Elsevier

Lung Cancer

Volume 32, Issue 2, May 2001, Pages 145-153
Lung Cancer

An evaluation of two techniques for beam intensity modulation in patients irradiated for stage III non-small cell lung cancer

https://doi.org/10.1016/S0169-5002(00)00214-2Get rights and content

Abstract

In locally advanced lung cancer, the use of high dose radiotherapy (RT) and/or concurrent chemo-RT is associated with significant pulmonary and esophageal toxicity. Despite a 3D conformal RT technique and the omission of elective mediastinal fields, three (of ten) patients with inoperable stage 3 NSCLC who were treated with induction chemotherapy (carboplatin–paclitaxel) followed by RT to 70 Gy, developed symptomatic radiation pneumonitis. In this planning study, the actual treatment plans of all ten patients were compared to plans derived using two beam intensity-modulated (BIM) techniques, for which similar geometrical beam setup parameters were used. In the first technique (BF-BIM), cranial and caudal boost fields were applied in order to allow field length reduction. The second technique (C-BIM) utilised 3-D missing-tissue compensators for all radiation beams. Both BIM techniques resulted in a significant sparing of critical normal tissues and the C-BIM technique was superior in all cases. When compared to the actual RT technique used for treatment, a reduction of 8.1±4.7% (1 S.D.) was observed in the mean lung dose for the BF-BIM plan, vs. 20.3±5.8% (1 S.D.) for the C-BIM plan. Similar reductions were observed in the percentage of the total lung volume exceeding 20 Gy (V20) for these techniques. BIM techniques appear to be a promising tool for enabling radiation dose-escalation and/or intensive concurrent chemo-RT in inoperable lung cancer.

Introduction

The local tumor control in locally advanced non-small cell lung cancer (NSCLC) is poor after radiotherapy (RT) alone and a pathological complete response was only seen in 20% of patients who were treated to 65 Gy [1]. Improvements in local control in this group of patients can improve survival [2], [3]. In locally advanced NSCLC, treatment with induction chemotherapy (CHT) followed by RT is superior to RT alone [1], [4], [5]. A recent study also suggested that the concurrent administration of CHT and RT is superior to sequential treatment [6]. However, concurrent CHT and RT can significantly increase the incidence of early and late radiation-induced grade 3/4 toxicities [7]. Data from patients treated using 3-dimensional (3D) conformal radiotherapy (CRT) suggest that a dose of 84 Gy may be required in order to achieve a likelihood of local control in excess of 50% [8]. However, efforts to escalate RT doses have been limited by toxicity [9] and/or have resulted in considerable dose inhomogeneity in the tumor [4].

The volume of healthy lung tissue receiving a dose exceeding 20 Gy (V20) has been shown to predict for radiation pneumonitis [10]. Similarly, the length of esophagus in the RT field has been shown to correlate with esophagitis [11], [12]. Consequently, steps to decrease the volume of irradiated normal tissue and/or improvements in radiation dose-homogeneity may reduce the toxicity of combined treatments. One approach to reduce normal tissue toxicity would be by not irradiating the radiologically normal mediastinum, i.e. ‘involved-field’ RT. In an ongoing phase II clinical trial at our center, patients with inoperable stage III NSCLC are treated with induction chemotherapy, followed by ‘involved-field’ RT to a dose of 70 Gy. This trial tests the hypothesis that isolated recurrences in the radiologically normal mediastinum are likely to be uncommon after ‘involved-field’ RT, thereby permitting radiation dose-escalation and/or CHT-RT with acceptable normal tissue toxicity. To achieve the latter, we are exploring alternative planning techniques, which can further increase normal tissue sparing.

Dose distributions delivered by multiple-field irradiation techniques can often be significantly improved by modulating the 2D intensity profile of the individual X-ray beams. Such intensity modulated beam (BIM) profiles can be realised in many ways including static multileaf collimation [13], dynamic multileaf collimation [14], [15] and tissue compensators [16]. Accurate missing tissue compensators can be made using a set of closely spaced CT scans to construct the 3D anatomy and to obtain density information for making tissue inhomogeneity corrections [17]. Such 3D tissue compensators have been shown to improve dose homogeneity in patients with tumors of the head and neck [18] and breast [19]. This technique is used at our institute in the treatment of patients with head and neck cancer [20].

In this planning study, our current 3D conformal radiation technique (CRT) for lung cancer patients was compared with 2 BIM techniques, namely: (a) BIM using missing tissue compensators (C-BIM), and (b) a technique using static BIM with a multileaf collimator to sharpen the penumbra at the superior and inferior ends of the planning target volume (PTV), thereby allowing a reduction in the penumbra margin between the field edge and the PTV [21], (BF-BIM). The resulting plans were compared using ICRU 50 homogeneity criteria, dose volume histograms (DVH's), mean lung dose [22], and percentage of the total lung volume exceeding 20 Gy (V20) [10]. The latter two parameters are often used to predict the risk of radiation pneumonitis.

Section snippets

Material and methods

In our phase II trial, patients with inoperable stage III tumors are treated with either two or four cycles (the latter when at least a partial response has been observed) of carboplatin and paclitaxel, followed by RT to a dose of 70 Gy, in once-daily fractions of 2 Gy, to the gross tumor volume (GTV), i.e. tumor and mediastinal nodes with a short-axis diameter ≥1 cm (Table 1). Patients with supraclavicular lymphadenopathy, malignant pleural effusions or tumors exceeding 6 cm were ineligible

Results

The ten patients had tumors located in the upper lobe or hilar regions, and details of the primary tumor, nodal metastases and PTV's are summarized in Table 1. The mean PTV of the ten patients was 407.4±144.7 cc (1 S.D.). Four of the ten patients had a partial response after the initial two cycles of CHT, but one patient in this cohort was unable to receive the fourth cycle of CHT due to an episode of (unrelated) cholecystitis. Six out of ten patients were judged to have stable disease but one

Discussion

Despite the omission of RT to the uninvolved mediastinum, 3/10 patients treated with conventional CRT to 70 Gy experienced acute radiation pneumonitis requiring treatment with glucocorticoids. Although concurrent CHT-RT with paclitaxel has been reported to result in an increase in the incidence of radiation pneumonitis [27], this has not been reported when RT follows the completion of paclitaxel-containing schemes. In any case, the high V20 values (Table 3) seen with CRT planning for our

Conclusions

Despite thoracic RT using an ‘involved-field’ approach in patients with mainly stage 3B inoperable NSCLC, acceptable V20 values (≤35%) were not achieved in the majority of patients. By enabling a consistent reduction in the V20, BIM techniques appear to be a promising tool in studies evaluating RT dose-escalation and/or intensive concurrent CHT-RT.

Acknowledgements

The assistance of Dr E. van Dieren with the ‘Optimize’ program is gratefully acknowledged.

References (39)

Cited by (24)

  • Radiotherapy for lung cancer: Beyond the limits!

    2011, Revue des Maladies Respiratoires Actualites
  • A potential to reduce pulmonary toxicity: The use of perfusion SPECT with IMRT for functional lung avoidance in radiotherapy of non-small cell lung cancer>

    2007, Radiotherapy and Oncology
    Citation Excerpt :

    Dose escalation to 95–100 Gy while maintaining V20 at 15–25% and MLD <16 Gy with IMRT is theoretically possible in selected patients with NSCLC [32]. However, the role of IMRT in treating NSCLC remains uncertain owing to the concern that IMRT may deliver a low, yet damaging, dose to significant volume of normal lung, due to an increase in monitor units (MUs) to deliver and due to multi-leaf collimator leakage [20,22]. The biologic effect of the trade-off between a reduction of the high dose volume and an increase of the low dose volume is not clear.

  • Literature-based recommendations for treatment planning and execution in high-dose radiotherapy for lung cancer

    2004, Radiotherapy and Oncology
    Citation Excerpt :

    As the 5-year disease-free survival in stage III NSCLC remains under 20%, these risks may not be an issue for most patients but prospective data from ongoing studies are awaited. There is presently no consensus available on contouring the heart for radiotherapy planning, although a recent publication on lung cancer suggested that the cranial extent of contouring should include the infundibulum of the right ventricle and the apex of both atria while excluding the great vessels as much as possible, and that the caudal border be defined by the lowest part of the left ventricle's inferior wall that is distinguishable from the liver [67]. The available data strongly support the use of set-up correction protocols in patients with lung cancer by using an electronic portal imaging device (EPID) [9].

View all citing articles on Scopus
View full text