PII S0360-3016(00)00783-5

CLINICAL INVESTIGATION Lung

RADIATION PNEUMONITIS IN LUNG CANCER PATIENTS:A RETROSPECTIVE STUDY OF RISK FACTORS AND THE

LONG-TERM PROGNOSIS

AKIRA INOUE, M.D.,*† HIDEO KUNITOH, M.D.,* IKUO SEKINE, M.D.,* MINAKO SUMI, M.D.,‡

KOICHI TOKUUYE, M.D.,‡ AND NAGAHIRO SAIJO, M.D.‡

Departments of *Medical Oncology and†Diagnostic Radiology, and‡Division of Radiation Oncology, National Cancer Center Hospital,Tokyo, Japan

Purpose: To retrospectively evaluate the risk factors for acute radiation pneumonitis (RP) and long-termprognosis of patients with lung cancer treated by thoracic radiotherapy.Methods and Materials: Of the 256 lung cancer patients who underwent definitive thoracic radiotherapy betweenJune 1988 and May 1998, the 191 patients who were capable of being evaluated were divided into three groupsaccording to the grade of RP. RP was defined as “severe,” when it caused severe clinical symptoms, such asintractable cough, dyspnea at rest, and the need for oxygen or steroid therapy. The definition was made by usinga modification of the Radiation Therapy Oncology Group and the European Organization for Research andTreatment of Cancer acute radiation morbidity scoring criteria. Factors that influenced the incidence of severeRP were assessed by using the Mantel–Haenszelx2 test in the univariate analysis and the logistic regression testin the multivariate analysis. Survival rates was calculated by using the Kaplan–Meier method, and thep valuesindicating the significance of differences between the RP groups were calculated by the log–rank test.Results: Of the 94 patients (49%) who experienced clinical RP in this study, the RP was mild in 69 (36%) andsevere in 25 (13%) patients. The 3-year survival rates of the patients who experienced no, mild, and severe RPwere 33.4%, 38.2%, and and 0%, respectively, and the survival rate of the patients who experienced severe RPwas significantly poorer than the other two groups combined (p 5 0.0028). The incidence of severe RP did notcorrelate with any of the baseline patient characteristics, radiotherapeutic factors, or chemotherapeutic vari-ables. Two clinical risk factors were identified from medical records before radiotherapy: low PaO2 (< 80 torr)and high C-reactive protein (CRP) (> 1.0 ng/mL). Both of them were significantly related to the development ofsevere RP in the univariate analysis (p 5 0.004 and 0.013, respectively), and low PaO2 remained a significant riskfactor in the multivariate analysis (p 5 0.034). Multivariate analysis also revealed the occurrence of severe RPto be the most important factor determining poor survival (p 5 0.0065). There was no significant difference insurvival rate according to whether the patients had been treated with corticosteroids.Conclusion: Mild and severe RP occurred in 69 (36%) and 25 (13%), respectively, of 191 lung cancer patientswho had undergone irradiation of the chest. Only severe RP was an adverse prognostic factor. Low PaO2 (< 80torr) before radiotherapy was a significant risk factor predictive of severe RP. The role of corticosteroids in RPcould not be accurately determined. © 2001 Elsevier Science Inc.

Lung cancer, Radiation pneumonitis, Risk factor, Prognosis, Corticosteroid.

INTRODUCTION

Lung cancer has become the leading cause of cancer deathsin both genders in both the United States (1) and Japan.Thoracic radiation therapy remains the mainstay of treat-ment in patients with locally advanced or medically inop-erable non–small-cell lung cancer (NSCLC), which ac-counts for 30–40% of the patient population. Thoracicradiotherapy has also been recognized as an indispensabletreatment modality for limited-stage small-cell lung cancer(SCLC) (2, 3).

However, irradiation of the thorax is not without harmfulside effects, and radiation pneumonitis (RP) is one of themost important and serious complications of this treatment.RP can lead to severe respiratory dysfunction and evendeath, and some patients may develop chronic respiratoryinsufficiency as a result of pulmonary fibrosis (4–7).

Previous reports have identified several risk factors forRP, including low performance status (PS) (8), low pulmo-nary function (8), history of smoking (8), changes in plasmaTGF-b levels (9), once-daily radiotherapy dose fraction-ation (10), radiotherapy combined with chemotherapy (10,

Reprint requests to Dr. Akira Inoue, M.D., Department ofMedical Oncology, National Cancer Center Hospital, 5-1-1Tsukiji, Chuo-ku, Tokyo 104-0045, Japan.E-mail: akinoue@gan2.ncc.go.jp

Supported in part by Grants-in-Aid for Cancer Research fromthe Ministry of Health and Welfare and from the Second-termComprehensive 10-year Strategy for Cancer Control.

Accepted for publication 13 July 2000.

Int. J. Radiation Oncology Biol. Phys., Vol. 49, No. 3, pp. 649–655, 2001Copyright © 2001 Elsevier Science Inc.Printed in the USA. All rights reserved

0360-3016/01/$–see front matter

649

11) and larger radiation doses (. 2.67 Gy) per fraction (12).However, large radiation dosages are no longer commonlyused in clinical practice, and the addition of systemic che-motherapy has been demonstrated to prolong survival inNSCLC patients who receive radiotherapy (13, 14), al-though exacerbation of pulmonary toxicity remains a majorconcern. Moreover, the long-term prognosis of patients withRP remains to be clearly defined.

In this study, we attempted to retrospectively evaluate thelong-term prognosis of patients who experienced acute RPand to identify factors that might allow prediction of RP.

METHODS AND MATERIALS

Evaluation of radiation pneumonitisA simplified set of toxicity criteria (Table 1) modified

from the Radiation Therapy Oncology Group (RTOG) andthe European Organization for Research and Treatment ofCancer (EORTC) acute radiation morbidity scoring criteriawas used to grade the severity of RP (15). Symptoms relatedto RP were obtained from patient medical records. When-ever a patient had clinical symptoms inconsistent with theRTOG score, the highest score was used to define the RPclass. Bacterial infection had largely been excluded on thebasis of negative sputum cultures and the clinical symp-toms. Patients with radiographic evidence of RP but noclinical symptoms was classified as “No RP.”

PatientsWe performed this retrospective analysis using a database

of 256 lung cancer patients who had undergone definitivethoracic radiotherapy in the Division of Radiation Oncologyof the National Cancer Center Hospital between June 1988and May 1998, 191 of the patients met the following inclu-sion criteria: (1) histologically or cytologically proven lungcancer without prior treatment, including by surgical resec-tion, radiation, or chemotherapy; (2) locally advanced ormedically inoperable cancer treated by radiotherapy, with orwithout chemotherapy, that was designed with a curativeintent to include the hilar and mediastinal lymph nodes inthe fields; and (3) full access to medical records with at leastthree month of follow-up after the completion of radiother-

apy, particularly with regard to administration of radiother-apy and its toxic effects. Of these, 171 patients (89.5%)were followed for more than 1 year. We excluded patientswho had undergone thoracic radiotherapy for any reasonswith a total dose under 40 Gy, even if treatment had beenstarted with curative intent, because radiation toxicity to thelung in such patients would be little problem. None of thepatients in our database who had been irradiated with under40 Gy developed clinical radiation pneumonitis.

Patient characteristics and the medical condition of pa-tients before radiotherapy, including the results of labora-tory respiratory function tests, are shown in Table 2. Weused the Eastern Cooperative Oncology Group (ECOG) PScriteria. We chose arbitrary cutoff points for the continuousvariables. The normal limits of the laboratory data in ourhospital were used as the range; mean6 2 SD and C-reac-tive protein (CRP) values below 1.0 ng/mL are considerednormal in our hospital. Respiratory function was consideredabnormal when either %vital capacity (VC) was less than80% or forced expiratory volume in one second (FEV1.0)%was less than 70.

TreatmentRadiotherapy, which ranged in energy from 6 to 21 MVs,

was delivered with linear accelerators. The volume treatedwas the gross tumor plus a 2-cm margin and the ipsilateralhilar and mediastinal lymph nodes. Whenever a patientunderwent radiotherapy with concurrent chemotherapy, thevolume was decided on the basis of the prechemotherapytumor margin, whereas the volume in the patients whounderwent radiotherapy after chemotherapy was based onthe postchemotherapy tumor margin. The total radiationdose was 40–80 (median, 58.5) Gy for NSCLC and 40–50(median, 47.5) Gy for SCLC. Whereas 121 of the 127(95.3%) NSCLC patients were treated with over 50 Gy, and61 of 64 (95.3%) SCLC patients were treated with over 45Gy, the others did not complete radiotherapy for variousreasons, such as disease progression or patient refusal. Themedian effective field size was 160.2 cm2 (range, 24.0–418.0 cm2). In 165 patients, treatment was performed oncedaily with 1.8–2 Gy per fraction, and 26 patients weretreated twice daily with 1.5 Gy. Each fraction was admin-

Table 1. Clinical radiation pneumonitis criteria (modified RTOG/EORTC radiation morbidity scoring criteria)

Class No RP Milder RP Severe RP

RTOG score 0 1 or 2 3 or 4

Clinical symptoms Asymptomatic Mild symptoms, such ascough, dyspnea on exertion

orPersistent cough requiringnarcotic, antitussive agents/dyspnea with minimaleffort but not at rest

Severe cough unresponsive to narcoticantitussive agents, dyspnea at rest/clinicalradiological evidence of acutepneumonitis/intermittent oxygen orsteroids may be required

orSevere respiratory insufficiency/continuous oxygen or assisted ventilation

RT 5 radiation pneumonitis.

650 I. J. Radiation Oncology● Biology ● Physics Volume 49, Number 3, 2001

Table 2. Risk factors for development of severe RP assessed by using univariate and multivariate analysis

Variable

No. of patients

p valueNo/Mild RP Severe RP

166 (87%) 25 (13%) univariate multivariate

Patient characteristicsGender

Male 133 (86%) 21 (14%) 0.648 NSFemale 33 (89%) 4 (11%)

Age median 63.9 (range, 36–86), 70 years 123 (87%) 19 (13%) 0.839 NS. 70 years 43 (88%) 6 (12%)

PS0–1 149 (88%) 21 (12%) 0.392 NS2–3 17 (81%) 4 (19%)

Histological diagnosisSCLC 59 (92%) 5 (8%) 0.126 NSNSCLC 107 (84%) 20 (16%)

Stage1 or 2 30 (91%) 3 (9%) 0.455 NS3 or 4 136 (86%) 22 (14%)

SmokingNever 17 (94%) 1 (6%) 0.317Yes 148 (86%) 24 (14%)

RadiotherapyRT total dose median 54.8 (range, 40–80)

, 60 Gy 98 (88%) 14 (12%) 0.774 NS. 60 Gy 68 (86%) 11 (14%)

RT volume median 145.8 (range, 24.0–380.3), 100 cm2 36 (90%) 4 (10%) 0.516 NS. 100 cm2 130 (86%) 21 (14%)

FractionationOnce-daily 143 (87%) 22 (13%) 0.801Twice-daily 23 (88%) 3 (12%)

Dose per fraction, 2.0 Gy 26 (90%) 3 (10%) 0.635 NS2.0 Gy 140 (86%) 22 (14%)

ChemotherapyYes 109 (89%) 14 (11%) 0.348 NSNo 57 (84%) 11 (16%)

Timing ofradiotherapy

Concurrent 51 (93%) 4 (7%) 0.400Sequential 53 (84%) 10 (16%)Prior to Cx 5 (100%) 0 (0%)

Anticancer drugsCDDP

Yes 96 (89%) 12 (11%) 0.356No 70 (84%) 13 (16%)

VP-16Yes 59 (92%) 5 (8%) 0.126No 107 (84%) 20 (16%)

VDSYes 35 (88%) 5 (12%) 0.901No 131 (87%) 20 (13%)

MMCYes 20 (80%) 5 (20%) 0.273No 146 (88%) 20 (12%)

CPT-11Yes 14 (93%) 1 (7%) 0.444No 152 (86%) 24 (14%)

(Continued)

651Radiation pneumonitis in lung cancer● A. INOUE et al.

istered for 5 days a week, and the daily fractions in the lattergroup were administered at intervals of 6 h.

Chemotherapy was administered to 123 patients; 63 hadundergone induction chemotherapy followed by radiation,55 had received concurrent chemoradiotherapy, and 5 hadbeen treated with adjuvant chemotherapy after radiotherapy.

The decisions to administer corticosteroids to the RPpatients were based on the following criteria: (1) severesymptoms refractory to usual medication and/or a rapidlyprogressive radiographic appearance with respiratory symp-toms, and (2) reasonable exclusion of severe bacterial in-fection based on negative sputum cultures and the absenceof clinical symptoms.

Statistical analysisWe assessed risk factors for the development of severe

RP only because it is important to identify high-risk patientsrequiring highly intensive clinical management. The differ-ences in variables between the severe RP and the othergroups were evaluated by using the Mantel–Haenszelx2 testin the univariate analysis and the logistic regression test inthe multivariate analysis. We included some variables in themultivariate analysis that had been suggested as prognosticfactors for radiation pneumonitis in previous studies, suchas PS, disease extent (stage), radiation total dose, dose perfraction, and combination with chemotherapy. We also in-cluded preradiotherapy laboratory data (WBC [white bloodcell count], LDH [lactic acid dehydrogenase], CRP, PaO2),tumor histology, age, and gender in the analysis. The overall

survival of the patients was calculated by using the Kaplan–Meier method, andp values indicating the significance ofdifferences between RP groups were calculated by the log–rank test. In the censored cases among the living patients,we used the date of last contact on their medical records asthe point of censor.

Multivariate analysis by the stepwise procedure was usedto assess the prognostic significance of several variables,including age, gender, stage, PS, preradiotherapy laboratorydata, and the RP grade. All statistical analyses in this studywere carried out with SAS programs (SAS Institute Inc.,Cary, NC). A p value of less than 0.05 was consideredstatistically significant.

RESULTS

Prognosis of patients with radiation pneumonitisThe 191 patients were divided into three groups: 97

(51%) with no RP (RP-free), 69 (36%) with mild RP, and 25(13%) with severe RP. We graded radiation pneumonitisbased on the clinical symptoms alone, as shown in Table 1,however, there were very few discrepancies between theradiographic appearance and the clinical manifestations.Figure 1 shows the survival data of these patients accordingto our classification of RP. Although the survival curve ofthe patients in the mild RP group was similar to that of theno RP group (data not shown), the severe RP group had asignificantly poorer survival rate (p 5 0.0028). The 3-yearsurvival rates of the no RP group, the mild RP group, and

Table 2. Risk factors for development of severe RP assessed by using univariate and multivariate analysis(Cont’d)

Variable

No. of patients

p valueNo/Mild RP Severe RP

166 (87%) 25 (13%) univariate multivariate

Clinical status before RTWBC

, 10000 141 (87%) 22 (13%) 0.735 NS. 10000 24 (89%) 3 (11%)

LDH, 450 120 (87%) 18 (13%) 0.902 NS. 450 44 (86%) 7 (14%)

CRP, 1.0 113 (91%) 11 (9%) 0.013 NS (0.141). 1.0 50 (78%) 14 (22%)

PaO2

. 80 89 (95%) 5 (5%) 0.004 0.034, 80 71 (81%) 17 (19%)

Respiratory functionNormal 76 (86%) 12 (14%) 0.838Abnormal 56 (88%) 8 (12%)

%DLCO. 60 112 (88%) 15 (12%) 0.182, 60 13 (76%) 4 (24%)

Abbreviations:RP 5 radiation pneumonitis; PS5 performance status; RT5 radiotherapy; Cx5 chemotherapy; CDDP5 cisplatin;VP-165 etoposide; VDS5 vindesine; MMC5 mitomycin C; CPT-115 irinotecan; WBC5 white blood cell count; LDH5 lactic aciddehydrogenase; CRP5 C-reactive protein; DLCO5 diffusing capacity of the lung for carbon monoxide.

652 I. J. Radiation Oncology● Biology ● Physics Volume 49, Number 3, 2001

the severe RP group were 33.4%, 38.2% and 0%, respec-tively. Onset of RP before the completion of radiotherapywas observed in 1 of the 69 patients in the mild RP groupand in 3 of the 25 patients in the severe RP group (p 50.025). Radiotherapy could be resumed and completed inthe patient with mild RP, but not in any of the 3 patientswith severe RP. There were 104 deaths during follow-up.The causes of death were cancer (55 patients), respiratoryfailure related directly to RP, or to complicating pneumonia(3 patients), and other (8 patients), including bacterial pneu-monia, acute myocardial infarction, cerebral infarction, andunknown (38 patients) (which occurred after at least 3months of follow-up at our hospital). There were 3 RP-related deaths in the severe RP group but none in the mildRP group. The causes of death among the patients withsevere RP and patients with no or mild RP were not signif-icantly different.

Risk factors for severe radiation pneumonitisTable 2 shows the relationship between baseline patient

characteristics such as sex, age, performance status, histo-logic tumor type, stage of disease, and smoking history, aswell as with radiotherapy factors, such as the total dose,field volume, fractionation mode and dose per fraction, andthe incidence of severe RP. None of the background patientcharacteristics or radiotherapy factors (including linear ac-celerator energy) were significantly correlated with the in-cidence of severe RP. Nor were there any significant cor-relations between the chemotherapy variables, including useof various anti-cancer drugs, such as cisplatin (CDDP),etoposide (VP-16), vindesine (VDS), irinotecan (CPT-11),and mitomycin C (MMC), or the timing of RT, and theincidence of severe RP.

We detected two significant risk factors for severe RPamong the baseline laboratory data. The incidence of severeRP was significantly higher among the patients with a lowPaO2 (, 80 torr) and among the patients with a high CRP

(. 1.0 ng/mL) (p 5 0.004 and 0.013, respectively). Multi-variate analysis also showed a significant correlation be-tween the incidence of severe RP and low PaO2 (p 50.034).

Table 3 shows the results of multivariate analysis con-ducted to determine the effect of RP grade on survival aftercorrection for other prognostic factors, such as age, sex,stage, PS, and pre-radiotherapy laboratory data. The occur-rence of severe RP remained a poor prognostic factor (p 50.0065).

Efficacy of corticosteroids for RPWe evaluated the effects of corticosteroids by making

comparisons between the group that received corticoste-roids and the group that did not received corticosteroidsamong the 94 patients who experienced mild or severe RP.Twenty-two of the patients were treated with corticoste-roids, 9 who had mild RP and 13 who had severe RP. Themedian starting corticosteroid dose administered was 27.5mg (range, 10–30 mg) in the mild RP group, and 34.6 mg(range, 20–60 mg) in the severe RP group. There was nosignificant difference in the survival rate according towhether the patients had received corticosteroid therapy(data not shown). Three patients died as a direct result ofRT-related toxicity, and all had received steroid pulse ther-apy during their clinical course, which was defined as treat-ment with over 1 g methyl-prednisolone for a few days. Noother patients had received steroid pulse therapy.

DISCUSSION

We have demonstrated a 49% incidence of clinical RP inlung cancer patients treated with RT. The incidence of RPreported in previous studies ranged from 0% to 58% (8, 10,13, 16–23), and this wide variation probably arose fromdifferences in the criteria for RP evaluation. In our series, 69(36%) patients were classified as having mild RP (RTOG

Fig. 1. Overall survival of lung cancer patients who experiencedradiation pneumonitis (RP). Survival was significantly pooreramong patients with severe RP than among those with no or mildRP. Thep value was calculated by the log–rank test.

Table 3. Multivariate analysis of risk factors related to survival

Variable Waldx2 p valueRiskratio

Radiation pneumonitisNo or mild RP* vs.severe RP 7.394 0.0065 2.320

CRP beforeradiotherapy,1.0* vs. . 1.0 (ng/mL) 3.460 0.0629 1.509

Clinical stageStage 1 or 2* vs.

Stage 3 or 4 2.794 0.0946 1.640PaO2 before

radiotherapy, 80*vs. , 80 (torr) 2.683 0.1014 1.420

Abbreviations:RP5 radiation pneumonitis; CRP5 C-reactiveprotein.

Asterisks indicate the better outcome group.

653Radiation pneumonitis in lung cancer● A. INOUE et al.

Grade 1 or 2) and 25 (13%) as having severe RP (RTOGGrade 3 or 4). The mild RP group in our study may wellhave included patients with symptomatic fibrosis inducedby radiotherapy, who would not have been classified as RPpatients in some other reports.

Few reports have referred to the long-term prognosis ofRP patients. We found no significant difference in survivalbetween patients without RP (RTOG Grade 0) and patientswith mild RP (RTOG Grade 1 or 2), although the prognosisof patients with severe RP (RTOG Grade 3 or 4) wasconsiderably worse than in the other two groups (p 50.0028) (Fig. 1).

Although previous studies have reported an associationbetween several factors and an increased risk of RP (8–12),the relationship between these factors and the grade of RPremains uncertain. In this study, we only assessed riskfactors for severe RP, because only the patients with severeRP were considered to be at a clinical disadvantage.

We were unable to identify any significant risk factors forsevere RP among the patient characteristics, radiotherapyfactors, or chemotherapy variables. This is not surprisingsince the dose fractions used in this study were below 2.67Gy, which has been reported to have the most significantassociation with the incidence of RP (12). In this study, wecompared two groups irradiated with different doses perfaction (,2.0 Gy and 2.0 Gy). In the smaller fraction group,patients were irradiated with 1.5 or 1.8 Gy, whereas patientsin the larger fraction group were all irradiated with 2.0 Gy.The differences between fractions may be too small to haveyielded any significant differences in our small sample.

Although some previous studies have suggested concur-rent chemotherapy as a risk factor for RP, no relationshipwas found between them in this study. The reason for this isdifficult to determine because of the small sample size inour study and because patient selection was probably per-formed before the radiotherapy.

We identified two significant risk factors for severe RPamong the baseline clinical factors assessed before radio-therapy: low PaO2 (, 80 torr) and high CRP (. 1.0ng/mL). These findings may reflect a pre-inflammatory con-dition, especially in the lung parenchyma, that predisposesto radiation-induced lung toxicity. For instance, a high CRPvalue may suggest increased activity of certain inflamma-tory cytokines, such as TGFb, which has been reported tobe a prognostic marker for the development of RP (9). Low%DLCO (, 80%) also tended to be more frequent in thesevere RP group. Although corticosteroids were often usedto treat RP, the optimal timing, dosage, and schedule forminimizing risk and maximizing benefit remains unknown.In our series, there was no significant difference in survivalaccording to whether patients had been treated with corti-costeroids. The standard corticosteroid regimen for RPshould be evaluated in future prospective trials.

In conclusion, mild RP and severe RP were found to haveoccurred in 69 (36%) and 25 (13%), respectively, of the 191lung cancer patients treated with radiation. Severe RP, andnot mild RP, was an adverse prognostic factor for survival.Low PaO2 (, 80 torr) and high CRP (. 1.0 ng/mL) beforeradiotherapy were significant risk factors for the develop-ment of severe RP.

REFERENCES

1. Landis SH, Murray T, Bolden S,et al. Cancer statistics.CACancer J Clin1998;48:6–9.

2. Saijo N. Combined modality therapy for small cell lung can-cer.Oncology1992;49(Suppl. 1):2–10.

3. Murray N, Coy P, Pater JL,et al. Importance of timing forthoracic irradiation in the combined modality treatment oflimited-stage small- cell lung cancer.J Clin Oncol 1993;11:336–344.

4. Morgan GW, Pharm B, Breit SN. Radiation and the lung: Areevaluation of the mechanisms mediating pulmonary injury.Int J Radiat Oncol Biol Phys1995;31:361–369.

5. McDonald S, Rubin P, Phillips TL,et al. Injury to the lungfrom cancer therapy: Clinical syndromes, measurable end-points, and potential scoring systems.Int J Radiat Oncol BiolPhys1995;31:1187–1203.

6. Nakayama Y, Makino S, Fukuda Y,et al. Activation of lavagelymphocytes in lung injuries caused by radiotherapy for lungcancer.Int J Radiat Oncol Biol Phys1996;34:459–467.

7. Rubin P, Johnston CJ, Williams JP,et al. A perpetual cascadeof cytokines postirradiation leads to pulmonary fibrosis.Int JRadiat Oncol Biol Phys1995;33:99–109.

8. Monson JM, Stark P, Reilly JJ,et al. Clinical radiation pneu-monitis and radiographic changes after thoracic radiation ther-apy for lung carcinoma.Cancer1998;82:842–850.

9. Anscher MS, Murase T, Prescott DM,et al. Changes inplasma TGF beta levels during pulmonary radiotherapy as apredictor of the risk of developing radiation pneumonitis.IntJ Radiat Oncol Biol Phys1994;30:671–676.

10. Segawa Y, Takigawa N, Kataoka M,et al. Risk factors fordevelopment of radiation pneumonitis following radiationtherapy with or without chemotherapy for lung cancer.Int JRadiat Oncol Biol Phys1997;39:91–98.

11. Mah K, Keane TJ, Dyk JV,et al. Quantitative effect ofcombined chemotherapy and fractionated radiotherapy on theincidence of radiation-induced lung damage: A prospectiveclinical study.Int J Radiat Oncol Biol Phys1994;28:563–574.

12. Roach M 3rd, Gandara DR, Yuo HS,et al. Radiation pneu-monitis following combined modality therapy for lung cancer:Analysis of prognostic factors.J Clin Oncol 1995;13:2606–2612.

13. Dillman RO, Seagren SL, Propert KJ,et al. A randomized trialof induction chemotherapy plus high-dose radiation versusradiation alone in stage 3 non-small-cell lung cancer.N EnglJ Med1990;323:940–945.

14. Dillman RO, Herndon J, Seagren SL,et al. Improved survivalin stage 3 non-small-cell lung cancer: seven-year follow-up ofCancer and Leukemia Group B (CALGB) 8433 trial.J NatlCancer Inst1996;88:1210–1215.

15. Cox JD, Stetz J, Pajak TF. Toxicity criteria of the RadiationTherapy Oncology Group (RTOG) and the European Organi-zation for Research and Treatment of Cancer (EORTC).Int JRadiat Oncol Biol Phys1995;31:1341–1346.

16. Perry MC, Eaton WL, Propert KJ,et al. Chemotherapy with orwithout radiation therapy in limited small-cell carcinoma ofthe lung.N Engl J Med1987;316:912–918.

17. Turrisi AT, Glover DJ, Mason B. A preliminary report: Con-

654 I. J. Radiation Oncology● Biology ● Physics Volume 49, Number 3, 2001

current twice-daily radiotherapy plus platinum-etoposide che-motherapy for limited small cell lung cancer.Int J RadiatOncol Biol Phys1988;15:183–187.

18. Morton RF, Jett JR, McGinnis WL,et al. Thoracic radiationtherapy alone compared with combined chemotherapy forlocally unresectable non-small cell lung cancer.Ann InternMed 1991;115:681–686.

19. Kies MS, Mira JG, Crowley JJ,et al. Multimodel therapyfor limited small-cell lung cancer: A randomized study ofinduction combination chemotherapy with or without tho-racic radiation in complete responders; And with wide-fieldversus reduced-field radiation in partial responders: ASouthwest Oncology Group study.J Clin Oncol 1987;5:592– 600.

20. Jett JR, Everson L, Therneau TM,et al. Treatment of limited-stage small-cell lung cancer with cyclophosphamide, doxoru-

bicin, and vincristine with or without etoposide: A randomizedtrial of the North Central Cancer Treatment Group.J ClinOncol 1990;8:33–39.

21. McCracken D, Janaki LM, Crowley JJ,et al. Concurrentchemotherapy/radiotherapy for limited small-cell lung carci-noma: A Southwest Oncology Group study.J Clin Oncol1990;8:892–898.

22. Schaake-Koning C, Maat B, Houtte P,et al. Radiotherapycombined with low-dose cis-diammine dichloroplatinum(2)(CDDP) in inoperable non-metastatic non-small cell lung can-cer(NSCLC): A randomized three arm phase 2 study of theEORTC Lung Cancer and Radiotherapy Cooperative Groups.Int J Radiat Oncol Biol Phys1990;19:33–39.

23. Verschoore J, Lagrange J, Boublil J,et al. Pulmonary toxicityof a combination of low-dose doxorubicin and irradiation forinoperable lung cancer.Radiother Oncol1987;9:281–288.

655Radiation pneumonitis in lung cancer● A. INOUE et al.