2446 - Association between Geotemporal PM2.5 and Outcomes in Unresectable NSCLC Treated with Chemoradiotherapy
Presenter(s)
A. Nalin1, T. Xu2, J. K. Bronk2, M. S. Ning2, S. Gandhi2, S. G. Chun2, A. B. Chen2, M. S. O'Reilly2, Q. N. Nguyen2, J. W. Welsh2, J. Y. Chang2, S. H. Lin2, Z. Liao2, Y. He3, and D. C. Qian2; 1Division of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, 2Department of Thoracic Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, 3Department of Epidemiology, UTHealth Houston School of Public Health, Houston, TX
Purpose/Objective(s): Many studies have reported a link between ambient fine particulate matter air pollution (PM2.5) and lung carcinogenesis and cardiopulmonary mortality. The effect of PM2.5 on lung cancer-specific treatment outcomes is not well established.
Materials/Methods: We identified patients with unresectable NSCLC (stages II–IVA and locoregional recurrent disease) who received concurrent chemoradiotherapy (CRT) with or without adjuvant durvalumab at a tertiary academic medical institution between 1/2019 and 6/2023. Using each patient’s home address and treatment start date, average location-specific PM2.5 during the 30 days preceding CRT was derived from daily PM2.5 values recorded by the nearest Environmental Protection Agency measurement site. Patients who lived >50 km from the nearest measurement site were excluded. Following partition at the bottom decile of observed PM2.5 (<6 µg/m3), clinicopathologic characteristics were compared using the ?2 test. PFS and OS were compared using Cox regression.
Results: There were 175 evaluable patients (45% female) with median age 70 (IQR 64–76), median follow-up 16 months (IQR 9–28), and pre-CRT 30-day estimated home PM2.5 ranging from 4.4 to 19.0 µg/m3 across 18 states. The lower- and higher-PM2.5 groups displayed similar clinicopathologic characteristics, though the lower-PM2.5 locations concentrated more in the western U.S. (P<0.01). PFS was shorter among the lower-PM2.5 group (median 13 vs. 29 months; HR 2.06, 95% CI 1.12–3.77, P=0.02) even after multivariable adjustment for sex, age, smoking history, ECOG performance, NSCLC histology, stage, use of proton therapy, and receipt of durvalumab. This association was notably more prominent among patients with squamous cell carcinoma (N=58, HR 2.67, 95% CI 1.16–6.12, P=0.021) than those with non-squamous NSCLC (N=117, HR 1.85, 95% CI 0.83–4.13, P=0.135). PM2.5 was not significantly associated with OS.
Conclusion: Following CRT for unresectable NSCLC in a cohort of patients whose local air quality is cleaner than most of the world, extremely low PM2.5 was independently associated with worse PFS. This paradoxical finding is hypothesis-generating for future studies of environmental influence on the anti-tumor immune response.
Table 1Average PM2.5 | |||
< 6 µg/m3, N=20 (%) | = 6 µg/m3, N=155 (%) | p | |
Sex | 0.64 | ||
Female | 10 (50) | 69 (45) | |
Male | 10 (50) | 86 (55) | |
Age | 0.94 | ||
=60 | 4 (20) | 23 (15) | |
61–70 | 7 (35) | 59 (38) | |
71–80 | 8 (40) | 63 (41) | |
=81 | 1 (5) | 10 (6) | |
Geographic region | <0.01 | ||
Northeast | 0 (0) | 1 (1) | |
Midwest | 0 (0) | 6 (4) | |
South | 15 (75) | 145 (94) | |
West | 5 (25) | 3 (2) | |
Smoking | 0.58 | ||
Never | 2 (10) | 24 (15) | |
Former | 16 (80) | 124 (80) | |
Current | 2 (10) | 6 (4) | |
Unknown | 0 (0) | 1 (1) | |
ECOG | 0.30 | ||
0 | 7 (35) | 55 (35) | |
1 | 13 (65) | 84 (54) | |
2 | 0 (0) | 16 (10) | |
Histology | 0.23 | ||
Adenocarcinoma | 10 (50) | 90 (58) | |
Squamous cell carcinoma | 10 (50) | 48 (31) | |
Other | 0 (0) | 12 (8) | |
Unspecified | 0 (0) | 5 (3) | |
Overall stage | 0.33 | ||
II | 0 (0) | 17 (11) | |
III | 15 (75) | 115 (74) | |
IVA | 1 (5) | 6 (4) | |
Locoregional recurrence | 4 (20) | 17 (11) | |
RT modality | 0.64 | ||
Photon | 14 (70) | 116 (75) | |
Proton | 6 (30) | 39 (25) | |
Adjuvant durvalumab | 1.00 | ||
No | 7 (35) | 55 (35) | |
Yes | 13 (65) | 100 (65) |