From January 2016 to December 2017, a total of 87 consecutive lung cancer patients with subclinical ILD, aged 48 to 86 years with a median age of 67 years, were enrolled in the study. The tumor stage was determined according to the 8th edition of the Union for International Cancer Control TNM staging system for lung cancer. None of the patients had been diagnosed with interstitial lung disease clinically or via lung biopsy prior to receiving chemotherapy or radiotherapy. None of the patients had received treatment with any of the currently available medicines for ILD. Seventy-six patients had received chemotherapy prior to radiotherapy. The median duration of chemotherapy was two cycles (range, 0–6 cycles). Two patients had previously received epidermal growth factor receptor tyrosine kinase inhibitor. The characteristics of the patients are shown in Table 1.
Table 1 Characteristics of the patients
Factors
|
N(%)
|
Gender
Male
Female
Age(years)
<70
≥70
Pathological types
NSCLC
Adenocarcinoma
Squamous cell carcinoma
Large cell carcinoma
Unclassified NSCLC
SCLC
Tumor stage
I
IIIA
IIIB
IIIC
IV
Postoperative
Chemotherapy
Concurrent
Sequential
|
81(93.1)
6 (6.9)
57(65.5)
30(34.5)
57(65.5)
17(19.5)
24(27.6)
1 (1.1)
15(17.2)
30(34.5)
2 (2.3)
24(27.6)
37(42.5)
3 (3.4)
10(11.5)
11(12.6)
19(21.8)
68(78.2)
|
NSCLC, non-small-cell lung cancer; SCLC, small-cell lung cancer
Table 2 Characteristics of patients who discontinued radiotherapy
No.
|
Age (years)/Gender
|
Smoking history (pack-years)
|
Pathological type/tumor stage
|
Treatment modality
|
Induction chemotherapy regimen
|
Radiotherapy dose (Gy/fraction)
|
Percentage of lung volume affected in subclinical ILD
|
Grade of RP
|
1
|
68/Male
|
40
|
Squamous cell carcinoma/IIIB
|
RT
|
Cisplatin+gemcitabine
|
44.0/22
|
≥25%
|
4
|
2
|
77/Male
|
50
|
Squamous cell carcinoma/IIIB
|
RT
|
Carboplatin+gemcitabine
|
49.5/22
|
≥25%
|
5
|
3
|
67/Male
|
No
|
Squamous cell carcinoma/postoperative stump recurrence
|
RT
|
Cisplatin+gemcitabine
|
24.0/12
|
<25%
|
2
|
4
|
63/Male
|
80
|
Adenocarcinoma/IIIB
|
RT
|
Cisplatin+gemcitabine
|
58.0/29
|
<25%
|
3
|
5
|
65/Male
|
40
|
NSCLC/IIIB
|
RT
|
Cisplatin+vinorelbine
|
38.0/19
|
<25%
|
4
|
6
|
69/Male
|
50
|
Squamous cell carcinoma/IIIB
|
RT
|
Carboplatin+gemcitabine
|
38.0/19
|
≥25%
|
5
|
7
|
52/Male
|
60
|
Adenocarcinoma/IIIB
|
RT
|
Cisplatin+pemetrexed
|
30.0/15
|
<25%
|
2
|
8
|
67/Male
|
No
|
SCLC/IIIB
|
RT
|
Carboplatin+etoposide
|
40.0/20
|
<25%
|
2
|
RP, radiation pneumonitis; ILD, interstitial lung disease; RT, radiation therapy; NSCLC, non-small-cell lung cancer; SCLC, small-cell lung cancer
The median follow-up time was 13.4 months (range, 1.2–33.6 months). RP was observed in 19 (21.8%), 27 (31.0%), 10 (11.5%), three (3.4%), and five (5.7%) patients with grades 1, 2, 3, 4, and 5 RP, respectively. Radiotherapy was discontinued in eight patients because grade ≥2 RP occurred during radiotherapy. Characteristics of the patients for whom radiotherapy was discontinued are shown in Table 2. Five patients developed grade 5 RP; their characteristics are shown in Table 3.
Table 3 Characteristics of patients with grade 5 RP
No.
|
Age (years)/Gender
|
Smoking history (pack-years)
|
Pathological types/Tumor stage
|
Treatment modality
|
Concurrent chemotherapy regimen
|
Induction chemotherapy regimen
|
Radiotherapy dose (Gy/fraction)
|
MLD
(Gy)
|
Percentage of lung volume affected in subclinical ILD
|
Time from treatment to grade 5 RP (months)
|
1
|
59/Male
|
20
|
Adenocarcinoma/postoperative recurrence
|
CCRT
|
Cisplatin+pemetrexed
|
Cisplatin+gemcitabine
|
60.0/30
|
12.12
|
<25%
|
3.3
|
2
|
77/Male
|
50
|
Squamous cell carcinoma/IIIB
|
RT
|
None
|
Carboplatin+gemcitabine
|
49.5/22
|
13.89
|
≥25%
|
4.1
|
3
|
69/Male
|
50
|
Squamous cell carcinoma/IIIB
|
RT
|
None
|
Carboplatin+gemcitabine
|
38.0/19
|
12.19
|
≥25%
|
1.2
|
4
|
71/Male
|
60
|
SCLC/IIIA
|
CCRT
|
Cisplatin+ etoposide
|
Cisplatin+etoposide
|
60.0/30
|
12.65
|
<25%
|
6.8
|
5
|
62/Male
|
30
|
Squamous cell carcinoma/postoperative recurrence
|
CCRT
|
Carboplatin+paclitaxel
|
None
|
60.0/30
|
11.92
|
<25%
|
3.0
|
RP, radiation pneumonitis; MLD, mean lung dose; ILD, interstitial lung disease; RT, radiation therapy; CCRT, concurrent chemoradiotherapy; SCLC, small-cell lung cancer
The cumulative incidence of grades ≥2 and 3 RP at one year was 51.0% and 20.9%, respectively. Although the percentage of lung volume affected in subclinical ILD did not significantly increase the cumulative incidence of grade ≥2 RP (69.4%vs.47.7%, p = 0.085,Fig. 1A.), the cumulative incidence of grade ≥3 RP was significantly higher in patients with subclinical ILD involving ≥25% of lung volume than those with <25% involvement of lung volume (46.1% vs. 16.3%, p = 0.004, Fig. 1B.). Chemotherapy with gemcitabine before radiotherapy did not significantly affect the cumulative incidence of grade ≥2 RP (53.2% vs. 49.4%, p = 0.524, Fig. 1C.), but the cumulative incidence of grade ≥3 RP was significantly higher in patients who had received chemotherapy with gemcitabine in the past than those who had not (32.3% vs. 13.3%, p = 0.023, Fig. 1D.).
Table 4 Correlation between factors and RP by univariate analysis
Factors
|
Grade ≥2 RP(N=45)
|
Grade ≥3 RP(N=18)
|
Grade <2 RP
|
Grade ≥2 RP
|
χ2/T
|
p value
|
Grade <3 RP
|
Grade ≥3 RP
|
χ2/T
|
p value
|
Gender
|
|
|
1.843
|
0.175
|
|
|
0.600
|
0.439
|
Male
|
37
|
44
|
|
|
63
|
18
|
|
|
Female
|
5
|
1
|
|
|
6
|
0
|
|
|
Age (years)
|
|
|
1.291
|
0.256
|
|
|
0.195
|
0.659
|
<70
|
25
|
32
|
|
|
46
|
11
|
|
|
≥70
|
17
|
13
|
|
|
23
|
7
|
|
|
Smoking history
|
|
|
1.931
|
0.165
|
|
|
0.470
|
0.493
|
No
|
15
|
10
|
|
|
21
|
4
|
|
|
Yes
|
27
|
35
|
|
|
48
|
14
|
|
|
Pathological type
|
|
|
0.448
|
0.503
|
|
|
1.510
|
0.219
|
NSCLC
|
29
|
28
|
|
|
43
|
14
|
|
|
SCLC
|
13
|
17
|
|
|
26
|
4
|
|
|
Tumor stage
|
|
|
6.433
|
0.092
|
|
|
1.457
|
0.692
|
I
|
2
|
0
|
|
|
2
|
0
|
|
|
III
|
28
|
36
|
|
|
51
|
13
|
|
|
IV
|
4
|
6
|
|
|
7
|
3
|
|
|
Postoperative
|
8
|
3
|
|
|
9
|
2
|
|
|
Tumor location
|
|
|
4.094
|
0.043
|
|
|
2.419
|
0.120
|
Upper lobe
|
32
|
25
|
|
|
48
|
9
|
|
|
Middle or lower lobe
|
10
|
20
|
|
|
21
|
9
|
|
|
Chemotherapy with gemcitabine in the past
|
|
|
0.033
|
0.856
|
|
|
4.627
|
0.031
|
No
|
26
|
27
|
|
|
46
|
7
|
|
|
Yes
|
16
|
18
|
|
|
23
|
11
|
|
|
Concurrent chemotherapy
|
|
|
2.156
|
0.142
|
|
|
0.076
|
0.782
|
No
|
30
|
38
|
|
|
53
|
15
|
|
|
Yes
|
12
|
7
|
|
|
16
|
3
|
|
|
Distribution of subclinical ILD
|
|
|
1.190
|
0.275
|
|
|
2.000
|
0.157
|
Lateral
|
7
|
4
|
|
|
11
|
0
|
|
|
Bilateral
|
35
|
41
|
|
|
58
|
18
|
|
|
Morphology of subclinical ILD
|
|
|
0.712
|
0.399
|
|
|
2.208
|
0.137
|
No honeycombing
|
34
|
33
|
|
|
56
|
11
|
|
|
Honeycombing
|
8
|
12
|
|
|
13
|
7
|
|
|
Percentage of lung volume affected in subclinical ILD
|
|
|
1.876
|
0.171
|
|
|
4.353
|
0.037
|
<25%
|
38
|
36
|
|
|
62
|
12
|
|
|
≥25%
|
4
|
9
|
|
|
7
|
6
|
|
|
Pulmonary emphysema
|
|
|
0.196
|
0.658
|
|
|
0.082
|
0.775
|
No
|
6
|
8
|
|
|
12
|
2
|
|
|
Yes
|
36
|
37
|
|
|
57
|
16
|
|
|
Total dose (Gy)
|
|
|
0.044
|
0.834
|
|
|
0.878
|
0.349
|
EQD2<60.0
|
13
|
13
|
|
|
19
|
7
|
|
|
EQD2≥60.0
|
29
|
32
|
|
|
50
|
11
|
|
|
Single fraction dose
|
|
|
0.049
|
0.826
|
|
|
0.000
|
1.000
|
2.0Gy
|
38
|
39
|
|
|
61
|
16
|
|
|
>2.0Gy,≤3.0Gy
|
4
|
6
|
|
|
8
|
2
|
|
|
FVC%
|
89.53 ± 14.50
|
92.15 ± 25.83
|
−0.481
|
0.632
|
89.93 ± 22.58
|
94.53 ± 17.75
|
−0.723
|
0.472
|
FEV1/ FVC (%)
|
74.71 ± 9.05
|
74.05 ± 9.89
|
0.277
|
0.782
|
74.57 ± 8.52
|
73.56 ± 12.40
|
0.359
|
0.720
|
MLD (Gy)
|
11.52 ± 3.36
|
12.97 ± 2.45
|
−2.307
|
0.024
|
12.00 ± 3.15
|
13.29 ± 2.11
|
−1.636
|
0.106
|
V5 (%)
|
46.07 ± 12.51
|
48.24 ± 9.54
|
−0.914
|
0.363
|
46.19 ± 11.69
|
51.06 ± 7.25
|
−1.679
|
0.097
|
V10 (%)
|
33.14 ± 9.84
|
35.00 ± 7.20
|
−1.010
|
0.316
|
33.68 ± 9.26
|
35.72 ± 5.04
|
−1.253
|
0.216
|
V20 (%)
|
20.62 ± 6.88
|
21.89 ± 4.46
|
−1.014
|
0.314
|
21.10 ± 6.14
|
21.94 ± 4.05
|
−0.551
|
0.583
|
V30 (%)
|
14.24 ± 5.64
|
16.07 ± 4.08
|
−1.741
|
0.085
|
14.91 ± 5.12
|
16.22 ± 4.22
|
−0.999
|
0.321
|
RP, radiation pneumonitis; NSCLC, non-small-cell lung cancer; SCLC, small-cell lung cancer; FVC, forced vital capacity; FEV1, forced expiratory volume in one second; FVC%, percentage forced vital capacity; EQD2, equivalent dose in 2.0 Gy/(fraction per day); MLD, mean lung dose; V5, percentage of lung volume receiving ≥5 Gy; V10, percentage of lung volume receiving ≥10 Gy; V20, percentage of lung volume receiving ≥20 Gy; V30, percentage of lung volume receiving ≥30 Gy
Table 4 shows the correlations between the risk factors and RP. In the univariate analysis, tumor location (upper lobe vs. middle or lower lobe) and MLD were associated with grade ≥2 RP (p = 0.043 and p = 0.024, respectively). The risk of grade ≥3 RP was higher in patients who had received chemotherapy with gemcitabine in the past and subclinical ILD involving ≥25% of lung volume (p = 0.031 and p = 0.037, respectively).
In the multivariate analysis, MLD≥12.0Gy was a significant risk factor for grade ≥2 RP (p = 0.049). Having received chemotherapy with gemcitabine in the past, V5≥50%, and subclinical ILD involving ≥25% of lung volume were significantly associated with grade ≥3 RP (p = 0.046, p = 0.040, and p = 0.024, respectively). The results of the binary logistic regression analysis for RP are shown in Table 5.
Table 5 Correlation between risk factors and RP using binary logistic regression analysis
Factors
|
Grade ≥2 RP
|
Grade ≥3 RP
|
Odds ratio (OR)
|
95% CI
|
p value
|
OR
|
95% CI
|
p value
|
MLD (≥12.0Gy vs. <12.0Gy)
|
2.480
|
1.006–6.113
|
0.049
|
-
|
-
|
-
|
Tumor located in lower lobe
|
2.311
|
0.898–5.943
|
0.082
|
-
|
-
|
-
|
Chemotherapy with gemcitabine in the past
|
-
|
-
|
-
|
3.209
|
1.018–10.113
|
0.046
|
V5(≥50%vs.<50%)
|
-
|
-
|
-
|
3.429
|
1.056–11.140
|
0.040
|
Percentage of lung volume affected in subclinical ILD ≥25%
|
-
|
-
|
-
|
4.861
|
1.237–19.104
|
0.024
|
RP, radiation pneumonitis; MLD, mean lung dose; V5, percentage of lung volume receiving ≥5Gy; ILD, interstitial lung disease; CI, confidence interval