheostomy at the 6-month and 12-month posttreatment intervals (40% vs 9%; p 5.022; and 37% vs 0%; p <.001, respectively; Figure 2). Two patients in the study who received a tracheostomy before treatment were decannulated within 6 months after primary chemoradiation (Figure 1). The same 2 patients remained decannulated at 12 months posttreatment. No other patients were decannulated between the 6-month and 12-month follow-up time intervals. Five patients who did not begin treatment with a tracheostomy received new tracheostomies for airway compromise in the first 12 months after primary chemoradiation (Table 3). Of the 4 patients with glottic primary tumors included in the study, 1 received a tracheostomy before treatment and the other 3 required tracheostomy placement within the first 6 months after completion of treatment. Among patients with higher T classification, the tracheostomy rate for patients did not vary as a function of neck nodal disease burden at any time point. Patients with T3/T4 disease and documented hemilarynx fixation FIGURE 2. Rate of tracheostomy in patients who presented with T1/T2 disease versus T3/T4 disease HEAD & NECK DOI /HED NOVEMBER 2014

4 PERSISTENT TRACHEOSTOMY AFTER PRIMARY CHEMORADIATION TABLE 3. Specific patient-related and disease-related variables for patients who did not begin treatment with a tracheostomy but subsequently received a new tracheostomy after completion of primary chemoradiation. Patient Age, y Sex Smoking history, pack-years Primary tumor site TNM classification TVC mobility status Timing of tracheostomy 1 46 Male 45 Right TVC T3N0M0 Immobile 4 mo s/p chemoxrt 2 59 Male 35 Right TVC T3N0M0 Immobile 6 mo s/p chemoxrt 3 49 Male Unknown Left pyriform sinus T4N2bM0 Immobile 3 wk s/p chemoxrt 4 54 Male 35 Left TVC T3N0M0 Immobile 5 mo s/p chemoxrt 5 58 Female 80 Right AE fold T3N0M0 Immobile 6 mo s/p chemoxrt Abbreviations: TVC, true vocal cord; s/p, status post; chemoxrt, chemoradiation; AE fold, aryepiglottic fold. at presentation (9 of 13 patients; 69%) were no more likely to receive a tracheostomy before treatment than T3/ T4 patients who retained true vocal fold mobility on the affected side (16 of 23 patients; 70%; p 5.097); however, patients who presented with advanced primary tumors and fixation of the involved hemilarynx had a higher rate of persistent tracheostomy requirement 12- months posttreatment (8 of 17 patients, 47% vs 1 of 8 patients, 13%; p ; Figure 3). Among all patients who underwent primary chemoradiation for advance-staged laryngeal or hypopharyngeal squamous cell carcinoma and did not have evidence of persistent or recurrent malignant disease during the initial 12-month follow-up time period, those who received a tracheostomy before initiation of treatment had a poorer subsequent 12-month overall survival than patients who did not require a tracheostomy before treatment. Patients who received a tracheostomy before treatment had a 65% 12-month overall survival, whereas patients who did not require tracheostomy had a 94% overall survival (p 5.003; Table 3; Figure 4). Among individuals who presented with T3 or T4 primary disease, 5 of 16 patients (31%) who initially required tracheostomy died within 12 months; in contrast, only 2 of 24 T3/T4 patients (8%) who did not receive a tracheotomy died within the 12- month follow-up time frame (p 5.032). This correlation weakened somewhat after adjusting for age at the time of presentation (p 5.072). Six of the 8 patients included in the study who died within the 12-month posttreatment time period remained tracheostomy dependent at the time of death. DISCUSSION Treatment of advanced laryngeal and hypopharyngeal cancer with combined chemotherapy and radiation therapy has gained widespread acceptance over the past 20 years. Although this regimen has demonstrated efficacy in preserving the larynx, a number of these patients have ultimately required salvage laryngectomy. In an effort to improve patient selection, different criteria have previously been evaluated for their prognostic significance on laryngeal function and survival. Tracheostomy requirement before treatment has been considered indicative of unfavorable disease; indeed, some institutions have historically viewed airway compromise requiring tracheostomy as an indication for primary laryngectomy. 12 Although the need for tracheostomy is acknowledged as an adverse prognostic factor, larynx-preserving strategies continue to be utilized in this population because of the confirmation of survival equivalence 13 and the powerful desire of patients to preserve their voice. 14 Speech and swallowing dysfunction is common in patients with advanced laryngeal and hypopharyngeal tumors. 9,15 Even among patients who remain clinically FIGURE 3. Rate of tracheostomy in T3/T4 patients who presented with a fixed hemilarynx versus T3/T4 patients who retained vocal fold mobility. FIGURE 4. Kaplan Meier survival curves representing patients who did versus did not receive a tracheostomy during the observation period of study. HEAD & NECK DOI /HED NOVEMBER

5 TENNANT ET AL. disease-free after primary chemoradiation for these malignancies, the morbidity of treatment can be significant. A meta-analysis of 230 patients who received primary chemoradiation treatment for head and neck cancers showed that a primary site in the larynx or hypopharynx was more strongly correlated with the presence of severe late toxicity in multivariate analysis than any other variable examined. 16 In another report, severe laryngopharyngeal dysfunction, defined as persistent need for gastrostomy or tracheostomy, was present in 36% of patients who underwent organ-preservation treatment for advanced laryngeal malignancies and remained disease-free at 6 months after completion of therapy. 17 Persistent tracheostomy requirement is also associated with poor functional outcome requiring salvage surgery. Hutcheson et al 18 recently studied the outcomes of patients who were rendered diseasefree with organ-sparing treatment but subsequently underwent elective total laryngectomy and found that 39% of these individuals were tracheostomy dependent before removal of their larynges. In our investigation of patients with stage III or IV laryngeal or hypopharyngeal squamous cell carcinoma, 34% received a tracheostomy before initiation of treatment. In a significant majority of patients, tracheostomy was performed for immediate or impending airway compromise. Among this cohort, patients who presented with T1 or T2 disease did not have a significantly lower tracheostomy rate at presentation than individuals with higher T classification; however, no patient with T1/T2 disease was tracheostomy dependent at the 12-month follow-up time interval. The opportunity for decannulation in patients with T1/T2 disease lends further support to the American Society of Clinical Oncology practice guidelines that advocate a larynx-preserving approach in virtually all cases of T1 or T2 laryngeal cancer. 19 Five patients who did not initially require tracheostomy subsequently received this intervention within the first year after treatment for airway compromise. Each of these patients had a fixed hemilarynx before treatment, likely indicating paraglottic space involvement. Three of these patients who required tracheostomy after completion of treatment had glottic primary tumors, and each of the 4 patients with advanced glottic laryngeal cancer included in the study were tracheostomy dependent at 12-month follow-up. It is postulated that patients with advanced glottic primary tumors are at particular risk of airway compromise because of radiation-induced laryngeal edema and consequent restriction of the glottic aperture; previous studies have proposed a similar phenomenon. 13 Although the rate of persistent tracheostomy requirement at 12 months posttreatment in patients with glottic primary tumors was significantly greater than in other laryngeal or hypopharyngeal subsites, this result must be interpreted with caution given the low number of glottic tumors included in the study. Of the 44 T3/T4 patients who met inclusion criteria, the regional metastatic disease burden did not affect initial tracheostomy requirement or persistent need for tracheostomy after treatment. This result agrees with findings from Staton et al 17 in their assessment of factors predictive of poor functional outcome in this patient population. Although T3/T4 patients with a mobile hemilarynx were equally likely to require initial tracheostomy as those with a fixed true vocal fold, their rate of tracheostomy dependence at 6-month and 12-month follow-up was significantly lower, again correlating well with a prior study. Hemilarynx fixation, considered predominantly a function of paraglottic space involvement, seems to be an important prognostic variable in determining which patients are ultimately suitable for decannulation after organ-preserving therapy. With regard to overall survival, it has been our perception that patients who require tracheostomy at the time of presentation and subsequently undergo primary chemoradiation have fared poorer in terms of survival at our institution. The results of this study corroborate this suspicion: during the initial 12-month posttreatment interval, patients who required a tracheostomy before treatment of their malignancy were more likely to die during this time period, even as they remained clinically diseasefree. In the smaller subgroup of T3/T4 tumors, this negative outcome remained significant, although the correlation weakened after adjusting for age at time of presentation; nevertheless, similar patterns in the data indicate that this finding still may have clinical relevance. Although the precise cause of mortality is speculative in this cohort because of the retrospective nature of this study, the decline in overall survival associated with persistent tracheostomy requirement could suggest a more significant initial disease burden and resultant morbidity of treatment that is not accurately predicted by current staging systems. The impact of tracheostomy upon disease-specific survival is unclear. Current evidence varies between Mendenhall et al 12 who reported that tracheostomy was an adverse prognostic factor for cause-specific survival in multivariate analysis of T3 glottic tumors treated with primary radiotherapy and MacKenzie et al 13 who do not distinguish tracheostomy as a predictor of locoregional control or cause-specific survival in a similar cohort. Our data suggest that pretreatment tracheostomy predisposes patients to increased mortality risk in the absence of biopsy-proven recurrent or persistent disease. This study had several limitations. Because it was conducted in retrospective fashion at a single institution, insufficient documentation was sometimes available regarding specific indications for initial or persistent tracheostomy requirement, introducing recall bias. The precise volume of initial primary disease was unknown in many cases, and TNM classification was used as a surrogate measure. Swallowing function and gastrostomy tube status were also not formally assessed, as they were not the focus of the study outcome. The 12-month follow-up period was possibly too brief to capture all late toxicities, and an analysis of medical comorbidities was not performed. Perhaps most significantly, although the patients included in this study received regular posttreatment surveillance, autopsies were not performed on any of the patients who died during the follow-up time period, and it is therefore impossible to know whether any of these individuals had histopathologic evidence of recurrent or persistent malignant disease at the time of death. Despite these shortcomings, it is clear that tracheostomy requirement before treatment generally implies a more 1632 HEAD & NECK DOI /HED NOVEMBER 2014

6 PERSISTENT TRACHEOSTOMY AFTER PRIMARY CHEMORADIATION complicated course after primary chemoradiation; it should therefore factor into clinical decision-making by the multidisciplinary teams caring for these patients. CONCLUSION Persistent tracheostomy requirement after primary chemoradiation is not accurately predicted by current staging mechanisms for advanced laryngeal and hypopharyngeal squamous cell carcinoma. Patients with T3/T4 disease and hemilarynx fixation who require pretreatment tracheostomy may represent poor candidates for organ-sparing treatment. Persistent tracheostomy dependence after primary chemoradiation is a significant morbidity and confers increased 12-month mortality risk to patients who have been treated for advanced laryngeal or hypopharyngeal cancer and otherwise remain clinically disease free. Further investigation is warranted to elucidate patient and disease factors that predict persistent need for tracheostomy, as this subset of patients may benefit from primary surgical treatment. REFERENCES 1. The Department of Veterans Affairs Laryngeal Cancer Study Group. Induction chemotherapy plus radiation compared with surgery plus radiation in patients with advanced laryngeal cancer. The Department of Veterans Affairs Laryngeal Cancer Study Group. N Engl J Med 1991;324: Lefebvre JL, Chevalier D, Luboinski B, Kirkpatrick A, Collette L, Sahmoud T. Larynx preservation in pyriform sinus cancer: preliminary results of a European Organization for Research and Treatment of Cancer phase III trial. EORTC Head and Neck Cancer Cooperative Group. J Natl Cancer Inst 1996;88: Weber RS, Berkey BA, Forastiere A, et al. Outcome of salvage total laryngectomy following organ preservation therapy: the Radiation Therapy Oncology Group trial Arch Otolaryngol Head Neck Surg 2003;129: Forastiere AA, Goepfert H, Maor M, et al. Concurrent chemotherapy and radiotherapy for organ preservation in advanced laryngeal cancer. N Engl J Med 2003;349: Cmelak AJ, Li S, Goldwasser MA, et al. Phase II trial of chemoradiation for organ preservation in resectable stage III or IV squamous cell carcinomas of the larynx or oropharynx: results of Eastern Cooperative Oncology Group Study E2399. J Clin Oncol 2007;25: Guadagnolo BA, Haddad RI, Posner MR, et al. Organ preservation and treatment toxicity with induction chemotherapy followed by radiation therapy or chemoradiation for advanced laryngeal cancer. Am J Clin Oncol 2005;28: Woodson GE, Rosen CA, Murry T, et al. Assessing vocal function after chemoradiation for advanced laryngeal carcinoma. Arch Otolaryngol Head Neck Surg 1996;122: Lazarus CL, Logemann JA, Pauloski BR, et al. Swallowing disorders in head and neck cancer patients treated with radiotherapy and adjuvant chemotherapy. Laryngoscope 1996;106(9 Pt 1): Carrara de Angelis E, Feher O, Barros AP, Nishimoto IN, Kowalski LP. Voice and swallowing in patients enrolled in a larynx preservation trial. Arch Otolaryngol Head Neck Surg 2003;129: Terrell JE, Fisher SG, Wolf GT. Long-term quality of life after treatment of laryngeal cancer. The Veterans Affairs Laryngeal Cancer Study Group. Arch Otolaryngol Head Neck Surg 1998;124: Terrell JE, Ronis DL, Fowler KE, et al. Clinical predictors of quality of life in patients with head and neck cancer. Arch Otolaryngol Head Neck Surg 2004;130: Mendenhall WM, Parsons JT, Mancuso AA, Pameijer FJ, Stringer SP, Cassisi NJ. Definitive radiotherapy for T3 squamous cell carcinoma of the glottic larynx. J Clin Oncol 1997;15: MacKenzie R, Franssen E, Balogh J, Birt D, Gilbert R. The prognostic significance of tracheostomy in carcinoma of the larynx treated with radiotherapy and surgery for salvage. Int J Radiat Oncol Biol Phys 1998;41: McNeil BJ, Weichselbaum R, Pauker SG. Speech and survival: tradeoffs between quality and quantity of life in laryngeal cancer. N Engl J Med 1981;305: Starmer H, Gourin C, Lua LL, Burkhead L. Pretreatment swallowing assessment in head and neck cancer patients. Laryngoscope 2011;121: Machtay M, Moughan J, Trotti A, et al. Factors associated with severe late toxicity after concurrent chemoradiation for locally advanced head and neck cancer: an RTOG analysis. J Clin Oncol 2008;26: Staton J, Robbins KT, Newman L, Samant S, Sebelik M, Vieira F. Factors predictive of poor functional outcome after chemoradiation for advanced laryngeal cancer. Otolaryngol Head Neck Surg 2002;127: Hutcheson KA, Alvarez CP, Barringer DA, Kupferman ME, Lapine PR, Lewin JS. Outcomes of elective total laryngectomy for laryngopharyngeal dysfunction in disease-free head and neck cancer survivors. Otolaryngol Head Neck Surg. 2012;146: American Society of Clinical Oncology, Pfister DG, Laurie SA, et al. American Society of Clinical Oncology clinical practice guideline for the use of larynx-preservation strategies in the treatment of laryngeal cancer. J Clin Oncol 2006;24: HEAD & NECK DOI /HED NOVEMBER