Therapeutic Irradiation Acute and Chronic Effects on the Lung

*Lawrence B. Marks, University of North Carolina 

Keywords: lung injury, radiation

Therapeutic Irradiation Acute and Chronic Effects on the Lung June 2012 Lawrence B Marks, MD: Department Radiation Oncology, University of North Carolina at Chapel Hill, NC, marks@med.unc.edu Introduction: Radiation therapy is commonly used when treating patients with tumors in/around the thorax. In these settings, radiation-induced acute and late lung injury is relatively common. Indeed, in some settings, thoracic radiation-induced lung injury limits the radiation dose/volume delivered. Endpoints: There are several endpoints generally considered for radiation-induced lung injury including, and thre incidence of injury depends on the endpoint considered (Marks 2000, 2010): Symptoms: This is arguably the most important endpoint, as this is what bothers the patient. However, symptoms are difficult to quantify, and are somewhat subjective. Indeed, several grading systems for toxicity for symptoms will include the intervention (e.g. placement of patient on steroids) in the grading of the toxicity. Therefore, physicians who tend to prescribe steroids more “liberally” will have a higher “complication rate.” Conversely, physicians who tend not to prescribe steroids as often would appear to have a “lesser complication rate.” Therefore, while symptoms are clinically very important, their subjective nature makes them somewhat challenging for studies. They reflect global lung function, but are non-specific (e.g. shortness of breath can occur from non-lung causes such as anemia or heart disease) Pulmonary function tests (PFTs): These are objective quantitative assessments of whole/global lung function and describe things such as the amount of air that can be moved in and out of the lung in a certain amount of time, or the ability of the lung alveoli to achieve gas exchange. Radiographs: A variety of tests, such as computed tomography (CT) or single photon emission computed tomography (SPECT) lung perfusion (or ventilation) imaging, have been used to quantify lung injury. These are very powerful tools to provide objective quantitative assessments of regional injury. However, as they reflect a regional endpoint, rather than a global endpoint, their clinical relevance has been questioned. Our Prospective Study: In approximately 1990, in order to better understand the predictors of radiation-induced lung injury, and the association between these various endpoints, we conducted a prospective clinical trial at Duke University. In this study, patients underwent pre-radiation and serial post-radiation assessments of all three endpoints. Study aims broadly are: Aim #1: To define the dose-response relationship for radiation-induced changes in regional lung function. Here, changes in regional perfusion (per SPECT) are taken as a surrogate for function. Perfusion is generally considered to be a reasonable surrogate for regional gas exchange, since the function of the alveoli-capillary unit is usually perfusion-limited. Sophisticated three-dimensional radiation treatment planning tools and image fusion software were used to allow pre and serial post-RT images to be registered to the three-dimensional dose distribution, and to each other. In this manner, we were able to define the dose-dependent nature of changes in regional perfusion. A similar approach was also taken for imaging of regional tissue density assessed by CT. Aim #2: To relate the ‘sum of the regional injuries’ (assessed by imaging) to changes in global lung function (e.g. symptoms and changes to pulmonary function tests). Since the lung is structured anatomically as a “parallel organ” it’s reasonable to hypothesize that the global function will be related to the sum of the injuries to the various regions within the lung. Other specific aims: Over the years, we have also considered the impact of biological markers (e.g. transforming growth factor beta), pre-treatment lung function, and the degree of cardiac irradiation, on these “lung” endpoints. Broad results: Regional dose response: We have successfully compared pre- vs. serial post-radiation SPECT images in many patients from 3 months to 12 years post-radiation (Marks 1997, Woel 2002, Zhang 2010). We demonstrated a nice dose response relationship between the regional radiation dose and the degree of reduction in regional perfusion. There is almost a linear relationship between regional dose and loss of function between approximately 5 and 50 Gy. (There are uncertainties in the low dose region, < 5-10 Gy) due to inherent normalization challenges in SPECT images). Comparing dose response curves over time, the degree of regional injury is largely manifest by approximately 6-9 months post-radiation. With longer follow-up, there is minimal further change in the degree of RT-associated regional perfusion changes. Relating the ‘sum of the regional injuries’ to changes in global function: Pulmonary function tests: There is always a significant, albeit weak, association between the sum of regional perfusion changes and declines in pulmonary function tests; (p =0.002-0.13) (Fan 2001a, b). These associations become somewhat stronger at longer time points post-RT, and are also stronger if we exclude the subset of patients who have tumor-associated abnormalities in their regional perfusion pre-RT. In other words, there are some patients who have central tumors that are compressing regional blood vessels that cause abnormalities in regional perfusion pre-radiation. After radiation is delivered, there is some regression of this tumor with re-perfusion of the “downstream” lung. Inclusion of these patients in our dose-response analyses is sub-optimal (since there are perfusion changes unrelated to the RT-induced tissue damage). When these patients are excluded from this analysis, the association between the sum of regional injuries and changes in PFTs becomes stronger. Declines in PFT’s were better correlated with the ‘sum of the regional injuries’ than with the mean lung dose (MLD) or percent of lung at = 30 Gy (V30). Using analogous techniques, similar results were obtained at the Netherlands Cancer Institute; NKI (p = 0.001) (Theuws 1999a). Nevertheless, correlations at both Duke and NKI were suboptimal (r ˜ 0.3 – 0.6). Shortness of Breath (SOB): Using modern 3D tools, the rate of RT-associated SOB has been related to metrics from the lung DVH, such as the mean lung dose (MLD), the % of lung receiving doses in excess of certain thresholds (e.g., =20 Gy; termed V20), and the NTCP (normal tissue complication probability). We and others have shown that these metrics can segregate patients into relatively high and low risk groups (e.g. Kwa 1998, Graham 1999, Hernando 2001, Marks 1997, 2010). The exact parameter extracted from the DVH (MLD, V20, V30, or NTCP) is not critical as these are well correlated with each other (Graham 1999; Fan 2001a). However, most of these data were collected retrospectively, and the cut-off values and model/parameters may have been chosen to better fit the observations. Can accurate predictions of symptomatic shortness of breath be made pre-RT? Based on an analysis of 162 patients in our initial cohort (1991-1999), and supporting data from institutions, we defined a multi-parameter predictive model for symptomatic pneumonitis based on: a) traditional DVH’s (dose volume histograms), b) the 3D dose distribution within perfused/functional lung (DFH; dose function histogram), and c) and the pre-RT PFT. We then prospectively tested the accuracy of this model in a subsequent group of patients (1999-2005). 94 patients were enrolled and 55 were evaluable (minimum 6-month followup, free of intra-thoracic tumor recurrence causing dyspnea). The patient demographics were similar to our prior cohort. The model was unable to accurately prospectively segregate high and low risk patients; RP rates 7/20 (30%) and 9/35 (25%) in the high and low-risk cohorts, respectively (p = 0.33). This model was also tested in an independent patient cohort from the NKI and failed to statistically segregate patients into high vs. low risk (4/21 vs. 6/44, p = 0.41, Kocak 2006). These results should NOT be interpreted to mean that the pre-RT PFT’s and DFH-based metrics are unimportant. The number of patients was modest, and hence the power was limited. Rather, it emphasizes the challenges of making these predictions prospectively. If considered retrospectively, these data are consistent with our and other’s prior data. Patients with a high-risk DVH (e.g. a high MLD), do have a higher rate of pneumonitis than patients with more a favorable DVH. Further, a different PFT/DVH/DFH model can be defined retrospectively that better segregates patients into high and low risk groups. Other considerations Biological factors: Several analyses from our group at Duke suggested that the risk of pneumonitis was higher in patients whose TGFb levels increased during RT, compared to those whose TGFb levels declined during RT (Anscher 2001, Fu 2000, 2001, Evans 2004, 2006, Hart 2005). The data on this topic from various institutions is conflicting. Cardiac irradiation: Some of our analyses suggest that the degree of incidental cardiac RT may also play a role in the development of “lung toxicity”, but these data are inconclusive and the data from different institutions inconsistent, in this regard. In Summary, there are presently no good methods to accurately predict a patient’s risk of toxicity (Marks 2003). Many studies have retrospectively related dosimetric lung parameters to symptoms, with sub-optimal results. The limited data regarding the prospective identification of patients at high- vs. low-risk for symptoms was disappointing. Further, the endpoint of shortness of breath may be non-specific for RT-induced lung injury (Kocak 2005). Nevertheless, there is a correlation between the sum of regional lung perfusion changes and changes in global “lung” function. References 1) Evans ES, Kocak Z, Zhou SM, Dahn DA, Huang H, Hollis DR, Light KL, Anscher MS, Marks LB. Does transforming growth factor-beta1 predict for radiation-induced pneumonitis in patients treated for lung cancer? Cytokine, 35(3-4):186-192, 2006. 2) Fan M, Marks LB, Hollis D, Bentel GC, Anscher MS, Sibley GS, Coleman RE, Jaszczak RJ, Munley MT. Can we predict radiation-induced changes in pulmonary function tests based on the sum of predicted regional dysfunction? J. Clin. Oncol. 19:543-550; 2001a. 3) Fan M, Marks LB, Lind P, Hollis D, Woel RT, Bentel GC, Anscher M, Shafman TD, Coleman RE, Jaszczak RJ, Munley MT: Relating Radiation-induced regional lung injury to changes in pulmonary function tests. Int. J. Radiat. Oncol. Biol. Phys. Int J Radiat Oncol Biol Phys 51(2):311-317, 2001b. 4) Fu XL, Huang H, Bentel GC, Clough R, Jirtle RL, Kong FM, Marks LB, Anscher MS: Predicting the risk of symptomatic radiation-induced lung injury using both the physical and biological parameters V30 and transforming growth factor ß. Int. J. Radiat. Oncol. Biol. Phys. 50(4):899-908, 2001. 5) Graham MV, Purdy JA, Emami B, Harris W, Bosch W, Lockett MA, Perez CA. clinical dose-volme histogram analysis for pneumonitis after 3D treatment for non-small cell lung cancer (NSCLC). Int J Radiat Oncol Biol Phys 1999 Sep 1;45(2):323-9. 6) Hernando ML, Marks LB, Bentel GC, Zhou S, Hollis D, Das SK, Fan M, Munle, MT, Shafman TD, Anscher MS, Lind PA. Radiation induced pulmonary toxicity: A dose volume histogrm analysis in 201 patients with lung cancer. Int. J. Radiat. Oncol. Biol. Phys. 51(3):650-659, 2001. 7) Kocak Z, Evans ES, Zhou SM, Miller KL, Folz RJ, Shafman TD, Marks LB. Challenges in defining radiation pneumonitis in patients with lung cancer. Int J Radiat Oncol Biol Phys 62:635-638, 2005. 8) Kocak Z, Gerben R. Borst, M.D.; Zeng Z, Zhou S, Hollis DR, Zhang J, Evans ES, Folz RJ, Wong TZ, Kahn D, Belderbos JS, Lebesque JV, Marks LB. Prospective Assessment of Dosimetric/Physiologic-Based Models for Predicting Radiation Pneumonitis. International Journal of Radiation Oncology Biology Physics, 2006. 9) Kwa SLS, Lebesque JV, Theuws JCM, Marks LB, Munley MT, Bentel G, Oetzel D, Spahn U, Graham MV, Drzymala RE, Purdy JA, Lichter AS, Martel MK, Ten Haken RK. Radiation pneumonitis as a function of dose: An analysis of pooled data of 540 patients. Int. J. Radiat. Oncol. Biol. Phys. 42:1-9; 1998. 10) Marks LB, Fan M, Clough R, Munley M, Bentel G, Coleman RE, Jaszczak R, Hollis D, Anscher M: Radiation induced pulmonary injury: symptomatic verses subclinical endpoints. Int. J. Rad. Biol. 76:469-475; 2000. 11) Marks LB, Lebesque JV. The challenge of predicting changes in pulmonary function tests after thoracic irradiation. Int J Radiat Oncol Biol Phys 55(5):1164-1165, 2003. 12) Marks LB, Munley MP, Bentel GC, Scarfone C, Zhou S-M, Hollis D, Jaszczak R, Sibley G, Coleman RE, Kong F-M, Jirtle R, Tapson V, Anscher M. Physical and biological predictors of changes in whole lung function following thoracic irradiation. Int. J. Radiat. Oncol. Biol. Phys. 39:563-570; 1997b. 13) Marks LB, Spencer DP, Sherouse GW, Bentel GC, Hoppenworth J, Munley MT, Chew M, Jaszczak RJ, Coleman RE, Prosnitz LR. Quantification of radiation-induced regional lung injury with perfusion imaging. Int. J. Radiat. Oncol. Biol. Phys. 38:399-409; 1997. 14) Theuws JCM, Kwa SLS, Wagenaar AC, et al.: prediction of overall pulmonary function loss in relation to the 3D dose distribution for patients with breast cancer and malignant lymphoma. Radiother Oncol 49:233-253; 1999a. 15) Woel RT, Munley MT, Hollis D, Fan M, Bentel G, Anscher MS, Shafman T, Coleman RE, Jaszczak RJ, Marks LB. The time course of radiation therapy-induced reductions in regional perfusion: A prospective study with > 5 years of followup.. Int. J. Radiat. Oncol. Biol. Phys. 52(1):58-67, 2002. 16) Zhang J, Jinli M, Sumin Z, et al. Radiation-Induced Reductions in Regional Lung Perfusion: 0.1-12 Year Data From a Prospective Clinical Study. Int J Radiat Oncol Biol Phys. 2010;76:425-432.