Self-reported sleep disturbance and survival in myelodysplastic syndromes

Transcription

1 short report Self-reported sleep disturbance and survival in myelodysplastic syndromes Marlise R. Luskin, 1,2 Angel M. Cronin, 1 Robert L. Owens, 3 Daniel J. DeAngelo, 1,2 Richard M. Stone, 1,2 Martha Wadleigh, 1,2 David P. Steensma 1,2 and Gregory A. Abel 1,2 1 Dana-Farber Cancer Institute, 2 Harvard Medical School, Boston, MA, and 3 University of California, San Diego, San Diego, CA, USA Received 18 October 2016; accepted for publication 6 December 2016 Correspondence: Marlise R. Luskin, Dana- Farber Cancer Institute, 450 Brookline Avenue, Dana 2056, Boston, MA 02215, USA. marlise_luskin@dfci.harvard.edu Summary Neither the prevalence of sleep disturbance nor its association with fatigue and overall survival (OS) are well understood for patients with myelodysplastic syndromes (MDS). New patients at our institution (n = 251; ) completed the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire, which includes questions about sleep and fatigue. Fifty-three per cent reported at least a little trouble sleeping. In multivariable models, anaemia and sleep disturbance were associated with fatigue (both P < 0001). Additionally, in separate models, sleep disturbance (P = 0002) and fatigue (P = 004) both predicted OS. Our data suggest that improving sleep quality may impact MDS-related fatigue and OS. Keywords: myelodysplastic syndromes, sleep, fatigue, quality of life, overall survival. The myelodysplastic syndromes (MDS) are a heterogeneous group of haematopoietic stem cell neoplasms characterized by bone marrow failure and risk of progression to acute myeloid leukaemia. The life expectancy of patients with MDS is variably compromised, with disease characteristics at diagnosis cytopenias, bone marrow blast burden, and genetics driving standard MDS risk classification schemes for overall survival (OS) (Greenberg et al, 1997, 2012). Patients at all levels of disease risk report that MDS negatively impacts their quality of life (QOL), with fatigue being the most prominent complaint (Jansen et al, 2003; Steensma et al, 2008). Fatigue in MDS has been attributed to symptomatic anaemia in the setting of a toxic milieu of inflammatory cytokines (Jansen et al, 2003; Meyers et al, 2005). Unfortunately, neither the administration of disease-modifying drugs nor treatment of anaemia with erythropoiesis-stimulating agents and transfusions successfully improves QOL for all patients, suggesting that other factors may contribute to fatigue (Pinchon et al, 2009). Recent work has demonstrated that severity of patient-reported fatigue is a stronger predictor of compromised QOL than anaemia in high-risk MDS patients, providing further evidence that factors beyond anaemia contribute to fatigue and impaired QOL for this population (Efficace et al, 2015a). An additional finding that selfreported fatigue is a prognostic factor for OS beyond standard disease risk factors adds urgency to efforts to identify contributors to MDS-associated fatigue (Efficace et al, 2015b). Sleep disturbance is widely recognized to contribute to fatigue and impaired QOL in patients with other cancers (Palesh et al, 2010). Additionally, an association between sleep disturbance and shortened OS has been documented in population-based cohorts, as well as in patients with cardiovascular disease (Li et al, 2014; Lallukka et al, 2016). On the other hand, little is known about the potential contribution of sleep disturbance to compromised outcomes for blood cancers. We hypothesized that for patients with MDS, selfreported sleep disturbance a potential target of medical intervention (Garland et al, 2014) would be associated with fatigue even after controlling for anaemia, and would also be independently associated with survival. Study design Beginning in 2006, adult patients ( 18 years) with biopsyconfirmed MDS presenting for initial evaluation at the Dana-Farber Cancer Institute (DFCI) were eligible for enrolment into the Dana-Farber/Harvard Cancer Center (DF/ HCC) Leukemia Clinical Research Information System (CRIS) database; 85% of eligible patients consented to enrolment. Patients enrolled up to June 2014 were included in this analysis. The CRIS database includes clinical data from initial evaluation at DFCI, which was used to determine the original and revised International Prognostic Scoring System (IPSS and IPSS-R) score (Greenberg et al, 1997, 2012) for each patient. Cytogenetics were classified per the three-level IPSS designations. First published online 8 March 2017 doi: /bjh ª 2017 John Wiley & Sons Ltd

2 At baseline, enrolled patients completed the European Organization for Research and Treatment Cancer Quality of Life Questionnaire (QLQ-C30), a validated 30-item instrument measuring QOL in patients with cancer, including MDS (Aaronson et al, 1993). The EORTC QLQ-C30 questionnaire is a copyrighted instrument; a sample is available for academic download at: The QLQ-C30 includes a 3-item subscale for measuring fatigue in cancer (lower score is better) and a single item question, Have you had trouble sleeping?, that measures sleep disturbance. Patients with incomplete clinical data were excluded from analysis. We created multivariable models to characterize (i) the association of self-reported sleep disturbance with selfreported fatigue, controlling for patient/clinical covariates including anaemia (haemoglobin <100 g/l), and (ii) the association of self-reported sleep disturbance with OS, controlling for patient/clinical covariates (Supplementary Methods). Study methods were approved by the DF/HCC Office for Human Research Studies. Results and discussion Of 303 patients with MDS who completed the baseline survey, 251 patients had baseline clinical data sufficient for inclusion (Table I). For the QLQ-C30 item Have you had trouble sleeping? 53% of patients reported some level of sleep disturbance by selecting an answer other than not at all. IPSS cytogenetic risk category and haemoglobin <100 g/l were not significantly associated with sleep disturbance, but Table I. Self-reported sleep disturbance among MDS patients and unadjusted associations with baseline clinical characteristics (N = 251). Self-reported sleep disturbance Overall Not at all A little Quite a bit Very much P Number of patients n = 251 n = 119 n = 80 n = 35 n = 17 % of cohort Median age (IQR) at presentation, years 68 (18, 89) <60 years years >70 years Gender Male Female Modified Charlson comorbidity index < IPSS-R cytogenetic risk category Good/very good Intermediate Poor/very poor Haemoglobin* 100 g/l or higher <100 g/l Marrow blast proportion* <5% % % >20% IPSS-R risk group Very low Low Intermediate High Very high Values are expressed in percentage. IPSS-R, Revised International Prognostic Scoring System. *Corresponding to values closest to first presentation to Dana-Farber Cancer Institute; some patients had progressed to blasts >20 with the closest marrow (World Health Organization-defined acute myeloid leukaemia) but were originally evaluated for myelodysplastic syndrome. P-values determined by Fisher s exact test. ª 2017 John Wiley & Sons Ltd 563

3 higher comorbidity index was (P < 0001). Components of the comorbidity index independently associated with sleep disturbance include congestive heart failure, peripheral vascular disease, cerebrovascular disease, and renal disease (Table SI). The mean fatigue score was 354 [standard deviation (SD) 269; median = 333; interquartile range (IQR) = ]. Among clinical factors, self-reported sleep disturbance, comorbidity and haemoglobin were strongly associated with fatigue (all P < 0005, Table II), with the two former factors exhibiting a clear dose response relationship with increased fatigue (Spearman s r of 021 and 043, respectively). In multivariable models, sleep disturbance and haemoglobin <100 g/l were independently associated with increased fatigue (both P < 0001; Tables II, SII). Importantly, reporting a little sleep disturbance compared to not at all had a similar effect size with respect to increase in fatigue score (125 points) as having a haemoglobin <100 g/l compared to 100 g/l (144 points). Median follow-up was 29 months; 150 of 251 patients died. In separate multivariable models, both self-reported sleep disturbance (P = 0002) and fatigue (P = 004) were independently associated with OS. Presence of very much sleep disturbance, in particular, was associated with an increased hazard of death [Hazard ratio (HR) 295, 95% confidence interval (CI) ] while the HR for a SD (i.e. 27 point) increase in fatigue was 120 (95% CI ). Sleep disturbance and fatigue severity had similar discriminative enhancements compared to a model with standard covariates alone (IPSS-R, comorbidity index, gender), although these were modest (bootstrap-corrected c-statistic: 0701 for standard covariates, 0702 for standard covariates plus fatigue severity, 0704 for standard covariates plus sleep disturbance; Table SIII). Table II. Unadjusted and adjusted associations with EORTC-QLQ-C30 fatigue scale, by multivariable linear regression modeling. Unadjusted Fatigue symptom score, mean (SD) Mean difference 95% CI P value Adjusted Mean difference 95% CI P value Overall 354 (269) Age* (per 10 years) 04 34, , <60 years 359 (267) years 355 (255) >70 years 350 (286) Gender Male 339 (264) Reference 018 Reference 028 Female 387 (280) 49 23, , 98 Modified Charlson comorbidity index (248) Reference 0004 Reference (271) 82 05, , (281) , , 141 IPSS-R cytogenetic risk category Good/very good 356 (273) Reference 078 Reference 076 Intermediate 331 (222) , , 73 Poor/very poor 368 (294) 12 70, , 44 Haemoglobin 100 g/l or higher 264 (254) Reference <0001 Reference <0001 <100 g/l 427 (260) , , 203 Marrow blast proportion <5% 344 (268) Reference 094 Reference % 371 (245) 27 55, , % 353 (317) 09 85, , 45 >20% 354 (256) , , 50 Sleep disturbance Not at all 244 (220) Reference <0001 Reference <0001 A little 390 (226) , , 193 Quite a bit 463 (299) , , 285 Very much 732 (239) , , 587 SD, standard deviation; CI, confidence interval; IPSS-R, Revised International Prognostic Scoring System. *For descriptive statistics, the mean (SD) fatigue symptom score is summarized by age categorized as <60 years, years and >70 years. In the statistical analyses, age was entered as a continuous variable with the mean difference presented for each 10-year increase in age. 564 ª 2017 John Wiley & Sons Ltd

4 Self-reported sleep disturbance was common in this MDS population, present in the majority (53%) of patients surveyed. Notably, sleep disturbance was not significantly associated with disease risk or haemoglobin level, illustrating that this pervasive symptom is not limited to specific subsets of this heterogeneous disease. On the other hand, we found sleep disturbance to be more prevalent in those with more non-mds comorbidities. We demonstrate that sleep disturbance may be an important predictor of fatigue even when controlling for comorbidity, anaemia, and other measures of disease risk (blast burden, cytogenetics). We also found that sleep disturbance was independently associated with worse OS after controlling for standard prognostic variables and comorbidity. Although the casual mechanism by which sleep disturbance may impact OS in MDS cannot be determined from this study, the role of sleep in promoting tumour growth is being increasingly recognized (Owens et al, 2016). In animal models of intermittent hypoxaemia, cancer cells have shown accelerated growth. Even absent intermittent hypoxaemia, sleep fragmentation has been associated with tumour growth in the laboratory. Finally, MDS cell colonies demonstrate enhanced growth in hypoxic conditions, a phenomenon not seen in acute or chronic myeloid leukaemia colonies (Thompson et al, 2007). Our analysis is consistent with the findings of two recent European studies which demonstrated that in higher-risk MDS patients, patient-reported fatigue was a better predictor of worse QOL than other typically assessed measures (including performance status and haemoglobin), and had independent prognostic value for survival beyond standard MDS risk systems (Efficace et al, 2015a,b). Our study confirms that self-reported fatigue contributes prognostic value independent of the IPSS-R for patients of all levels of disease risk. Moreover, by identifying a strong association between sleep disturbance and fatigue, our data suggest that sleep disturbance is a potential target of intervention to improve MDS-related QOL and potentially lengthen OS. There are limitations to our analyses. First, we focus on patient-reported outcomes from a single centre. On the other hand, our methods include community-based patients with MDS presenting to tertiary care for the first time and a variety of risk groups. Second, we did not include a multi-item measure of sleep, instead relying on a single question on the QLQ-C30. We do note, however, that although the question regarding sleep disturbance and the 3-item fatigue assessment are both included in the QLQ-C30, they are analysed separately in our survival analyses, following methods used in prior MDS QOL analyses (Efficace et al, 2015a,b). Still, sleep disturbance was assessed by self-report, rather than directly via polysomnography or other methods that could more precisely characterize the nature of sleep disruption. Third, anaemia and disordered iron signalling may themselves be associated with confounding physiological derangements such as restless leg syndrome that lead to poor sleep quality. Finally, we defined anaemia as a haemoglobin of <100 g/l, although some patients experience anaemia-related fatigue at higher levels. The association we found between sleep disturbance and OS is novel for patients with haematological malignancies and deserves further attention. Awareness of the role of sleep disturbance in patients with cancer is increasing, and several investigations have suggested that management of sleep disturbance can improve QOL (Palesh et al, 2010). Correction of anatomical causes, such as obstructive sleep apnoea, may be particularly important, as this co-morbidity is common in older patients, particularly men (who are, in turn, more likely to develop MDS). Depending on the underlying cause of disordered sleep, pharmacological and behavioural interventions may be helpful, and formal sleep studies to characterize the nature of sleep disturbance in patients with MDS can focus treatment approaches. Moreover, our analysis suggests that further study of the contribution of sleep disturbance to QOL and OS outcomes in MDS is warranted, including assessment of specific interventions to ameliorate it. Acknowledgement The authors gratefully acknowledge funding from an American Cancer Society Research Scholar Grant RSG (G.A.A). Author contributions G.A.A. and A.M.C. designed the study. M.R.L., A.M.C., R.L.O., D.J.D., R.M.S., M.W., D.P.S. and G.A.A. analysed the data. M.R.L., A.M.C., R.L.O., D.J.D., R.M.S., M.W., D.P.S. and G.A.A. wrote the manuscript. D.J.D., R.M.S., M.W., D.P.S. and G.A.A. contributed data. All authors reviewed the manuscript. Disclosure of conflict of interest The authors declare no competing financial interests. Supporting Information Additional Supporting Information may be found in the online version of this article: Table SI. Self-reported sleep disturbance among MDS patients and unadjusted associations with baseline comorbidity (N = 251). Table SII. Adjusted associations with clinically relevant fatigue (EORTC-QLQ-C30 fatigue scale 40), by multivariable logistic regression modeling. Table SIII. Multivariable Cox proportional hazards models for overall survival following initial presentation. Supplemental Methods. Further specification. ª 2017 John Wiley & Sons Ltd 565

5 References Aaronson, N.K., Ahmedzai, S., Bergman, B., Bullinger, M., Cull, A., Duez, N.J., Filiberti, A., Flechtner, H., Fleishman, S.B., de Haes, J.C., Kaasa, S., Klee, M., Osoba, D., Razavi, D., Rofe, P.B., Schraub, S., Sneeus, K., Sullivan, M. & Takeda, F. (1993) The European Organization for Research and Treatment of Cancer QLQ- C30: a quality-of-life instrument for use in international clinical trials in oncology. Journal of the National Cancer Institute, 85, Efficace, F., Gaidano, G., Breccia, M., Criscuolo, M., Cottone, F., Caocci, G., Bowen, D., Lubbert, M., Angelucci, E., Stauder, R., Selleslag, D., Platzbecker, U., Sanpaolo, G., Jonasova, A., Buccisano, F., Specchia, G., Palumbo, G.A., Niscola, P., Wan, C., Zhang, H., Fenu, S., Klimek, V., Beyne-Rauzy, O., Nguyen, K. & Mandelli, F. (2015a) Prevalence, severity and correlates of fatigue in newly diagnosed patients with myelodysplastic syndromes. British Journal of Haematology, 168, Efficace, F., Gaidano, G., Breccia, M., Voso, M.T., Cottone, F., Angelucci, E., Caocci, G., Stauder, R., Selleslag, D., Sprangers, M., Platzbecker, U., Ricco, A., Sanpaolo, G., Beyne-Rauzy, O., Buccisano, F., Palumbo, G.A., Bowen, D., Nguyen, K., Niscola, P., Vignetti, M. & Mandelli, F. (2015b) Prognostic value of self-reported fatigue on overall survival in patients with myelodysplastic syndromes: a multicentre, prospective, observational, cohort study. The Lancet. Oncology, 16, Garland, S.N., Carlson, L.E., Stephens, A.J., Antle, M.C., Samuels, C. & Campbell, T.S. (2014) Mindfulness-based stress reduction compared with cognitive behavioral therapy for the treatment of insomnia comorbid with cancer: a randomized, partially blinded, noninferiority trial. Journal of Clinical Oncology, 32, Greenberg, P., Cox, C., LeBeau, M.M., Fenaux, P., Morel, P., Sanz, G., Sanz, M., Vallespi, T., Hamblin, T., Oscier, D., Ohyashiki, K., Toyama, K., Aul, C., Mufti, G. & Bennett, J. (1997) International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood, 89, Greenberg, P.L., Tuechler, H., Schanz, J., Sanz, G., Garcia-Manero, G., Sole, F., Bennett, J.M., Bowen, D., Fenaux, P., Dreyfus, F., Kantarjian, H., Kuendgen, A., Levis, A., Malcovati, L., Cazzola, M., Cermak, J., Fonatsch, C., Le Beau, M.M., Slovak, M.L., Krieger, O., Luebbert, M., Maciejewski, J., Magalhaes, S.M., Miyazaki, Y., Pfeilstocker, M., Sekeres, M., Sperr, W.R., Stauder, R., Tauro, S., Valent, P., Vallespi, T., van de Loosdrecht, A.A., Germing, U. & Haase, D. (2012) Revised international prognostic scoring system for myelodysplastic syndromes. Blood, 120, Jansen, A.J., Essink-Bot, M.L., Beckers, E.A., Hop, W.C., Schipperus, M.R. & Van Rhenen, D.J. (2003) Quality of life measurement in patients with transfusion-dependent myelodysplastic syndromes. British Journal of Haematology, 121, Lallukka, T., Podlipskyte, A., Sivertsen, B., Andruskiene, J., Varoneckas, G., Lahelma, E., Ursin, R., Tell, G.S. & Rahkonen, O. (2016) Insomnia symptoms and mortality: a register-linked study among women and men from Finland, Norway and Lithuania. Journal of Sleep Research, 25, Li, Y., Zhang, X., Winkelman, J.W., Redline, S., Hu, F.B., Stampfer, M., Ma, J. & Gao, X. (2014) Association between insomnia symptoms and mortality: a prospective study of U.S. men. Circulation, 129, Meyers, C.A., Albitar, M. & Estey, E. (2005) Cognitive impairment, fatigue, and cytokine levels in patients with acute myelogenous leukemia or myelodysplastic syndrome. Cancer, 104, Owens, R.L., Gold, K., Gozal, D., Peppard, P.E., Jun, J., Lippman, S.M. & Malhotra, A. (2016) Sleep and breathing... and cancer?: A report from the University of California San Diego Sleep and Cancer Symposium, January 5 6, 2016 in La Jolla, CA. Cancer Prevention Research, 9, Palesh, O.G., Roscoe, J.A., Mustian, K.M., Roth, T., Savard, J., Ancoli-Israel, S., Heckler, C., Purnell, J.Q., Janelsins, M.C. & Morrow, G.R. (2010) Prevalence, demographics, and psychological associations of sleep disruption in patients with cancer: University of Rochester Cancer Center-Community Clinical Oncology Program. Journal of Clinical Oncology, 28, Pinchon, D.J., Stanworth, S.J., Doree, C., Brunskill, S. & Norfolk, D.R. (2009) Quality of life and use of red cell transfusion in patients with myelodysplastic syndromes. A systematic review. American Journal of Hematology, 84, Steensma, D.P., Heptinstall, K.V., Johnson, V.M., Novotny, P.J., Sloan, J.A., Camoriano, J.K., Niblack, J., Bennett, J.M. & Mesa, R.A. (2008) Common troublesome symptoms and their impact on quality of life in patients with myelodysplastic syndromes (MDS): results of a large internet-based survey. Leukemia Research, 32, Thompson, J.E., Conlon, J.P., Yang, X., Sanchez, P.V. & Carroll, M. (2007) Enhanced growth of myelodysplastic colonies in hypoxic conditions. Experimental Hematology, 35, ª 2017 John Wiley & Sons Ltd