Do volume metrics predict pulmonary function changes in lung irradiation

Abstract

Purpose: With conventional doses and field arrangements there is no accurate way to predict the change in pulmonary function, although several volume based metrics do correlate with the risk of pneumonitis. We hypothesized that in a series of patients treated with nonconventional beam arrangements on a dose escalation study for non-small cell lung cancer (NSCLC) that one of these metrics might correlate with late pulmonary function changes as measured by pulmonary function tests (PFTs). Materials and Methods: 42 patients with NSCLC treated on a lung dose escalation study had pre-treatment and post-treatment PFTs 3 or more months after treatment, and were available for evaluation. The study has increased the dose based on the volume of normal lung irradiated by stratifying patients based on the effective volume (Veff) of normal lung irradiated. There was no minimal pulmonary function requirement. Each patient received 2.1 Gy per day in a single fraction, with lung dose inhomogeneity correction. The PTV included the gross tumor plus a 1 cm expansion. The median dose delivered was 76.9 Gy (range 63 to 102.9 Gy). Field arrangements were not specified and varied widely. The median number of fields was 4 with a range from 1 to 6. This cohort had a median age of 66.5 yrs (range 40 to 80). Eight patients received induction chemotherapy with cisplatin and vinorelbine. All patients had the volume of normal lung treated to greater than 20 Gy (V20), the Veff and mean lung dose (MLD) calculated for both lungs. A linear regression analysis was done to determine if there was a correlation between the metrics and changes in FEV1, % Predicted FEV1, and DLCO. Results: The analyzed group had pre-treatment PFTs similar to the entire study group. 42 patients had FEV1 available, 38 had %predicted FEV1 pre-and posttreatment, and 21 had pre- and posttreatment DLCO available. Post-treatment PFTs were measured at a median of 208 days (range 84-471). Median values for Pretreatment PFTs were as follows. FEV1 1.72 (.84to 3.52),%predicted FEV1 65.3 (25 to 117%), DLCO 65.4 (27 to 116). Median values for metrics were Veff .19 (.076 to .43), MLD 15.4 Gy (6.2 to 28.5 Gy, V20 20.6% (0 to 54%). Median posttreatment PFTs were FEV1 1.67 (0.6 to 5.0), % predicted FEV1 60.9 (23 to 119%), DLCO 53.8 (25 to 108). No Patients experienced Grade 3 or 4 pneumonitis. There was no correlation between Veff, V20 or MLD and the change in FEV1, the percent change in FEV1, the change in the % predicted FEV1, the % change in % predicted FEV1, the change in DLCO or the % change in DLCO. Subgroup analyses of patients with and without atelectasis prior to irradiation, induction chemotherapy and stage I or III only patients, were reviewed and there was no correlation between the volume metrics and degree of PFT changes. Conclusions: None of the volume metrics shown to correlate with the risk of pneumonitis were successful in predicting long term changes in pulmonary function as assessed with standard pulmonary function studies in patients treated to high dose using 3D-CRT. A different model and/or parameters are likely necessary to accurately estimate the long-term change in pulmonary function. Tabled 1ParameterMedianRangeDecrease in FEV1.02Decline .9 to increase 3.14%Decrease in FEV12%Decline 45 % to increase 168%%Decrease in %FEV1-2.2%Decline 35 % to increase 56%Decrease in DLCO6.9Decline 9% to increase 31%%Decrease in DLCO10%Decline 28% to increase 29% Open table in a new tab