Can Patient Reporting Affect Radiation Pneumonitis?

January 1, 2008

Radiation therapy (RT) is an important treatment modality for multiple thoracic malignancies. Incidental irradiation of the lungs, which are particularly susceptible to injury, is unavoidable and often dose-limiting. The most radiosensitive subunit of the lung is the alveolar/capillary complex, and RT-induced lung injury is often described as diffuse alveolar damage. Reactive oxygen species generated by RT are directly toxic to parenchymal cells and initiate a cascade of molecular events that alter the cytokine milieu of the microenvironment, creating a self-sustaining cycle of inflammation and chronic oxidative stress. Replacement of normal lung parenchyma by fibrosis is the culminating event. Depending on the dose and volume of lung irradiated, acute radiation pneumonitis may develop, characterized by dry cough and dyspnea. Fibrosis of the lung, which can also cause dyspnea, is the late complication. Imaging studies and pulmonary function tests can be used to quantify the extent of lung injury. While strict dose-volume constraints to minimize the risk of injury are difficult to impose, substantial data support some general guidelines. New modalities such as intensity-modulated radiation therapy and stereotactic body radiation therapy provide new treatment options but also pose new challenges in safely delivering thoracic RT.

Ghafoori et al have provided a useful and timely update on radiation-induced lung injury. As they point out, the most consistent factors associated with an increased risk of developing radiation pneumonitis (RP) are dosimetric parameters, such as the V20 (volume of both lungs receiving > 20 Gy) or the mean lung dose. Yet, in a large Radiation Therapy Oncology Group (RTOG) analysis, Movsas et al examined the association of sociodemographic factors (in addition to treatment parameters) with RP among 1,450 patients treated on nine prospective RTOG nonoperative non–small-cell lung cancer (NSCLC) studies activated during the 1990s.[1] On multivariate analysis, they found that the following factors were significant predictors of grade ≥ 3 RP: lower lobe primaries, married/live-in relationship, presence of family/friends with cancer, and the interaction of high Karnofsky performance status (KPS) and female gender.

Questions Arising From These Results

The nature of such sociodemographic factors suggests that these variables may influence the degree of reporting and/or diagnosis of RP. Is it possible, for example, that RP is more readily reported or diagnosed in patients who have a stronger "voice" (married/live-in, family/friends, high KPS/female)? As RP is a subacute toxicity that typically manifests several months after completion of radiation therapy (RT), do the current dosimetric predictive models primarily reflect risk factors for those who are actually reporting they have toxicity? What does this mean for patients without a prominent voice who may not be consistently following up with their radiation oncologist? This RTOG analysis illustrates the need to include sociodemographic factors in risk models for RT toxicity for lung cancer. It also suggests increased vigilance is necessary to avoid underreporting or underdiagnosis of RP.

If such sociodemographic factors are indeed important, why have they not been previously appreciated? Perhaps the reason is that investigators in the past typically have not assessed these key factors. Recently, in a separate RTOG analysis, Konski et al reported a significant survival disadvantage for men living alone who were treated on RTOG head and neck trials.[2] In light of this, RTOG has expanded its outcomes model to now include key parameters including clinical, humanistic, and economic endpoints, as well as biologic markers and physical parameters, such as imaging and dosimetry.

Sensitivity of Toxicity Measurement

To date, no agent has been approved to prevent RP. This limitation, though, may be partly related to our inability to sensitively enough measure toxicity, rather than the inactivity of the compounds being studied. For example, the randomized RTOG 9801 trial of amifostine (Ethyol), a radioprotector, showed no significant differences in the rates of esophageal or pulmonary toxicity.[3] However, patients reported significant improvement in swallowing symptoms and pain, demonstrating a "disconnect" between physician- vs patient-reported outcomes (PROs). This suggests that our current clinical toxicity endpoints, primarily reflecting the physician's and not the patient's perspective, may not be sufficiently sensitive.

Interestingly, when pulmonary function testing was closely followed, Komaki et al found a significant difference in the degree of reduction in diffusing capacity of the lung for carbon monoxide (DLCO), favoring patients who received amifostine.[4] This finding was observed despite the fact that this trial,[4] which also employed hyperfractionated radiation, used amifostine only twice per week vs four times per week in RTOG 9801.[3] Of note, Trotti et al[5] elegantly demonstrated the overall lack of sensitivity of standard toxicity reporting in the RTOG head and neck database and developed a more sophisticated methodology (TAME, an acronym for three areas of risk evaluation-acute toxicity [T], adverse late events [A], treatment-related mortality risk [M]-in an end-result [E] summary). We are planning to study the TAME system within the RTOG lung database as well.

Evolving Treatments

While we await more sensitive toxicity measurements that better reflect the patient perspective, it's important to recognize that radiation treatments also keep evolving and improving. Based on the precision of stereotactic body RT (SBRT) for lung cancer, most would agree that the V20 has essentially become moot, and investigators are searching for other parameters to assess lung toxicity in this setting that may be more relevant. Such analyses will clearly require precise calculations of the radiation dose to the lung, using not only heterogeneity corrections, but also sophisticated algorithms, such as the Monte Carlo method. Proper accounting of respiratory-induced tumor and lung motion is also necessary to calculate dosimetric indices accurately.[6]

Whereas SBRT appears quite safe for peripheral lung lesions (with minimal lung toxicity), Timmerman et al[7] recently reported a high rate of grade 3–5 pulmonary toxicity with SBRT (using 20–22 Gy × 3 fractions) for central NSCLC lesions (with a median follow up of ~1 year). This alarming finding highlights the need to always remain vigilant when applying a novel RT technique, particularly involving high-dose fraction sizes. It also raises a key question: Is it possible to safely treat central lung lesions with SBRT?

At Henry Ford Hospital, we now have a minimum follow-up of 1 year on nine consecutive patients with central NSCLC lesions treated with SBRT (with 12 Gy mostly × 4 fractions) on an institutional review board (IRB) study. The actuarial 1-year local control rate is 87% with no cases of grade 3 or higher lung toxicity.[8] These preliminary results suggest that using up to 12 Gy × 4, one can safely treat central NSCLC lesions with SBRT. RTOG is currently developing a dose-escalation trial to further study the optimal dose fractionation schema for central lung lesions.

Ultimately, in order to maximize the therapeutic ratio in lung cancer, we must not only improve the efficacy of our treatments, but also miminize the potential toxicity to the lung. This will require a concerted effort to more accurately measure lung toxicity, while always keeping in mind the perspective of patient reporting, or "PR on RP."

-Benjamin Movsas, MD
-Indrin Chetty, PhD
-Munther Ajlouni, MD

Financial Disclosure: The authors have no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.


1. Movsas B, Swann S, Curran W. Sociodemographic factors are significant predictors of toxicity in RTOG non-Operative NSCLC trials. Int J Radiat Oncol Biol Phys 66(3):S62, 2006.

2. Konski A, Pajak T, Movsas B, et al: Disadvantage of men living alone participating in Radiation Therapy Oncology Group head and neck trials. J Clin Oncol 24:4177-4183, 2006.

3. Movsas B, Scott C, Langer C, et al: Randomized trial of amifostine in locally advanced non-small-cell lung cancer patients receiving chemotherapy and hyperfractionated radiation: Radiation therapy oncology group trial 98-01. J Clin Oncol 23:2145-2154, 2005.

4. Komaki R, Lee JS, Kaplan B, et al: Randomized phase III study of chemoradiation with or without amifostine for patients with favorable performance status inoperable stage II-III non-small cell lung cancer: Preliminary results. Semin Radiat Oncol 12(1 suppl 1):46-49, 2002.

5. Trotti A, Pajak T, Gwede C, et al: TAME: Development of a new method for summarising adverse events of cancer treatment by the Radiation Therapy Oncology Group. Lancet Oncol 8:613-624, 2007.

6. Jin J, Ajlouni M, Chen Q, et al: Quantification of incidental dose to potential clinical target volume (CTV) under different stereotactic body radiation therapy (SBRT) techniques for non-small cell lung cancer: Tumor motion and using internal target volume (ITV) could improve dose distribution in CTV. Radiother Oncol Sept 28, 2007 [epub ahead of print].

7. Timmerman R, McGarry R, Yiannoutsos C, et al: Excessive toxicity when treating central tumors in a phase II study of stereotactic body radiation therapy for medically inoperable early-stage lung cancer. J Clin Oncol 24:4833-4839, 2006.

8. Movsas B, Patel A, Ajlouni M, et al: Is it safe to treat central lung lesions with SBRT? Presented at the American Society of Therapeutic Radiology (ASTRO) Presidential Session, October 2007.