Prediction model to predict critical weight loss in patients with head and neck cancer during (chemo)radiotherapy

Oral Oncology xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Oral Oncology journal homepage: www.elsevier.com/locate/oraloncology

Prediction model to predict critical weight loss in patients with head and neck cancer during (chemo)radiotherapy Jacqueline A.E. Langius a,b,⇑, Jos Twisk c, Martine Kampman a, Patricia Doornaert d, Mark H.H. Kramer e, Peter J.M. Weijs a,f, C. René Leemans g a

Department of Nutrition and Dietetics, Internal Medicine, VU University Medical Center Amsterdam, PO Box 7057, 1007 MB Amsterdam, The Netherlands Faculty of Health, Nutrition and Sport, The Hague University of Applied Sciences, Joh. Westerdijkplein 75, 2521EN The Hague, The Netherlands Department of Health Science, Faculty of Earth and Life Sciences, VU University Amsterdam, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands d Department of Radiation Oncology, VU University Medical Center Amsterdam, PO Box 7057, 1007 MB Amsterdam, The Netherlands e Department of Internal Medicine, VU University Medical Center Amsterdam, PO Box 7057, 1007 MB Amsterdam, The Netherlands f Department of Nutrition and Dietetics, Amsterdam University of Applied Sciences, Dr. Meurerlaan 8, 1067 SM Amsterdam, The Netherlands g Department of Otolaryngology/Head and Neck Surgery, VU University Medical Center Amsterdam, PO Box 7057, 1007 MB Amsterdam, The Netherlands b c

a r t i c l e

i n f o

Article history: Received 25 July 2015 Received in revised form 24 October 2015 Accepted 26 October 2015 Available online xxxx Keywords: Head and neck cancer Oral cancer Malnutrition Weight loss Nutrition Radiation Chemoradiotherapy

s u m m a r y Objectives: Patients with head and neck cancer (HNC) frequently encounter weight loss with multiple negative outcomes as a consequence. Adequate treatment is best achieved by early identification of patients at risk for critical weight loss. The objective of this study was to detect predictive factors for critical weight loss in patients with HNC receiving (chemo)radiotherapy ((C)RT). Materials and methods: In this cohort study, 910 patients with HNC were included receiving RT (±surgery/concurrent chemotherapy) with curative intent. Body weight was measured at the start and end of (C)RT. Logistic regression and classification and regression tree (CART) analyses were used to analyse predictive factors for critical weight loss (defined as >5%) during (C)RT. Possible predictors included gender, age, WHO performance status, tumour location, TNM classification, treatment modality, RT technique (three-dimensional conformal RT (3D-RT) vs intensity-modulated RT (IMRT)), total dose on the primary tumour and RT on the elective or macroscopic lymph nodes. Results: At the end of (C)RT, mean weight loss was 5.1 ± 4.9%. Fifty percent of patients had critical weight loss during (C)RT. The main predictors for critical weight loss during (C)RT by both logistic and CART analyses were RT on the lymph nodes, higher RT dose on the primary tumour, receiving 3D-RT instead of IMRT, and younger age. Conclusion: Critical weight loss during (C)RT was prevalent in half of HNC patients. To predict critical weight loss, a practical prediction tree for adequate nutritional advice was developed, including the risk factors RT to the neck, higher RT dose, 3D-RT, and younger age. Ó 2015 Published by Elsevier Ltd.

Introduction Patients with advanced head and neck cancer (HNC) frequently experience weight loss, often already before diagnosis, as well as during and after treatment [1–4]. Weight loss is one of the main symptoms of malnutrition. Malnutrition is a subacute or chronic state in which a combination of varying degrees of undernutrition and inflammatory activity has led to a change in body composition and diminished function [5]. It is associated with a higher risk of ⇑ Corresponding author at: Department of Nutrition and Dietetics, VU University Medical Center, P.O. Box 7057, 1007 MB Amsterdam, The Netherlands. Tel.: +31 204443410; fax: +31 204444143. E-mail address: [email protected] (J.A.E. Langius).

complications, less tolerance and a lower response to treatment, lower quality of life, morbidity and mortality [3,6–12]. Patients with head and neck cancer are predominantly at risk for malnutrition due to anorexia, side effects of the treatment which may hamper food intake (e.g. xerostomia or dysphagia) and metabolic alterations as a result of inflammation induced by the tumour or therapy [13–16]. Prevalence of critical weight loss during (chemo)radiotherapy ((C)RT) varies from 37%, in a mixed group of HNC patients, up to 88% in patients with nasopharyngeal cancer [17,18,1,19]. The prevalence of malnutrition in head and neck cancer patients differs considerably between studies due to differences in tumour locations and stages, the intensity of the anti-tumour treatment, the availability of nutritional therapy and the different criteria used

http://dx.doi.org/10.1016/j.oraloncology.2015.10.021 1368-8375/Ó 2015 Published by Elsevier Ltd.

Please cite this article in press as: Langius JAE et al. Prediction model to predict critical weight loss in patients with head and neck cancer during (chemo) radiotherapy. Oral Oncol (2015), http://dx.doi.org/10.1016/j.oraloncology.2015.10.021

2

J.A.E. Langius et al. / Oral Oncology xxx (2015) xxx–xxx

in the literature for the identification of malnourished patients [20]. Knowledge of risk factors for malnutrition may contribute to effective management, or preferably prevention of malnutrition in these patients. Over the last years, several studies have been undertaken to investigate predictive risk factors for weight loss during (C)RT [21]. However, only few well powered studies conducted multivariable analyses in the total group of head and neck cancer patients and included radiotherapeutic factors, like dose and technique. Those treatment factors might have a high predictive value for weight loss during (C)RT [22]. In a previous study we investigated predictive factors for critical weight loss during RT in early stage laryngeal cancer patients [2]. In this group of HNC patients, RT to the elective neck regions was found to be a simple and useful predictor for malnutrition, identifying 72% of patients who experienced critical weight loss during treatment. However, the total HNC population is a rather heterogeneous group with differences in diagnoses, stages and treatment schedules. Analysing predictive factors could lead to a complex model. Therefore, alongside classical logistic regression analyses we explored a more practical prediction tree for critical weight loss [23]. Material and methods In this retrospective analysis of prospectively collected data, we included newly diagnosed adult HNC patients receiving primary or postoperative RT with or without chemotherapy with curative intent at the VU University Medical Center Amsterdam between 2000 and 2012. All patients with tumours of the oral cavity, oropharynx, hypopharynx and larynx were included for analysis. Patients receiving reirradiation or brachytherapy were excluded. Radiotherapy Patients received (C)RT for 6–7 weeks using megavoltage equipment. Since October 2004, the Radiotherapy department implemented intensity-modulated RT (IMRT) as follow up for the three-dimensional conformal RT (3D-RT) [24,25]. RT on the primary tumour consisted of 5–6 fractions of 2–2.5 Gy a week, to a total dose of 56–70 Gy. Patients with locally advanced cancer received prophylactic irradiation to the neck nodes to a total dose of 46 Gy and a boost on the primary tumour and lymph nodes metastases to a total dose of 70 Gy. In case of concomitant chemoradiation, 100 mg/m2 cisplatin was given in week 1, 4 and 7, or equivalent in a weekly dose. Patients received protocolised dietary counselling by a dietitian at the start of (C)RT and in the third, fifth and last week of (C)RT. Dietary counselling was aimed at achieving individual nutritional requirements. If nutritional requirements could not be reached by regular food products, energy-enriched oral nutritional supplements and/or enteral tube feeding by nasogastric tube or percutaneous endoscopic gastrostomy were prescribed. Body weight Body weight was measured around the start of (C)RT and weekly during (C)RT. Body weight was assessed on a calibrated digital scale (Seca, Alpha 770), with an accuracy of 0.1 kg. Patients were wearing light indoor clothes and shoes. Body weight was corrected for clothes and shoes by subtracting 2.0 kg for men and 1.3 kg for women [26]. Critical weight loss was defined as weight loss >5% between the start and end of (C)RT. As we earlier observed that acute toxicity and weight loss arise after two weeks of RT, we expected weight loss to occur in the last month of (C)RT and

applied the ASPEN/AND consensus criteria for severe weight loss [27]. Predictors of weight loss Possible predictors of critical weight loss during (C)RT were recorded at the start of (C)RT. These included gender, age, WHO performance status (score 2 or 3 versus score 0 or 1) [28], tumour location (oral cavity, oropharynx, hypopharynx, and larynx), staging (I, II, III, and IV) [29], treatment modality (RT alone, RT + surgery, and RT + chemotherapy ± surgery), RT technique (3D-RT versus IMRT), total dose on primary tumour (>65 Gy versus 665 Gy), and RT on the elective lymph node regions (none, unilateral, bilateral). The toxicity scoring system of the Radiation Therapy Oncology Group (RTOG) was used to score dysphagia and xerostomia at the start of (C)RT [30]. Weight loss before (C)RT was categorized into four groups: no weight loss, 65% weight loss, >5–10% weight loss, and >10% weight loss. Statistical analysis Two methods were used to detect possible predictors of critical weight loss during (C)RT. (1) Logistic regression analyses. The possible predictor variables were first tested for their association in univariable analyses. Then a forward selection procedure with all candidate predictors was used to create the ’best’ prediction model. (2) Classification and regression tree (CART) analysis was used to build a prediction tree. CART analysis is based on the method of recursive-partitioning analysis [31]. In this analysis, the total group is split into subgroups based on the variable with the highest risk for critical weight loss. These subgroups are again partitioned, until no further significant partitioning was possible or when the group size became below 50, resulting in a tree with the best predictive accuracy. In case of continuous candidate predictors, groups were split at the cutoff point with the highest discriminative value for critical weight loss. In case of missing values for one of the candidate predictors, these missing values were handled per predictor as a separate category. For the patients with the missing value the risk for critical weight loss was therefore separately calculated. Based on their risk, these patients were classified to the best fitting node, i.e. the value with the highest or the lowest risk for critical weight loss. The terminal nodes in the prediction tree were used to classify patients’ risk for critical weight loss based on the predictor variables. The CART analysis was conducted with the same possible predictor variables as used in the logistic regression analyses. For both methods the C-statistic was used to indicate the quality of the final model. Statistical analyses were carried out with SPSS version 20.0 (2011, IBM Corporation, Armonk, New York, USA). p-values less than 0.05 were considered as statistically significant. Results In this study, 910 patients with HNC were included. Complete data for all candidate predictors were available in 895 patients (98% of the total group). Mean age was 62.2 ± 10.9 year and 71% of patients were male (Table 1). Tumours were mainly located at the larynx (40%), and the oropharynx (32%). Most tumours were classified as stage IV (50%). Weight loss Before (C)RT, 16% of the patients had weight loss of more than 5%. During (C)RT, 87% of patients lost weight. Mean body weight decreased with 5.1 ± 4.8%, corresponding to 3.9 ± 3.7 kg.

Please cite this article in press as: Langius JAE et al. Prediction model to predict critical weight loss in patients with head and neck cancer during (chemo) radiotherapy. Oral Oncol (2015), http://dx.doi.org/10.1016/j.oraloncology.2015.10.021

3

J.A.E. Langius et al. / Oral Oncology xxx (2015) xxx–xxx Table 1 Patient characteristics at start of radiotherapy for patients with or without critical weight loss (>5%) during (chemo)radiotherapy. Characteristics

Gender Female Male Age Mean (SD) WHO performance status Grade 0 or 1 Grade 2 or 3 Weight loss before RT None 65% >5–10% >10% Site Oral cavity Oropharynx Hypopharynx Larynx TNM stage I II III IV Treatment modality Surgery + RT RT alone ChemoRT (+surgery; n = 10) RT technique 3D-RT IMRT RT dose 665 Gy >65 Gy Median (interquartile range) RT to the neck No Unilateral Bilateral

Critical weight loss n (%) 450 (49)

No critical weight loss n (%) 460 (51)

OR (95% CI)

138 (52) 312 (48)

126 (48) 334 (52)

ref. 0.85 (0.64–1.14)

61.0 ± 10.8

63.5 ± 10.8

0.98 (0.97–0.99)

399 (51) 48 (38)

379 (49) 79 (62)

ref. 0.58 (0.39–0.85)

302 (50) 80 (54) 40 (48) 21 (40)

301 (50) 68 (46) 43 (52) 31 (60)

ref. 1.17 (0.82–1.68) 0.93 (0.59–1.47) 0.68 (0.38–1.20)

89 (45) 177 (60) 26 (44) 158 (44)

109 (55) 117 (40) 33 (56) 201 (56)

ref. 1.85 (1.29–2.67) 0.97 (0.54–1.73) 0.96 (0.68–1.37)

41 (32) 92 (54) 84 (53) 233 (52)

86 (68) 80 (46) 76 (47) 218 (48)

ref. 2.41 (1.50–3.89) 2.32 (1.43–3.76) 2.24 (1.48–3.40)

122 (44) 208 (48) 120 (60)

158 (56) 222 (52) 80 (40)

ref. 1.12 (0.90–1.64) 1.94 (1.34–2.81)

315 (49) 132 (51)

329 (51) 127 (49)

ref. 1.09 (0.81–1.45)

109 (34) 341 (58)

211 (66) 248 (42)

ref. 2.66 (2.01–3.53)

38 (23) 66 (48) 345 (57)

127 (77) 71 (52) 261 (43)

ref. 3.11 (1.90–5.01) 4.42 (2.97–6.57)

p-value

0.276

0.001 0.005

0.389

<0.001

<0.001

0.001

0.577

<0.001

<0.001

Note: p-values < 0.05 are shown in bold. OR, 95%CI and p-values were obtained by univariable logistic regression analyses, with critical weight loss as dependent variable. Abbreviations: OR, odds ratio; CI, confidential interval; ref., reference; RT, radiotherapy; SD, standard deviation; 3D-RT, three-dimensional conformal radiotherapy; IMRT, intensity-modulated radiotherapy.

Critical weight loss during (C)RT was diagnosed in 50% of the patients. Patients with critical weight loss lost on average 8.9 ± 3.1% of their body weight, whereas patients without critical weight loss lost on average 1.4 ± 3.0%. Of the patients with critical weight loss, 32% lost more than 10% of body weight between start and end of (C)RT. Body weight of patients with critical weight loss already decreased in the first weeks of (C)RT, whereas it remained stable in patients without critical weight loss (Fig. 1). Critical weight loss was most common in patients with oropharyngeal cancer (60%), in patients receiving chemoradiotherapy (60%), a RT dose above 65 Gy (58%) and RT to the neck (57% for bilateral RT; Table 1). Predictive factors of weight loss during (C)RT In the univariable logistic regression analyses 7 possible predictive factors were significantly associated with critical weight loss during (C)RT (Table 1). RT to the neck was the main predictor for critical weight loss. Odds ratios were 3.2 and 4.1 for respectively uni- and bilateral irradiation, with no significant difference between unilateral and bilateral. Table 2 shows the final multivariable predictive model for critical weight loss during (C)RT. The independent predictors for

critical weight loss during (C)RT were RT to the neck, higher RT doses on the primary tumour, receiving 3D-RT versus IMRT, having better WHO performance score, and lower age. Because 80 patients received tube feeding at the start of (C)RT, we reanalysed the data without these patients. The multivariate prediction model was comparable to the model shown in Table 2 with comparable odds ratios, except for WHO performance score, which was no longer significantly associated with critical weight loss (Appendix I). Fig. 2 presents the prediction tree based on recursivepartitioning analysis. RT to the neck was the principal discriminator, with uni- and bilateral RT merged together because of comparable risks. Patients with uni- or bilateral irradiation to the neck had a probability of 55% for critical weight loss compared to patients without neck irradiation that had a probability of 23%. Among the group with RT to the neck, the next division was by the amount of RT on the primary tumour. Patients who received a RT dose above 65 Gy were further partitioned by RT technique. For patients with an RT dose of 65 Gy or less, age (i.e. lower or higher than 56 years) was the last determinant of critical weight loss. The C-statistic for the final logistic regression model was 0.69 (95% CI 0.66–0.73) and for the prediction tree 0.68 (95% CI 0.64–0.71), representing an acceptable ability to predict critical weight loss [32].

Please cite this article in press as: Langius JAE et al. Prediction model to predict critical weight loss in patients with head and neck cancer during (chemo) radiotherapy. Oral Oncol (2015), http://dx.doi.org/10.1016/j.oraloncology.2015.10.021

4

J.A.E. Langius et al. / Oral Oncology xxx (2015) xxx–xxx

Critical weight loss

Fig. 1. Weight change (%) during radiotherapy for head and neck cancer patients with or without critical weight loss (>5% weight loss between start and end of radiotherapy).

Table 2 Final multivariable logistic regression model for the prediction of critical weight loss in patients with head and neck cancer during (chemo)radiotherapy. Predictors RT on neck nodes Unilateral vs no Bilateral vs no RT dose on primary tumour >65 Gy vs 665 Gy RT technique, 3D vs IMRT WHO performance status Grade 2 or 3 vs 0 or 1 Age, per 10 years

OR

95% CI

p-value

3.19 4.05

1.90–5.37 2.61–6.29

<0.001 <0.001

2.10 1.70

1.52–2.89 1.23–2.36

<0.001 0.001

0.55 0.84

0.37–0.83 0.74–0.97

0.004 0.010

Dependent variable is >5% weight loss during RT. Abbreviations: OR, odds ratio; CI, confidence interval; RT, radiotherapy; Gy, gray; 3D, three-dimensional conformal radiotherapy; IMRT, intensity-modulated radiotherapy.

Discussion Patients with head and neck cancer are at high risk of critical weight loss during treatment. In this study the prevalence of critical weight loss was 50%. Independent predictors for critical weight loss during (C)RT are RT to the neck nodes, higher RT doses on the primary tumour, receiving 3D-RT versus IMRT, and lower age, as demonstrated in both multivariable logistic regression and CART analysis. We demonstrated previously that RT to the elective neck node regions is the main predicting factor for RT-induced critical weight loss in patients with early stage laryngeal cancer [2]. Also in the current cohort of patients with head and neck cancers from different sites, RT to the neck is the most prominent predictor for critical weight loss. These patients have a 3–4 fold higher risk for critical weight loss than patients without RT to the neck. By irradiation to the neck, the major salivary glands are (partially) included in the radiation field, leading to salivary dysfunction. As a consequence patients may experience xerostomia, taste alterations and swallowing difficulties [33]. These complaints may negatively impact food intake [34] and nutritional status [35]. The salivary function is often already reduced by 50–60% in the first week of (C)RT [36]. Although this is speculative, xerostomia

might be a possible explanation why weight already decreased in the first weeks of (C)RT. Although radiation-induced toxicity is more severe when treating the bilateral neck compared to unilateral irradiation [37], we did not observe a difference in the amount of critical weight loss between these groups. The majority of HNC patients receive either uni- or bilateral radiation to the neck, resulting in a large group of patients with high risk for malnutrition. Besides RT to the neck, a high RT dose and 3D-RT were important factors, together accounting for a 66% risk for critical weight loss. We found age to be inversely associated with critical weight loss during (C)RT. Older patients had lower risk for critical weight loss. The reason for this finding is not clear. We performed posthoc analyses to evaluate whether age was related to other factors which might explain the association with critical weight loss. Weight at start of (C)RT was not different between the age-groups. Patients older than 56 years had more often cancer of the larynx and less often cancer of the oropharynx, and they received more often primary RT instead of postoperative RT than patients aged below 56 years. However, in the subgroup defined by the CART analyses these differences were no longer relevant. We therefore think that the inverse association between age and weight loss might be related to other factors, like a higher activity level, i.e. higher energy expenditure in younger, mostly still working patients, and/or a greater decline of physical activity level in the younger patients with subsequently loss of large amounts of muscle mass. Another explanation might be more intense supportive care in older patients anticipating on the expectation that they are going to do worse during therapy. However, further studies on this topic should be undertaken. Because a small part of the patients received tube feeding at the onset of (C)RT, we controlled for this intervention. Without these patients the logistic model was nearly the same, excluding the influence of this early tube feeding. Part of the patients, especially those with chemoradiation, started tube feeding during therapy. This might explain why chemoradiation was no risk factor for critical weight loss in the final model of our study, while it was in other studies [1,22,38,39]. We choose to conduct a prediction tree by CART analysis. In this analysis, interaction between the predictors for critical weight loss were evaluated recursively, whereas it was evaluated

Please cite this article in press as: Langius JAE et al. Prediction model to predict critical weight loss in patients with head and neck cancer during (chemo) radiotherapy. Oral Oncol (2015), http://dx.doi.org/10.1016/j.oraloncology.2015.10.021

J.A.E. Langius et al. / Oral Oncology xxx (2015) xxx–xxx

5

Fig. 2. Prediction tree for critical weight loss in head and neck cancer patients during (chemo)radiotherapy. Nodes display sample size (n) and the percentage of patients with critical weight loss within the (sub)group. The potential predictors in the CART analyses were gender, age, WHO/ECOG performance status, weight loss before radiotherapy, tumour location, TNM classification, treatment modality, RT technique, total RT dose on primary tumour, and RT on the elective or macroscopic lymph nodes. Abbreviations: CWL, critical weight loss; RT, radiotherapy; 3D-RT, three-dimensional conformal radiotherapy; IMRT, intensity-modulated radiotherapy.

simultaneously in logistic regression analyses. These trees are considered to represent complex diagnostic and treatment strategies in a way that is ideally suited to mimic actual thinking processes [40]. The advantage of this method is that it is easier to use in clinical practice than multivariate logistic regression models, because no calculation of regression scores is necessary. Moreover, this approach is transparent for practicing clinicians and patients [41]. One of the disadvantages of CART analysis is the low external validity of the prediction tree. The final prediction tree consisted of 5 end nodes, classifying patients into categories with respect to the risk of critical weight loss during (C)RT. Risks ranged from 23% for patients without RT to the neck, to 66% for patients with >65 Gy 3D-RT with neck irradiation. The same predictors were found by CART analysis and logistic regression analysis, which strengthened our finding. Studies reporting on factors associated with weight loss before therapy found diagnosis and stage mostly significantly related to critical weight loss [4,42,43]. It seems that the situation at diagnosis may not be predictive for weight loss during therapy. We also found that diagnosis and tumour stage are important factors in univariable, but not in multivariable analysis. This was in line with other studies including therapeutic factors in their multivariate analyses [1,22,38]. We intuitively expected that laryngeal cancer patients would have less acute toxicity relating to food intake than pharyngeal cancer patients. Probably, the effort undertaken to reduce the radiation dose to the swallowing structures has led to less differences between the tumour sites. A strength of our study was the large sample size and the completeness of the data. However, due to the large cohort we could not delineate organs at risk for weight loss for each patient. We used target volume in general (local versus unilateral neck versus bilateral neck) as a surrogate for dose to the swallowing structures, because earlier this has been found a useful indicator for dose distribution to these organs at risk. We realise that using irradiated volumes like the planning target volume (PTV) might even be a better predictor. Mallick et al. recently showed that a PTV receiving

the prescription dose above 235 cc or a PTV receiving the elective dose above 615 cc was, besides chemoradiation, the main predictor for critical weight loss [22]. The cut off point for critical weight loss was based on the international consensus statement of the Academy of Nutrition and Dietetics and the American Society for Parenteral and Enteral Nutrition [27]. It was thereby assumed that (C)RT induced acute toxicity and weight loss would arise after 2 weeks of (C)RT [44,45]. Therefore the time span for critical weight loss during (C)RT was expected to be in the last month of (C)RT. In retrospect, we found that weight loss started already after 1 week and the period of weight loss might therefore have been approximately 5–6 weeks instead of one month. Several studies in head and neck cancer patients used the same cut off value for critical weight loss during radiotherapy as we did [22,38,46,47], however the best criteria for describing malnutrition is still under debate [20]. Despite dietary counselling, and supplements or tube feeding if necessary, half of the HNC patients still had critical weight loss during (C)RT. Although weight loss already started within 1 week after the start of (C)RT, patients were not counselled in the first weeks, because we expected acute toxicity and weight loss later during (C)RT. The findings of this study support dietary intervention [48] from the commencement of treatment and evaluation of weight loss after the first week of (C)RT, especially in those patients with high risk for malnutrition during (C)RT. Furthermore, the dietary intervention should become more proactive instead of anticipating on decreased intake and lost weight, so before the problem has arisen. In the group of patients with the lowest risk for malnutrition, still almost a quarter of the group suffered from critical weight loss. In these patients, weekly monitoring of weight, and dietary intervention by ongoing weight loss might be more efficient and functional. The prediction model, including the predictive factors RT to the neck, higher RT doses on the primary tumour, receiving 3D-RT versus IMRT and lower age might help to identify head and neck cancer patients with high risk for critical weight loss. Early

Please cite this article in press as: Langius JAE et al. Prediction model to predict critical weight loss in patients with head and neck cancer during (chemo) radiotherapy. Oral Oncol (2015), http://dx.doi.org/10.1016/j.oraloncology.2015.10.021

6

J.A.E. Langius et al. / Oral Oncology xxx (2015) xxx–xxx

intensive dietary counselling might prevent critical weight loss in these patents during (C)RT. Conflict of interest statement None declared. Acknowledgements We like to thank Hanneke Wijnhoven, Ans van Stijgeren and Loes van Aken for their valuable advice. This study was funded by the Cancer Center Amsterdam Foundation, Amsterdam, The Netherlands. The funding source was not involved in study design, data collection, data-analysis and interpretation, manuscript writing or in the decision to submit the manuscript for publication. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.oraloncology. 2015.10.021. References [1] Munshi A, Pandey MB, Durga T, Pandey KC, Bahadur S, Mohanti BK. Weight loss during radiotherapy for head and neck malignancies: what factors impact it? Nutr Cancer 2003;47(2):136–40. [2] Langius JAE, Doornaert P, Spreeuwenberg MD, Langendijk JA, Leemans CR, Schueren MA. Radiotherapy on the neck nodes predicts severe weight loss in patients with early stage laryngeal cancer. Radiother Oncol 2010;97(10):80–5. [3] Chasen MR, Bhargava R. A descriptive review of the factors contributing to nutritional compromise in patients with head and neck cancer. Support Care Cancer 2009;17(11):1345–51. [4] Jager-Wittenaar H, Dijkstra PU, Vissink A, van der Laan BF, van Oort RP, Roodenburg JL. Critical weight loss in head and neck cancer-prevalence and risk factors at diagnosis: an explorative study. Support Care Cancer 2007;15 (9):1045–50. [5] Soeters PB, Reijven PL, van Bokhorst-de van der Schueren MA, Schols JM, Halfens RJ, Meijers JM, et al. A rational approach to nutritional assessment. Clin Nutr 2008;27(5):706–16. [6] Di FF, Lecleire S, Pop D, Rigal O, Hamidou H, Paillot B, et al. Baseline nutritional status is predictive of response to treatment and survival in patients treated by definitive chemoradiotherapy for a locally advanced esophageal cancer. Am J Gastroenterol 2007;102(11):2557–63. [7] Cho YW, Roh JL, Jung JH, Kim SB, Lee SW, Choi SH, et al. Prediction of posttreatment significant body weight loss and its correlation with diseasefree survival in patients with oral squamous cell carcinomas. Nutr Cancer 2013;65(3):417–23. [8] Hammerlid E, Wirblad B, Sandin C, Mercke C, Edstrom S, Kaasa S, et al. Malnutrition and food intake in relation to quality of life in head and neck cancer patients. Head Neck 1998;20(6):540–8. [9] Nguyen TV, Yueh B. Weight loss predicts mortality after recurrent oral cavity and oropharyngeal carcinomas. Cancer 2002;95(3):553–62. [10] Langius JA, Bakker S, Rietveld DH, Kruizenga HM, Langendijk JA, Weijs PJ, et al. Critical weight loss is a major prognostic indicator for disease-specific survival in patients with head and neck cancer receiving radiotherapy. Br J Cancer 2013;109(5):1093–9. [11] Ravasco P, Monteiro GI, Camilo M. Cancer wasting and quality of life react to early individualized nutritional counselling! Clin Nutr 2007;26(1):7–15. [12] Trotti A, Bellm LA, Epstein JB, Frame D, Fuchs HJ, Gwede CK, et al. Mucositis incidence, severity and associated outcomes in patients with head and neck cancer receiving radiotherapy with or without chemotherapy: a systematic literature review. Radiother Oncol 2003;66(3):253–62. [13] Baracos VE. Cancer-associated cachexia and underlying biological mechanisms. Annu Rev Nutr 2006;26:435–61. [14] Richey LM, George JR, Couch ME, Kanapkey BK, Yin X, Cannon T, et al. Defining cancer cachexia in head and neck squamous cell carcinoma. Clin Cancer Res 2007;13(22 Pt 1):6561–7. [15] Silver HJ, Dietrich MS, Murphy BA. Changes in body mass, energy balance, physical function, and inflammatory state in patients with locally advanced head and neck cancer treated with concurrent chemoradiation after low-dose induction chemotherapy. Head Neck 2007;29(10):893–900. [16] Van Cutsem E, Arends J. The causes and consequences of cancer-associated malnutrition. Eur J Oncol Nurs 2005;9(Suppl. 2):S51–63. [17] Brookes GB. Nutritional status–a prognostic indicator in head and neck cancer. Otolaryngol Head Neck Surg 1985;93(1):69–74. [18] Collins MM, Wight RG, Partridge G. Nutritional consequences of radiotherapy in early laryngeal carcinoma. Ann R Coll Surg Engl 1999;81(6):376–81.

[19] Ng K, Leung SF, Johnson PJ, Woo J. Nutritional consequences of radiotherapy in nasopharynx cancer patients. Nutr Cancer 2004;49(2):156–61. [20] Dechaphunkul T, Martin L, Alberda C, Olson K, Baracos V, Gramlich L. Malnutrition assessment in patients with cancers of the head and neck: a call to action and consensus. Crit Rev Oncol Hematol 2013;88(2):459–76. [21] Zhao JZ, Zheng H, Li LY, Zhang LY, Zhao Y, Jiang N. Predictors for weight loss in head and neck cancer patients undergoing radiotherapy: a systematic review. Cancer Nurs 2015;38(6):E37–45. [22] Mallick I, Gupta SK, Ray R, Sinha T, Sinha S, Achari R, et al. Predictors of weight loss during conformal radiotherapy for head and neck cancers – how important are planning target volumes? Clin Oncol (R Coll Radiol) 2013;25 (9):557–63. [23] Shariat SF, Kattan MW, Vickers AJ, Karakiewicz PI, Scardino PT. Critical review of prostate cancer predictive tools. Future Oncol 2009;5(10):1555–84. [24] Fang FM, Tsai WL, Chen HC, Hsu HC, Hsiung CY, Chien CY, et al. Intensitymodulated or conformal radiotherapy improves the quality of life of patients with nasopharyngeal carcinoma: comparisons of four radiotherapy techniques. Cancer 2007;109(2):313–21. [25] Mendenhall WM, Amdur RJ, Palta JR. Intensity-modulated radiotherapy in the standard management of head and neck cancer: promises and pitfalls. J Clin Oncol 2006;24(17):2618–23. [26] Frank E, Dunlop AL. What does a patient’s outfit weight? Fam Med 2000;32 (9):595–6. [27] White JV, Guenter P, Jensen G, Malone A, Schofield M. Consensus statement: Academy of Nutrition and Dietetics and American Society for Parenteral and Enteral Nutrition: characteristics recommended for the identification and documentation of adult malnutrition (undernutrition). JPEN J Parenter Enteral Nutr 2012;36(3):275–83. [28] Health and social care: data dictionary. WHO/ECOG Perform Status; 2010. [29] Green FL, Page DL, Fleming ID, Fritz A, Balch CM, Haller DG, et al. AJCC cancer staging manual. 6th ed. New York: Springer-Verslag; 2002. [30] Cox JD, Stetz J, Pajak TF. Toxicity criteria of the Radiation Therapy Oncology Group (RTOG) and the European Organization for Research and Treatment of Cancer (EORTC). Int J Radiat Oncol Biol Phys 1995;31(5):1341–6. [31] Breiman L, Friedman JH, Olshen RA, Stone RJ. Classification and regression trees. New York: Chapman and Hall; 1984. [32] Hosmer DW, Lemeshow S. Applied logistic regression. 2nd ed. New York: John Wiley & Sons; 2000. [33] Dirix P, Nuyts S. Evidence-based organ-sparing radiotherapy in head and neck cancer. Lancet Oncol 2010;11(1):85–91. [34] Kubrak C, Olson K, Baracos VE. The head and neck symptom checklist(c): an instrument to evaluate nutrition impact symptoms effect on energy intake and weight loss. Support Care Cancer 2013;21(11):3127–36. [35] Jager-Wittenaar H, Dijkstra PU, Vissink A, Langendijk JA, van der Laan BF, Pruim J, et al. Changes in nutritional status and dietary intake during and after head and neck cancer treatment. Head Neck 2011;33(6):863–70. [36] Konings AW, Coppes RP, Vissink A. On the mechanism of salivary gland radiosensitivity. Int J Radiat Oncol Biol Phys 2005;62(4):1187–94. [37] Jellema AP, Slotman BJ, Doornaert P, Leemans CR, Langendijk JA. Unilateral versus bilateral irradiation in squamous cell head and neck cancer in relation to patient-rated xerostomia and sticky saliva. Radiother Oncol 2007;85 (1):83–9. [38] Palazzi M, Tomatis S, Orlandi E, Guzzo M, Sangalli C, Potepan P, et al. Effects of treatment intensification on acute local toxicity during radiotherapy for head and neck cancer: prospective observational study validating CTCAE, version 3.0, scoring system. Int J Radiat Oncol Biol Phys 2008;70(2):330–7. [39] Qiu C, Yang N, Tian G, Liu H. Weight loss during radiotherapy for nasopharyngeal carcinoma: a prospective study from northern China. Nutr Cancer 2011;63(6):873–9. [40] Ciampi A, Negassa A, Lou Z. Tree-structured prediction for censored survival data and the Cox model. J Clin Epidemiol 1995;48(5):675–89. [41] Ebell M. AHRQ white paper: use of clinical decision rules for point-of-care decision support. Med Decis Making 2010;30(6):712–21. [42] Righini CA, Timi N, Junet P, Bertolo A, Reyt E, Atallah I. Assessment of nutritional status at the time of diagnosis in patients treated for head and neck cancer. Eur Ann Otorhinolaryngol Head Neck Dis 2013;130(1):8–14. [43] Farhangfar A, Makarewicz M, Ghosh S, Jha N, Scrimger R, Gramlich L, et al. Nutrition impact symptoms in a population cohort of head and neck cancer patients: multivariate regression analysis of symptoms on oral intake, weight loss and survival. Oral Oncol 2014;50(9):877–83. [44] Raber-Durlacher JE, Elad S, Barasch A. Oral mucositis. Oral Oncol 2010;46 (6):452–6. [45] Rodriguez-Caballero A, Torres-Lagares D, Robles-Garcia M, Pachon-Ibanez J, Gonzalez-Padilla D, Gutierrez-Perez JL. Cancer treatment-induced oral mucositis: a critical review. Int J Oral Maxillofac Surg 2012;41(2):225–38. [46] Nourissat A, Bairati I, Samson E, Fortin A, Gelinas M, Nabid A, et al. Predictors of weight loss during radiotherapy in patients with stage I or II head and neck cancer. Cancer 2010;116(9):2275–83. [47] Aaronson NK, Ahmedzai S, Bergman B, Bullinger M, Cull A, Duez NJ, et al. The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst 1993;85(5):365–76. [48] Langius JA, Zandbergen MC, Eerenstein SE, Van Tulder MW, Leemans CR, Kramer MH, et al. Effect of nutritional interventions on nutritional status, quality of life and mortality in patients with head and neck cancer receiving (chemo)radiotherapy: a systematic review. Clin Nutr 2013;32(5):671–8.

Please cite this article in press as: Langius JAE et al. Prediction model to predict critical weight loss in patients with head and neck cancer during (chemo) radiotherapy. Oral Oncol (2015), http://dx.doi.org/10.1016/j.oraloncology.2015.10.021