Abstract
Grade ≥ 3 radiation pneumonitis (RP) is generally severe and life-threatening. Predictors of grade ≥ 2 are usually used for grade ≥ 3 RP prediction, but it is unclear whether these predictors are appropriate. In this study, predictors of grade ≥ 2 and grade ≥ 3 RP were investigated separately. The increased risk of severe RP in elderly patients compared with younger patients was also evaluated. Material and methods. A total of 176 consecutive patients with locally advanced non-small cell lung cancer were followed up prospectively after three-dimensional conformal radiotherapy. RP was graded according to Common Terminology Criteria for Adverse Events version 3.0. Results. Mean lung dose (MLD), mean heart dose, ratio of planning target volume to total lung volume (PTV/Lung), and dose-volume histogram comprehensive value of both heart and lung were associated with both grade ≥ 2 and grade ≥ 3 RP in univariate analysis. In multivariate logistic regression analysis, age and MLD were predictors of both grade ≥ 2 RP and grade ≥ 3 RP; receipt of chemotherapy predicted grade ≥ 3 RP only; and sex and PTV/Lung predicted grade ≥ 2 RP only. Among patients who developed high-grade RP, MLD and PTV/Lung were significantly lower in patients aged ≥ 70 years than in younger patients (p < 0.05 for both comparisons). Conclusions. The predictors were not completely consistent between grade ≥ 2 RP and grade ≥ 3 RP. Elderly patients had a higher risk of severe RP than younger patients did, possibly due to lower tolerance of radiation to the lung.
Despite recent technological advances in radiation therapy (RT) such as three-dimensional (3D) conformal RT and intensity-modulated RT, radiation pneumonitis (RP) remains an important dose-limiting toxicity in thoracic RT. Many dose-volume parameters, such as V5 [Citation1], V20 [Citation2], V30 [Citation3], mean lung dose [Citation4,Citation5], and normal tissue complication probability (NTCP) [Citation6], have been reported to be correlated with the development of RP. In the National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology (version 2.2011) for non-small cell lung cancer, the upper limits for lung dose-volume are V20 < 37% and mean lung dose < 20 Gy. However, patient characteristics such as age [Citation7], history of chronic obstructive pulmonary disease [Citation2], and performance status [Citation8] vary greatly, leading to differences in the tolerance of lung tissue to radiation. For this reason, evaluating the presence and severity of RP in patients by using unified values of dose-volume parameters is not suitable. Combining dose-volume parameters with patient characteristics to predict individual risk of developing RP is more appropriate.
Most scholars [Citation3,Citation4,Citation7,Citation9,Citation10] have evaluated factors associated with grade ≥ 2 RP. However, grade ≥ 3 RP is generally severe and life-threatening [Citation11], so identifying predictive factors of grade ≥ 3 RP is more important. Whether the factors associated with grade ≥ 2 RP are necessarily predictive of grade ≥ 3 RP is unclear. To address this gap in knowledge, we performed a prospective investigation to identify clinical and dose-volume factors correlated with the development of grade ≥ 2 RP and grade ≥ 3 RP in patients with non-small cell carcinoma who were treated with 3D conformal RT. We also evaluated the higher risk of RP in elderly patients compared with younger patients and the possible mechanism for this difference.
Methods and materials
Patients
From October 2009 through October 2011, 214 consecutive patients with locally advanced non-small cell lung cancer received 3D conformal RT at the Radiotherapy Department of The First Hospital of China Medical University. These patients were recruited into our study. Patients were excluded from the study if they had had a break from RT of more than five days (n = 8 patients), radiation with inconsistent doses per fraction (n = 3), a total radiation dose less than 50 Gy (n = 14), or a post-RT performance status score > 2 (n = 13). Thus, 176 patients were followed up prospectively after RT was completed.
Chemotherapy
Among the 176 patients included in our prospective study, 51 received RT alone and 125 received chemoradiotherapy. Among the latter, 107 patients received concurrent chemoradiotherapy followed by consolidation chemotherapy (1 or 2 cycles administered during RT, and 2 or 3 cycles administered after RT) and 18 patients received only sequential chemotherapy (3 or 4 cycles administered after RT). Most of the regimens for concurrent chemoradiotherapy followed by consolidation chemotherapy consisted of vinorelbine (20 mg/m2 on days 1 and 5) and cisplatin (20 mg/m2 on days 1–3). All patients who went sequential chemotherapy had the same regimen: docetaxel (75 mg/m2 on day 1) and cisplatin (20–25 mg/m2 on days 1–3).
Radiation therapy
All patients were treated with 6-MV x-rays with any combination of coplanar or non-coplanar 3D conformal fields. Treatments planning computed tomography (CT) scans with slices 5 mm thick were obtained from the mandible to the lower edge of the liver before RT and again when the dosage reached 35–40 Gy. Gross tumor volume (GTV) was defined as the total volume of the primary and nodal tumor masses visualized on any planning CT images. The clinical target volume was defined as the GTV plus a 0.6 cm × 0.8 cm margin, and the planning target volume (PTV) was defined as the clinical target volume plus a 0.5 cm × 1.0 cm margin for setup uncertainty and respiratory motion. The prescribed dose per fraction was 62.5–65 Gy for patients who received RT alone and 2.0–2.5 Gy with a total dosage of 60–62.5 Gy for patients who underwent concurrent or sequential chemoradiotherapy. Five or six fractions in one week and four or five fields were usually used in the treatment plan.
Dose-volume histogram (DVH) parameters
The total normal lung volume was defined as the total lung volume minus the primary GTV and the volume of the trachea and main bronchi. We extracted a wide range of dose-volume parameters for heart and total lung for modeling: Vx (V5, V10, V15, V20, V25, V30, V35, V40, V45, V50), MLD, mean heart dose (MHD), and the ratio of PTV to total lung volume (PTV/Lung). Vx was defined as the percentage of total normal lung (LVx) or heart (HVx) volume receiving at least x Gy of radiation.
Due to colinearity among the dose-volume variables (specifically, V5–V50), we could not induct them into the regression model simultaneously when the multivariate analysis was carried out. We used principal component analysis, a statistical method to convert a large amount of related data to a smaller amount of unrelated data, to compress the data and thus glean pertinent information. As the accumulation of variance was 94.74% for LVx and 92.32% for HVx, we used the contribution rate of two principal components (component 1 of LVx including LV30–LV50 and component 2 of LVx including LV5–LV25; and component 1 of HVx including HV5–LV25 and component 2 of HVx including HV30–HV50) as weight numbers to calculate the comprehensive value (DVH-CV) and to analyze its correlation with RP instead of with every Vx variable (i.e. V5–V50) ().
Table I. Principal component analysis of DVH parameters.
Evaluation of RP and follow-up
Early RP and late lung fibrosis are different stages of radiation-induced lung injury. Early RP usually occurs one to six months after RT [Citation5], whereas late lung fibrosis usually occurs 6–24 months after RT [Citation12]. Because we were interested in early RP, we used six months as the cut-off for diagnosis.
All patients were followed up by their treating radiation oncologist fortnightly in the first three months after RT and then every four weeks until six months after RT. Chest x-ray or CT was performed at each follow-up session. The diagnosis of RP was made by the radiation oncologists and was based on clinical symptoms and CT images. Cases difficult to diagnose were referred to other physicians to exclude other diseases. Patients with a diagnosis of grade ≥ 2 RP received immediate intervention.
As RP is a dynamic, evolutionary process, we graded each case based on the worst clinical features during the six-month follow-up period. Grading was conducted according to Common Terminology Criteria for Adverse Events version 3.0 [Citation13]. Of the 176 patients in this study, 69 developed grade ≥ 2 RP (39.2%) (48 had grade 2 RP [27.3%]), and 21 developed grade ≥ 3 RP [11.9%]). For statistical analysis, patients were grouped based on the development of grades 0–1 and ≥ 2 RP, or grades 0–2 and ≥ 3 RP.
Statistical analysis
The SPSS 12.0 statistical software package (Chicago, IL) was used to analyze the data. Patient characteristics and dose-volume variables were assessed for correlations with the risk for RP. Univariate analysis was done with the independent samples t-test or the χ2-test. Multivariate analysis was done by logistic regression. Principal component analysis was used to fit DVH parameters into an aggregative indicator which was applied to the analysis of RP occurrence. The receiver operating characteristic (ROC) curve was used to identify the best cut points of dose-volume variables with which to assess the risk of RP. P-value < 0.05 was considered statistically significant.
Results
Two patients died of grade 5 RP (at two and six weeks after RT). All other patients were successfully followed up through the end point of six months after RT. RP was frequently observed to occur within one to three months after RT.
Patient characteristics
Patient demographic and tumor characteristics by RP grade are shown in . There were no statistically significant differences in variables for patients with RP grade ≥ 2. The only statistically significant difference for patients with grade ≥ 3 RP was for age: the incidence of grade ≥ 3 RP was significantly higher in age group ≥ 70 years than in age group < 70 years (18.5% vs. 8.1%, p = 0.041). The results of the multivariate logistic regression analysis () suggested that age was the strongest predictor in both the grade ≥ 2 group [odds ratio (OR) = 2.99, p = 0.014) and the grade ≥ 3 RP group (OR = 11.46, p = 0.001). Chemotherapy was a predictor for the grade ≥ 3 group only (OR = 5.38, p = 0.030), and sex was a predictor for the grade ≥ 2 group only (OR = 0.32, p = 0.028).
Table II. Univariate analysis of patient demographic and tumor characteristics and dose-volume factors, by RP grade (N = 176).
Table III. Multivariate logistic regression analysis of factors associated with RP.
Dose-volume factors
In univariate analysis, the dose-volume variables of DVH-CV of lung, DVH-CV of heart, MLD, MHD, and PTV/Lung were all significantly associated with grade ≥ 2 RP and grade ≥ 3 RP (p < 0.0001 for each comparison) (). The results of the multivariate logistic regression analysis suggested that MLD was the strongest predictor for both the grade ≥ 2 group (OR = 1.32) and the grade ≥ 3 group (OR = 1.55) and that PTV/Lung was a predictor for the grade ≥ 2 group only (OR = 3E + 011). (p < 0.0001 for each comparison).
Comparison of RP-related dose-volume variables in patients with grade ≥ 2 RP or grade ≥ 3 RP corrected for age
For patients who developed grade ≥ 2 or grade ≥ 3 RP, the mean values of MLD and PTV/Lung in the age ≥ 70 group were significantly lower than those in the age < 70 group (p < 0.05 for each comparison). There were no significant differences in MHD, DVH-CV of lung, or DVH-CV of heart between the two groups ().
Table IV. RP-related dose-volume variables in patients with grade ≥ 2 or grade ≥ 3 RP, corrected for age.
Comparison of RP rate by cut point of dose-volume variables
Using ROC curves, we assessed the cut points of dose-volume variables (including MLD and PTV/Lung) to assess the risk of RP. Regardless of whether patients had grade ≥ 2 RP or grade ≥ 3 RP, for both patients aged ≥ 70 years and those aged < 70 years, the incidence of RP increased significantly when MLD and PTV/Lung were higher than their cut points (p ≤ 0.002 for each comparison) ( and ). In patients who developed grade ≥ 3 RP, the cut points of MLD and PTV/Lung were noticeably lower in age group ≥ 70 than in age group < 70 (MLD: 16.06 Gy vs. 22.71 Gy; PTV/Lung: 0.119 Gy vs. 0.157 Gy).
Table V. Comparison of RP rate by cut point of dose-volume variables (age ≥ 70) (n = 65).
Table VI. Comparison of RP rate by cut point of dose-volume variables (age＜70) (n = 111).
Discussion
In this prospective study, we identified age, receipt of chemotherapy, MLD, PTV/Lung, and sex as predictors of grade ≥ 2 or grade ≥ 3 RP in patients with non-small cell carcinoma. Age was a predictor for both grade ≥ 2 RP and grade ≥ 3 RP. Other researchers have also reported age to be correlated with RP. Parashar et al. [Citation7] showed that patients aged 60–70 years had a significantly increased risk of grade ≥ 2 RP compared with younger patients. Wang et al. [Citation14] showed that age > 65 years was significantly associated with symptomatic radiation-induced lung toxicity. However, neither of these research groups analyzed the mechanism of the association. With multivariate analysis, we found that age was an independent predictor for both grade ≥ 2 RP and grade ≥ 3 RP. The incidence of grade ≥ 3 RP was nearly three times higher in patients aged in ≥ 70 years than in younger patients. To determine the reason for this discrepancy by age group, we compared dose-volume parameters between the two groups and found that, for both grade ≥ 2 RP and grade ≥ 3 RP, the mean values of MLD and PTV/Lung were significantly lower in older patients than in younger patients. These results suggested that the higher risk of RP among older patients is associated with lower lung tolerance of radiation.
In our study, chemotherapy was a predictor for grade ≥ 3 RP only. Chemotherapy had been previously reported to be correlated with severe RP. Wang et al. [Citation1] reported a higher incidence of grade ≥ 3 RP (32%) in patients who had been treated with concurrent chemoradiotherapy than those who had not. Onishi et al. [Citation15] showed that concurrent docetaxel chemotherapy led to a higher risk of grade ≥ 3 RP (47%) and that survival rate was not satisfactory due to severe RP.
Results from studies of the correlation between PTV and the risk of RP have been inconsistent. Sunyach et al. [Citation16] showed that a PTV > 200 cm3 was a risk factor for RP, whereas Huang et al. [Citation17] found no correlation between tumor size and RP. We reasoned that this discrepancy might be related to differences in total lung volume of different patients; that is, a lower percentage of lung would be irradiated in patients with a larger lung volume and a higher percent of lung would be irradiated in patients with a smaller lung volume when their PTVs were equal. For this reason, we assessed the correlation between RP and the ratio of PTV to total lung volume (PTV/Lung) instead of PTV.
MLD was a predictor for both grade ≥ 2 and grade ≥ 3 RP in our study. Other studies have also identified dose-volume predictors, such as Vx [Citation1–3,Citation6,Citation10], MLD [Citation3–5], and NTCP [Citation6] for grade ≥ 2 or grade ≥ 3 RP (). A single DVH parameter (Vx) was usually used by other researchers to investigate its correlation with RP. In previous studies [Citation1–3,Citation6]. In this study, we used a more comprehensive measure (DVH-CV) instead of a single DVH variable to analyze its correlation with RP. Both DVH-CV of lung and DVH-CV of heart were significantly associated with the risk of grade ≥ 2 RP or grade ≥ 3 RP in univariate analysis.
Table VII. Study comparison of RP.
Several recent reports have shown that the heart undergoing overly radiation increases the risk of RP. The Groningen group studied the effect of combined heart and lung irradiation on post-RT lung function [Citation18,Citation19] and showed that the severity of respiratory dysfunction after partial thoracic radiation was partially governed by an interaction between pulmonary and cardiac functional deficits. The tolerance dose for early lung function damage depended not only on the irradiated lung region but also the concomitant irradiation of the heart, which severely reduced the tolerance of the lung. Huang et al. [Citation16] showed an important correlation between heart DVH parameters and the incidence of RP and identified heart V65 as an important predictor of RP. In the current study, we also found a correlation between both DVH-CV of heart and MHD and the risk of grade ≥ 2 or ≥ 3 RP in univariate analysis.
Dose-volume parameters are clearly associated with the risk of RP. Most scholars have been interested in finding the best predictors of RP among dose-volume parameters, but a few have explored the threshold values of reported predictors. Barriger et al. [Citation4] used a cut-off point near the median values of dosimetric variables to assess the risk of RP and found that only MLD ≥ 18 Gy was correlated with grade ≥ 2 RP. In our more detailed study using ROC curves, we found that the dose-volume parameters of MLD and PTV/Lung, which were the predictors of high-grade RP in the multivariate analysis, were notably lower in patients aged ≥ 70 years than in patients aged < 70 years.
In our study, sex and PTV/Lung were predictors for grade ≥ 2 RP only in our analysis. Other studies have reported these to be predictors of grade ≥ 2 RP [Citation9,Citation15], but not of grade ≥ 3 RP [Citation6,Citation8]. In previous reports, there were also different results when grade ≥ 2 or grade ≥ 3 was used for evaluating RP. Chemotherapy [Citation4,Citation7,Citation9], age [Citation7], sex [Citation9], V30 [Citation3]and tobacco use [Citation10] have been identified as predictors for grade ≥ 2 RP only; pulmonary emphysema [Citation2], performance status [Citation8], forced expiratory volume in 1 second [Citation8], V5 [Citation1], and NTCP [Citation6] have been identified as predictors for grade ≥ 3 RP only (). Both our study and the work of other researchers suggest that factors associated with the risk of grade ≥ 2 RP are not necessarily predictors of grade ≥ 3 RP.
In conclusion, the predictors of RP were not completely consistent for grade ≥ 2 RP and grade ≥ 3 RP. That is, factors associated with the risk of grade ≥ 2 RP were not necessarily predictors of grade ≥ 3 RP. In addition, the risk of severe RP was higher among patients ≥ 70 years old than among younger patients. The mechanism for this difference might be related to the lower lung tolerance to radiation in older patients.
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References
- Wang S, Liao Z, Wei X, Liu H, Tucker S, Hu C, et al. Analysis of clinical and dosimetric factors associated with treatment-related pneumonitis (TRP) in patients with non-small-cell lung cancer (NSCLC) treated with concurrent chemotherapy and three-dimensional conformal radiotherapy (3D-CRT). Int J Radiat Oncol Biol Phys 2006;66:1399–407.
- Kimura T, Togami T, Takashima H, Nishiyama Y, Ohkawa M, Nagata Y. Radiation pneumonitis in patients with lung and mediastinal tumours: A retrospective study of risk factors focused on pulmonary emphysema. Br J Radiol 2012;85: 135–41.
- Fay M, Tan A, Fisher R, Mac Manus M, Wirth A, Ball D. Dose-volume histogram analysis as predictor of radiation pneumonitis in primary lung cancer patients treated with radiotherapy. Int J Radiat Oncol Biol Phys 2005;61:1355–63.
- Barriger RB, Fakiris AJ, Hanna N, Yu M, Mantravadi P, Mcgarry RC. Dose-volume analysis of radiation pneumonitis in non-small-cell lung cancer patients treated with concurrent cisplatinum and etoposide with or without consolidation docetaxel. Int J Radiat Oncol Biol Phys 2010;78:1381–6.
- Kim TH, Cho KH, Pyo HR, Lee JS, Han JY, Zo JI, et al. Dose-volumetric parameters for predicting severe radiation pneumonitis after three-dimensional conformal radiation therapy for lung cancer. Radiology 2005;235:208–15.
- Shi A, Zhu G, Wu H, Yu R, Li F, Xu B. Analysis of clinical and dosimetric factors associated with severe acute radiation pneumonitis in patients with locally advanced non-small cell lung cancer treated with concurrent chemotherapy and intensity-modulated radiotherapy. Radiat Oncol 2010;12:35.
- Parashar B, Edward A, Mehta R, Pasmantier M, Wernicke AG, Sabbas A, et al. Chemotherapy significantly increases the risk of radiation pneumonitis in radiation therapy of advanced lung cancer. Am J Clin Oncol 2011;34:160–4.
- Robnett TJ, Machtay M, Vines EF, McKenna MG, Algazy KM, McKenna WG. Factors predicting severe radiation pneumonitis in patients receiving definitive chemoradiation for lung cancer. Int J Radiat Oncol Biol Phys 2000;48: 89–94.
- Dang J, Li G, Lu XB, Yao L, Zhang S, Yu Z. Analysis of related factors associated with radiation pneumonitis in patients with locally advanced non-small-cell lung cancer treated with three-dimensional conformal radiotherapy. J Cancer Res Clin Oncol 2010;136:1169–78.
- Jin H, Tucker SL, Liu HH, Wei X, Yom SS, Wang S, et al. Dose-volume thresholds and smoking status for the risk of treatment-related pneumonitis in inoperable non-small cell lung cancer treated with definitive radiotherapy. Radiother Oncol 2009;91:427–32.
- Semrau S, Bier A, Thierbach U, Virchow C, Ketterer P, Fietkau R. Concurrent radiochemotherapy with vinorelbine plus cisplatin or carboplatin in patients with locally advanced non-small-cell lung cancer (NSCLC) and an increased risk of treatment complications: Preliminary results. Strahlenther Onkol 2003;179:823–31.
- Gravs PR, Siddiqui F, Anscher MS, Movsas B. Radiation pulmonary toxicity: From mechanisms to management. Semin Radiat Oncol 2010;20:201–7.
- Trotti A, Colevas AD, Setser A, Rusch V, Jaques D, Budach V, et al. CTCAE v3.0: Development of a comprehensive grading system for the adverse effects of cancer treatment. Semin Radiat Oncol 2003;13:176–81.
- Wang JB, Cao JZ, Yuan SH, Wei J, Douglas AB, Dai JR, et al. Poor baseline pulmonary function may not increase the risk of radiation-induced lung toxicity. Int J Radiat Oncol Biol Phys Epub 2012 Jul 24.
- Onishi H, Kuriyama K, Yamaguchi M, Komiyama T, Tanaka S, Araki T, et al. Concurrent two-dimensional radiotherapy and weekly docetaxel in the treatment of stage III non-small cell lung cancer: A good local response but no good survival due to radiation pneumonitis. Lung Cancer 2003; 40:79–84.
- Sunyach MP, Falchero L, Pommier P, Perol M, Arpin D, Vincent M, et al. Prospective evaluation of early lung toxicity following three-dimensional conformal radiation therapy in non-small-cell lung cancer: Preliminary results. Int J Radiat Oncol Biol Phys 2000;48:459–63.
- Huang EX, Hope AJ, Lindsay PE, Trovo M, El Naqa I, Deasy JO, et al. Heart irradiation as a risk factor for radiation pneumonitis. Acta Oncol 2011;50:51–60.
- Van Luijk P, Faber H, Meertens H, Schippers JM, Langendijk JA. The impact of heart irradiation on dose-volume effects in the rat lung. Int J Radiat Oncol Biol Phys 2007;69: 552–9.
- Van Luijk P, Novakova-Jiresova A, Faber H, Steneker MN, Kampinga HH, Meertens H, et al. Relation between radiation-induced whole lung functional loss and regional structural changes in partial irradiated rat lung. Int J Radiat Oncol Biol Phys 2006;64:1495–502.