Clinical investigation
Significance of plasma transforming growth factor-β levels in radiotherapy for non–small-cell lung cancer

https://doi.org/10.1016/j.ijrobp.2003.09.078Get rights and content

Abstract

Purpose

In dose-escalation studies of radiotherapy (RT) for non–small-cell lung cancer (NSCLC), radiation pneumonitis (RP) is the most important dose-limiting complication. Transforming growth factor-β1 (TGF-β1) has been reported to be associated with the incidence of RP. It has been proposed that serial measurements of plasma TGF-β1 can be valuable to estimate the risk of RP and to decide whether additional dose-escalation can be safely applied. The aim of this study was to evaluate prospectively the time course of TGF-β1 levels in patients irradiated for NSCLC in relation to the development of RP and dose–volume parameters.

Methods and materials

Plasma samples were obtained in 68 patients irradiated for medically inoperable or locally advanced NSCLC (dose range, 60.8–94.5 Gy) before and 4, 6, and 18 weeks after the start of RT. Plasma TGF-β1 levels were determined using a bioassay on the basis of TGF-β1-induced plasminogen activator inhibitor-1 expression in mink lung cells. All patients underwent chest computed tomography scans before RT that were repeated at 18 weeks after RT. The computed tomography data were used to calculate the mean lung dose (MLD) and to score the radiation-induced radiologic changes. RP was defined on the basis of the presence of either radiographic changes or clinical symptoms. Symptomatic RP was scored according to the Common Toxicity Criteria (Grade 1 or worse) and the Southwestern Oncology Group criteria (Grade 2 or worse). Multivariate analyses were performed to investigate which factors (pre- or posttreatment TGF-β1 level, MLD) were associated with the incidence of RP. To improve our understanding of the time course of TGF-β1 levels, we performed a multivariate analysis to investigate which factors (pre-RT TGF-β1 level, MLD, RP) were independently associated with the posttreatment TGF-β1 levels.

Results

The pre-RT TGF-β1 levels were increased in patients with NSCLC (median 21 ng/mL, range, 5–103 ng/mL) compared with healthy individuals (range, 4–12 ng/mL). On average, the TGF-β1 levels normalized toward the end of treatment and remained stable until 18 weeks after RT. In 29 patients, however, TGF-β1 was increased at the end of RT with respect to the pre-RT value. The multivariate analyses revealed that the MLD was the only variable that correlated significantly with the risk of both radiographic RP (p = 0.05) and symptomatic RP, independent of the scoring system used (p = 0.05 and 0.03 for Southwestern Oncology Group and Common Toxicity Criteria systems, respectively). The TGF-β1 level at the end of RT was significantly associated with the MLD (p <0.001) and pre-RT TGF-β1 level (p = 0.001).

Conclusion

The MLD correlated significantly with the incidence of both radiographic and symptomatic RP. The results of our study did not confirm the reports that increased levels of TGF-β1 at the end of RT are an independent additional risk factor for developing symptomatic RP. However, the TGF-β1 level at the end of a RT was significantly associated with the MLD and the pre-RT level.

Introduction

Radiotherapy (RT) is the primary treatment modality for inoperable non–small-cell lung cancer (NSCLC). When delivered at conventional doses of 60–66 Gy, it has, however, limited ability to eradicate intrathoracic disease, with local control rates of only 15–17% at 1 year (1). RT dose escalation is a promising approach 2, 3, 4 but is limited by the presence of the surrounding normal lung tissue. Radiation injury to normal lung tissue can be divided into an early phase of radiation pneumonitis (RP), occurring 1 to 6 months after RT and a late phase of fibrosis, developing from 6 months onward. The symptoms of RP range from fever, dyspnea, and cough to death from respiratory failure. Pulmonary fibrosis may result in a mild to severe impairment of pulmonary function.

Several investigators have reported that the risk of developing RP depends on treatment-related factors such as the radiation dose (5), volume of lung irradiated 6, 7, fractionation schedule 8, 9 and use of concurrent chemotherapy 10, 11. A variety of dosimetric parameters, such as the mean lung dose (MLD) 12, 13 or the percentage of lung volume receiving more than a threshold dose of 20 Gy 14, 15, 25 Gy (16), or 30 Gy (17), can be derived from three-dimensional dose distributions. These parameters have shown good correlation with radiation-induced lung injury and have been used to segregate patients into risk groups 2, 17 and to derive normal (lung) tissue complication probability models 12, 13, 14, 16, 18, 19. These normal (lung) tissue complication probability models provide a risk estimate for a patient population, but suffer from considerable uncertainties when predicting the risk for an individual patient. This is not surprising, because the pathogenesis of RP is complex, and it has been reported that in addition to treatment-related factors, patient-related factors such as preexisting lung disease and poor pulmonary function (20), smoking 21, 22, and tumor location 15, 23, 24 may explain interpatient variation in the sensitivity for RP. Furthermore, biologic factors such as surfactant proteins (25), interleukin-6 (26), and interleukin-1α (27) have been identified as early circulating cytokine markers for RP. Anscher et al. (28) have extensively evaluated in clinical studies the role of plasma transforming growth factor-β1 (TGF-β1) in the development of RP.

TGF-β1 is a multifunctional cytokine and an important mediator of a myriad of biologic effects (29). Its activation plays a central role in the pathogenesis of inflammatory processes and is considered a master switch for the initiation, development, and persistence of radiation fibrosis (30). Anscher et al. (28) reported that after thoracic RT, the development of pulmonary injury was associated with persistently elevated plasma TGF-β1 levels at the end of RT. The authors proposed using plasma TGF-β1 levels to discriminate between patients with a low and high risk of developing RP and adjusting the radiation prescription dose on the basis of this risk. They showed that combining dosimetric parameters such as the volume of lung receiving more than 30 Gy (V30) with biologic information (TGF-β1) facilitated the stratification of patients into low-, intermediate-, and high-risk groups for RP development (31).

The purpose of this study was to evaluate prospectively the time course of plasma TGF-β1 levels in patients irradiated for NSCLC in relation to the development of RP and dose–volume parameters.

Section snippets

Patient and treatment characteristics

Sixty-eight patients with medically inoperable or locally advanced NSCLC were eligible for this study. Patients were referred to the Netherlands Cancer Institute for RT with radical or curative intent between 1997 and 2001. The pretreatment evaluation included medical history, physical examination, complete blood counts, chest radiography, and computed tomography (CT) of the chest and upper abdomen. Since 2000, 18fluoro-2-deoxy-glucose positron emission tomography was performed in all patients

Patient, tumor, and treatment characteristics

The patient, tumor, and treatment characteristics are summarized in Table 1. The median age was 74 years (range, 48–88 years). Of the 68 patients, 49 (72%) were men; 34 patients (50%) had Stage I or II and 34 (50%) had Stage III NSCLC. Forty-six patients were treated within the context of a Phase I-II dose-escalation trial (4). The average dose delivered to the tumor was 76 Gy (range, 60.8–94.5 Gy). Most patients (78%) received involved field RT. The average MLD was 15 Gy (range, 4–24 Gy).

Discussion

This is the first report showing that the MLD is significantly associated with the plasma TGF-β1 level at the end of RT. In addition, the results of this study showed that the MLD is the most important prognostic factor for the development of RP, whether scored according to symptoms or radiographic changes. These findings suggest that the MLD correlates with the incidence of radiation-induced injury, among others, through its effect on TGF-β1 levels.

TGF-β1 is a cytokine secreted as a

Conclusion

Our study results showed that changes in the plasma TGF-β1 levels during RT for NSCLC were dose and volume dependent. Increased levels of TGF-β1 at the end of RT were not predictive of the risk of RP. The overall dosimetric parameter, MLD, was the most predictive parameter of the risk of RP after RT for NSCLC, independent of the scoring system used.

Acknowledgements

We thank Annette van Assen for the measurements of TGF-β and Guus Hart for statistical advice.

References (45)

  • Y Seppenwoolde et al.

    Comparing different NTCP models that predict the incidence of radiation pneumonitis

    Int J Radiat Oncol Biol Phys

    (2003)
  • M.V Graham et al.

    Clinical dose-volume histogram analysis for pneumonitis after 3D treatment for non-small cell lung cancer (NSCLC)

    Int J Radiat Oncol Biol Phys

    (1999)
  • E.D Yorke et al.

    Dose-volume factors contributing to the incidence of radiation pneumonitis in non-small-cell lung cancer patients treated with three-dimensional conformal radiation therapy

    Int J Radiat Oncol Biol Phys

    (2002)
  • J.G Armstrong et al.

    Strategy for dose escalation using 3-dimensional conformal radiation therapy for lung cancer

    Ann Oncol

    (1995)
  • L.B Marks

    Dosimetric predictors of radiation-induced lung injury

    Int J Radiat Oncol Biol Phys

    (2002)
  • M.K Martel et al.

    Dose-volume histogram and 3-D treatment planning evaluation of patients with pneumonitis

    Int J Radiat Oncol Biol Phys

    (1994)
  • L.B Marks et al.

    Physical and biological predictors of changes in whole-lung function following thoracic irradiation

    Int J Radiat Oncol Biol Phys

    (1997)
  • P.A Lind et al.

    Receiver operating characteristic curves to assess predictors of radiation-induced symptomatic lung injury

    Int J Radiat Oncol Biol Phys

    (2002)
  • S Johansson et al.

    Effects of ongoing smoking on the development of radiation-induced pneumonitis in breast cancer and oesophagus cancer patients

    Radiother Oncol

    (1998)
  • V.V Moiseenko et al.

    In-field and out-of-field effects in partial volume lung irradiation in rodentsPossible correlation between early DNA damage and functional endpoints

    Int J Radiat Oncol Biol Phys

    (2000)
  • M.A Khan et al.

    Partial volume rat lung irradiation; assessment of early DNA damage in different lung regions and effect of radical scavengers

    Radiother Oncol

    (2003)
  • R Sasaki et al.

    Clinical significance of serum pulmonary surfactant proteins a and d for the early detection of radiation pneumonitis

    Int J Radiat Oncol Biol Phys

    (2001)
  • Cited by (0)

    Supported by Grant 99-2043 from the Dutch Cancer Society.

    View full text