Virtual Library
Start Your Search
G. Dhami
Author of
-
+
P2.03 - Poster Session/ Treatment of Locoregional Disease – NSCLC (ID 213)
- Event: WCLC 2015
- Type: Poster
- Track: Treatment of Locoregional Disease – NSCLC
- Presentations: 1
- Moderators:
- Coordinates: 9/08/2015, 09:30 - 17:00, Exhibit Hall (Hall B+C)
-
+
P2.03-028 - Functional Lung Imaging with Perfusion SPECT/CT Improves Prediction of Radiation Pneumonitis (ID 2477)
09:30 - 09:30 | Author(s): G. Dhami
- Abstract
Background:
Current standard of care relies on CT-based lung dose-volume parameters to predict risk of pulmonary toxicity from radiotherapy. However, this approach remains imperfect as patients treated with radiotherapy for lung cancer still experience up to 20% clinically significant pneumonitis despite efforts to avoid it. We propose to improve prediction accuracy for radiation pneumonitis through incorporation of functional lung radiation dose parameters defined on ventilation/perfusion SPECT/CT.
Methods:
Pre-treatment [99m]Tc-MAA perfusion and[ 99m]Tc-DTPA ventilation SPECT/CT scans were co-registered to planning CT scans in 12 patients who received thoracic radiotherapy: 11 with lung cancer and 1 with lung metastasis. Five patients were treated with IMRT/3DCRT, 3 SBRT, and 4 proton RT. Two patients had clinical grade 2 and one patient had grade 3 pneumonitis (G2+ PNM) defined by CTCAE v4. Total lung minus GTV (TL-GTV) mean dose, V5Gy, V20Gy, and V30Gy were calculated. Threshold percentages of maximum ventilation and perfusion (10-90%) within TL-GTV defined functional dose-volume regions, from which the mean dose (D~PERF10-90%~, D~VENT10-90%~) and volume fraction of TL-GTV (V~PERF10-90%~, V~VENT10-90%~) were extracted. Mann-Whitney tests were conducted between patients with and without G2+ PNM. Receiver operating characteristic (ROC) curves identified functional dose-volume thresholds that could predict for G2+ PNM status. Logistic regression of G2+ PNM incidence from anatomic and functional dose-volume parameters was modeled. Spearman rank correlation between predictive anatomic and functional dose-volume parameters was calculated.
Results:
Anatomic TL-GTV parameters were significantly higher in G2+ PNM patients following independent testing, particularly mean lung dose (MLD 23 vs. 9 Gy RBE EQD2~α/β=3~, p=0.03) and V20Gy (33% vs. 9%, p=0.03). Perfused lung dose was also higher in patients with G2+ PNM (D~PERF70%~ 18 vs. 3 Gy, p=0.04). Using a cutoff value of MLD>20 Gy, 3/4 patients had G2+ PNM. The addition of mean perfused lung dose D~PERF70%~>17.5 Gy to MLD>20Gy improved specificity, with all 3/3 patients who received high MLD and D~PERF70%~ developing G2+ PNM (See Figure 1). Anatomic and perfused lung metrics were statistically correlated (Spearman R2=0.6, p=0.03). Figure 1. Mean dose to perfused lung vs. anatomic lung for patients with(black circle)/without(gray circle) G2+PNM. Adding perfused lung dose to anatomic lung dose increases specificity for pneumonitis. Radiotherapy plan is highlighted for one false positive case, in which dose to perfused lung is low(white arrow) while dose to anatomic lung is high. Figure 1
Conclusion:
Our data suggests that incorporating perfused lung metrics into radiotherapy planning objectives may improve our ability to predict and mitigate the risk of pneumonitis. Given the limited sample size, further investigation is warranted in a larger population.
