Results of multifield conformal radiation therapy of nonsmall-cell lung carcinoma using multileaf collimation beams

Is IMRT or VMAT superior or inferior to 3D conformal therapy in the treatment of lung cancer? A brief literature review

Aim:

To identify treatment outcome, dose uniformity, treatment time, toxicity among 3D conformal therapy (3D-CRT), intensity-modulated radiation therapy (IMRT), volumetric-modulated arc therapy (VMAT) for non-small-cell lung cancer (NSCLC) based on literature review.

Methods:

A literature search was conducted using PubMed/MEDLINE, BMC—part of Springer Nature, Google Scholar and iMEDPub Ltd with the following keywords for filtering: 3D-CRT, IMRT, VMAT, lung cancer, local control and radiobiology. A total of 14 publications were finally selected for the comparison of 3D-CRT, IMRT and VMAT to determine which technique is superior or inferior among these three.

Results:

Compared to 3D-CRT, IMRT delivers more precise treatment, has better conformal dose coverage to planning target volume (PTV) that covers gross tumour with microscopic extension, respiratory tumour motion and setup margin. 3D-CRT has large number of limitations: low overall survival (OS), large toxicity, secondary malignancies.

Conclusions:

It is difficult to choose the best technique for treating NSCLC due to patient conditions and technique availability. A high-precision treatment may improve tumour control probability (TCP) and patient’s quality of life. VMAT, whether superior or not, needs more clinical trials to treat NSCLC and requires longer dose optimisation time with the greatest benefit of rapid treatment delivery, improved patient comfort, reduced intrafraction motion and increased patient throughput compared to IMRT and 3D-CRT.

Comparison of intensity modulated radiotherapy and dynamic three-dimensional conformal radiotherapy with regard to dose distribution and sparing of organs at risk

Dose escalation to the target while sparing the organs at risk near the lesion has been difficult over the last decade. However, recent radiotherapy techniques can deliver more sophisticated doses to the target. This study evaluated whether intensity modulated radiotherapy can deliver more homogeneous and conformal doses to the target than dynamic three-dimensional conformal radiotherapy while sparing organs at risk near the lesion in 13 patients with central nervous system tumors and other tumors around the central nervous system. Dynamic three-dimensional conformal radiotherapy and intensity modulated radiotherapy plans were calculated and dose distributions were compared for all patients with regard to the planning target volume and organs at risk. The plan of intensity modulated radiotherapy was significantly superior to that of dynamic three-dimensional conformal radiotherapy in target dose conformity (p = 0.0006) and organs at risk sparing (p = 0.0257). Intensity modulated radiotherapy could deliver more homogeneous and conformal doses to the target than dynamic three-dimensional conformal radiotherapy with sparing organs at risk near the lesion and may improve local control of radioresistant tumors via dose escalation.

A phase I study of concurrent chemotherapy (paclitaxel and carboplatin) and thoracic radiotherapy with swallowed manganese superoxide dismutase plasmid liposome protection in patients with locally advanced stage III non-small-cell lung cancer

Manganese superoxide dismutase (MnSOD) is a genetically engineered therapeutic DNA/liposome containing the human MnSOD transgene. Preclinical studies in mouse models have demonstrated that the expression of the human MnSOD transgene confers protection of normal tissues from ionizing irradiation damage. This is a phase I study of MnSOD plasmid liposome (PL) in combination with standard chemoradiation in surgically unresectable stage III non-small-cell lung cancer. Chemotherapy (carboplatin and paclitaxel) was given weekly (for 7 weeks), concurrently with radiation. MnSOD PL was swallowed twice a week (total 14 doses), at three dose levels: 0.3, 3, and 30 mg. Dose escalation followed a standard phase I design. Esophagoscopy was done at baseline, day 4, and 6 weeks after radiation with biopsies of the squamous lining cells. DNA was extracted and analyzed by PCR for the detection of the MnSOD transgene DNA. Ten patients with AJCC stage IIIA (three) and IIIB (seven) completed the course of therapy. Five had squamous histology, two adenocarcinoma, one large cell, and two not specified. Patients were treated in three cohorts at three dose levels of MnSOD PL: 0.3 (three patients), 3 (three patients), and 30 mg (four patients). The median dose of radiation was 77.7 Gy (range 63-79.10 Gy). Overall response rate for the standard chemoradiation regimen was 70% (n = 10). There were no dose-limiting toxicities reported in all three dosing tiers. It is concluded that the oral administration of MnSOD PL is feasible and safe. The phase II recommended dose is 30 mg.

Predictive factors for radiation-induced pulmonary toxicity after three-dimensional conformal chemoradiation in locally advanced non-small-cell lung cancer

BACKGROUND AND PURPOSE

Radiation pneumonitis (RP) is a restricting complication of non-small-cell lung cancer irradiation. Three-dimensional conformal radiotherapy (3D-CRT) represents an advance because exposure of normal tissues is minimised. This study tries to identify prognostic factors associated with severe RP.

MATERIALS AND METHODS

Eighty patients with stage IIIA (20%) and IIIB (80%) NSCLC treated with cisplatin- based induction chemotherapy followed by concurrent chemotherapy and hyperfractionated 3D-CRT (median dose: 72.4 Gy, range: 54.1-85.9) were retrospectively evaluated. Acute and late RP were scored using RTOG glossary. Potential predictive factors evaluated included clinical, therapeutic and dosimetric factors. The lungs were defined as a whole organ. Univariate and multivariate analyses were performed.

RESULTS

Early and late RP grade>or=3 were observed in two patients (2%) and 10 patients (12%), respectively. Five patients (6%) died of pulmonary toxicity, 3 of whom had pre-existing chronic obstructive pulmonary disease (COPD). Median time to occurrence of late RP was 4.5 months (range: 3-8). Multivariate analysis showed that COPD (OR=10.1, p=0.01) and NTCPkwa>30% (OR=10.5, p=0.007) were independently associated with late grade>or=3 RP. Incidence of RP>or=3 grade for patients with COPD and/or NTCPkwa>30% was 25% vs. 4% for patients without COPD and NTCPkwa<30% (p=0.01). Risk of severe RP was higher for patients with COPD and/or NTCPkwa>30% (OR=7.3; CI 95%=1.4-37.3, p=0.016).

CONCLUSIONS

COPD and NTCP are predictive of severe RP. Careful medical evaluation and meticulous treatment planning are of paramount importance to decrease the incidence of severe RP.

Multivariate analysis of survival, local control, and time to distant metastases in patients with unresectable non-small-cell lung carcinoma treated with 3-dimensional conformal radiation therapy with or without concurrent chemotherapy

BACKGROUND

Three-dimensional (3D) conformal radiation therapy (CRT) and chemotherapy have recently improved lung cancer management.

PATIENTS AND METHODS

We reviewed outcomes in 68 patients with unresectable stage I-III non-small-cell lung cancer. Treatment consisted of 3D CRT alone or with concurrent chemotherapy (CCR).

RESULTS

Concurrent chemotherapy improved survival, to a median of 17 months +/- 4.9 months, compared with 8 months+/- 4.1 months for the radiation therapy (RT) alone group (P=0.0347). The 2- and 5-year survival rates were 40.3%+/-7.7% and 14.1%+/-6.4%, respectively, with CCR, compared with 19.6%+/- 9.6% and 0, respectively, for RT alone. In a subgroup analysis for age > 65, patients who received CCR (n=20) had significantly improved survival and local control (P=0.005 and P=0.0286, respectively). Acute esophageal toxicity Radiation Therapy Oncology Group grade >or= 3 was significantly higher in the CCR group and correlated with the RT dose (19% in CCR vs. 0 in RT, P=0.0234; P=0.050). The overall incidences of esophageal and pulmonary toxicity grade >or= 3 were 20.6% and 5.9%, respectively.

CONCLUSION

Our study confirms that CCR is associated with improved survival over RT alone, with a tolerable increase in acute toxicity.

Inhalation delivery of manganese superoxide dismutase-plasmid/liposomes protects the murine lung from irradiation damage

Intratracheal injection of manganese superoxide dismutase-plasmid/liposome (MnSOD-PL) complexes has been demonstrated to delay the onset and reduce the extent of ionizing irradiation-induced murine pulmonary organizing alveolitis/fibrosis. To facilitate translation of this modality to clinical fractionated radiotherapy, inhalation delivery of MnSOD-PL was developed using an ultrasonic nebulizer. Transgene product was quantitated by immunohistochemical quantitation and pulmonary tissue levels of MnSOD biochemical activity. C57BL/6NHsd female mice demonstrated a plasmid dose-dependent increased expression of MnSOD transgene product over the range of 250 microg-2.5 mg of MnSOD-PL administered over a constant 5 min interval. Delivery of a constant concentration of 500 microg of MnSOD-PL with varying times of administration ranging from 0.5 to 10 min demonstrated optimal MnSOD expression at 5 min. Mice pretreated by inhalation delivery of MnSOD-PL demonstrated significantly improved survival after 20 Gy single fraction irradiation to both lungs compared to LacZ-PL inhalation-treated or irradiated control mice. Mice receiving 10 fractions of 3.5 cGy demonstrated increased pulmonary MnSOD transgene product activity by a protocol of every Monday-Wednesday or daily inhalation of MnSOD-PL. Thus, inhalation radioprotective gene therapy using MnSOD-PL provides a practical and effective method for delivery of lung-specific radioprotection during fractionated radiotherapy protocols in a mouse model.

Radiotherapy alone in technically operable, medically inoperable, early-stage (I/II) non-small-cell lung cancer

PURPOSE

To investigate the effectiveness of high-dose, curative radiotherapy (RT) given alone in technically operable, but medically inoperable, patients with early-stage (I-II) non-small-cell lung cancer (NSCLC).

METHODS AND MATERIALS

Computerized and manual searches were done to identify published reports dealing with curative RT for NSCLC. Relevant studies were identified and the information provided therein was extracted regarding patient and treatment characteristics, treatment outcome, and various pretreatment and treatment-related factors influencing outcome, as well as toxicity and quality-of-life issues.

RESULTS

Although a large variation of pretreatment and treatment characteristics was noted in the available studies, a median survival time of >30 months and a 5-year survival rate of up to 30% had been achieved. Accumulated experience seems to suggest that doses of at least 65 Gy with standard fractionation, or its equivalent when altered fractionation is used, are necessary for control of NSCLC. Smaller tumors seem to have a favorable prognosis, and the issue of elective nodal RT continues to be controversial. Analyses of patterns of failure have clearly identified local failure as the predominant pattern. Although a number of potential pretreatment patient- and tumor-related prognostic factors have been examined, none has been shown to clearly influence survival. Toxicity was usually low, but very high doses (e.g., 80 Gy) given with a conventional approach may carry a risk of an excessive rate of side effects.

CONCLUSION

High-dose, curative RT is an effective treatment modality in technically operable, but medically inoperable, patients with early-stage NSCLC.

Update on Results of Multifield Conformal Radiation Therapy of Non—Small-Cell Lung Cancer Using Multileaf Collimated Beams

We evaluated the treatment outcome for 5-field 3-dimensional conformal radiation therapy (3D-CRT) in 46 consecutive patients with unresectable, nonmetastatic non-small-cell lung cancer treated from 1993-2001. Four percent of the patients had stage I tumors, 6% had stage II, 44% had stage IIIA, and 46% had stage IIIB tumors. The median radiation therapy (RT) dose to the gross tumor volume with a median of 467.5 cc (range, 75.0-3073.0 cc) was 6120 cGy (range, 3000-6840 cGy). Thirty-one of 46 patients (67.4%) received combined chemoradiotherapy. Mean follow-up was 13.2 months (range, 3-159 months). Survival for stage III patients was 48.7% +/-9.1% at 1 year and 25.0% +/-8.4% at 2 years, with a median survival of 12.0 months+/-4.4 months. The local control rate for stage III patients was 66.8%+/- 9.4% at 1 year and 28.5%+/- 10.4% at 2 years. Patients who received chemotherapy had better survival (P = 0.0533) and local control (P = 0.0984) compared with patients receiving RT alone. Esophageal toxicity >or= grade 3 was significantly greater in combined chemoradiotherapy patients (29% early, 13% late) compared to the patients receiving RT alone (0% early and late). Pulmonary toxicity (early and late) was limited to grades 1/2 in 24% of patients and early grade 3 in 2% of patients. Chemotherapy appears to improve survival and local control when added to 3D-CRT in this series. The addition of concurrent chemotherapy to RT significantly increased esophageal toxicity (within acceptable levels) and did not effect pulmonary toxicity in this series.

Induction chemotherapy plus three-dimensional conformal radiation therapy in the definitive treatment of locally advanced non-small-cell lung cancer

PURPOSE

To evaluate our institution's experience using chemotherapy in conjunction with three-dimensional conformal radiation therapy (3D-CRT).

METHODS AND MATERIALS

From 1991 to 1998, 152 patients with Stage III non-small-cell lung cancer (NSCLC) were treated with 3D-CRT at Memorial Sloan-Kettering Cancer Center. A total of 137 patients (90%) were surgically staged with either thoracotomy or mediastinoscopy. The remainder were staged radiographically. Seventy patients were treated with radiation therapy alone, and 82 patients received induction chemotherapy before radiation. The majority of chemotherapy-treated patients received a platinum-containing regimen. Radiation was delivered with a 3D conformal technique using CT-based treatment planning. The median dose in the radiation alone group was 70.2 Gy, while in the combined modality group, it was 64.8 Gy.

RESULTS

The median follow-up time was 30.5 months among survivors. Stage IIIB disease was present in 36 patients (51%) in the radiation-alone group and 57 patients (70%) in the combined-modality group. Thirty-nine patients had poor prognostic factors (KPS < 70 or weight loss > 5%), and they were equally distributed between the two groups. The median survival times for the radiation-alone and the combined-modality groups were 11.7 months and 18.1 months, respectively (p = 0.001). The 2-year rates of local control in the radiation-alone and combined-modality groups were 35.4% and 43.1%, respectively (p = 0.1). Grade 3 or worse nonhematologic toxicity occurred in 20% of the patients receiving radiation alone and in 16% of those receiving chemotherapy and radiation. Overall, there were only 4 cases of Grade 3 or worse esophagitis.

CONCLUSION

Despite more Stage IIIB patients in the combined-modality group, the addition of chemotherapy to 3D-CRT produced a survival advantage over 3D-CRT alone in Stage III NSCLC without a concomitant increase in toxicity. Chemotherapy thus appears to be beneficial, even in patients who are receiving higher doses of radiation therapy than are typically given with conventional techniques. Because locoregional failure remains a major challenge in patients with advanced disease, 3D-CRT in conjunction with chemotherapy may allow safe treatment to the dose levels required to further enhance local control.

Manganese superoxide dismutase-plasmid/liposome (MnSOD-PL) administration protects mice from esophagitis associated with fractionated radiation

Intraesophageal administration of manganese superoxide dismutase-plasmid/liposome (MnSOD-PL) prior to single fraction radiation has been shown to protect mice from lethal esophagitis. In our study, C3H/HeNsd mice received fractionated radiation in two protocols: (i) 18 Gy daily for four days with MnSOD-PL administration 24 hr prior to the first and third fraction, or (ii) 12 Gy daily for six days with MnSOD-PL 24 hr prior to the first, third, and fifth fraction. Control radiated mice received either no liposomes only or LacZ (bacterial beta-galactosidase gene)-plasmid/liposome (LacZ-PL) by the same schedules. We measured thiol depletion and lipid peroxidation (LP) in whole esophagus and tested the effectiveness of a new plasmid, hemagglutinin (HA) epitope-tagged MnSOD (HA-MnSOD). In fractionation protocols, mice receiving MnSOD-PL, but not LacZ-PL (200 microl of plasmid/liposomes containing 200 microg of plasmid DNA), showed a significant reduction in morbidity, decreased weight loss, and improved survival. Four and seven days after 37 Gy single fraction radiation, the esophagus demonstrated a significant increase in peroxidized lipids and reduction in overall antioxidant levels, reduced thiols, and decreased glutathione (GSH). These reductions were modulated by MnSOD-PL administration. The HA-MnSOD plasmid product was detected in the basal layers of the esophageal epithelium 24 hr after administration and provided significant radiation protection compared to glutathione peroxidase-plasmid/liposome (GPX-PL), or liposomes containing MnSOD protein, vitamin E, co-enzyme Q10, or 21-aminosteroid. Thus, MnSOD-PL administration significantly improved tolerance to fractionated radiation and modulated radiation effects on levels of GSH and lipid peroxidation (LP). These studies provide further support for translation of MnSOD-PL treatment into human esophageal radiation protection.

Modulation of redox signal transduction pathways in the treatment of cancer

Reactive oxygen species (ROS)-mediated damage to DNA is associated with induction of stress-activated protein kinases leading to secondary and tertiary effects on the nuclear matrix, cytoplasmic transport mechanisms, and altered mitochondrial and cell membranes. The cellular defenses against ROS damage are associated with up-regulation of gene products that can significantly alter cell biology, including antiapoptotic Bax family proteins and inflammatory proteins. Altered cell integrity can occur either directly or by indirect paracrine and juxtacrine interactions within tissues. Previous approaches toward therapeutic intervention against ROS damage have included administration of radical scavenger compounds, use of novel drugs that increase cellular production of constitutive antioxidants, or pharmacologic agents that modify the intracellular transport of antioxidants. Strategies to modify the cellular effects of ROS in hyperbaric oxygen injury to the lung, reperfusion injury to transplanted organs, and cancer have led to novel approaches of gene therapy in which the transgenes for antioxidant proteins can be expressed in specific tissues. Reducing tissue-damaging effects of ROS may have relevance to cancer patients by ameliorating normal tissue damage from ionizing irradiation therapy, photodynamic therapy, and cancer chemotherapy.

Modulation of radiation-induced cytokine elevation associated with esophagitis and esophageal stricture by manganese superoxide dismutase-plasmid/liposome (SOD2-PL) gene therapy

Radiation of the esophagus of C3H/HeNsd mice with 35 or 37 Gy of 6 MV X rays induces significantly increased RNA transcription for interleukin 1 (Il1), tumor necrosis factor alpha (Tnf), interferon gamma inducing factor (Ifngr), and interferon gamma (Ifng). These elevations are associated with DNA damage that is detectable by a comet assay of explanted esophageal cells, apoptosis of the esophageal basal lining layer cells in situ, and micro-ulceration leading to dehydration and death. The histopathology and time sequence of events are comparable to the esophagitis in humans that is associated with chemoradiotherapy of non-small cell lung carcinoma (NSCLC). Intraesophageal injection of clinical-grade manganese superoxide dismutase-plasmid/liposome (SOD2-PL) 24 h prior to irradiation produced an increase in SOD2 biochemical activity in explanted esophagus. An equivalent therapeutic plasmid weight of 10 microgram ALP plasmid in the same 500 microliter of liposomes, correlated to around 52-60% of alkaline phosphatase-positive cells in the squamous layer of the esophagus at 24 h. Administration of SOD2-PL prior to irradiation mediated a significant decrease in induction of cytokine mRNA by radiation and decreased apoptosis of squamous lining cells, micro-ulceration, and esophagitis. Groups of mice receiving 35 or 37 Gy esophageal irradiation by a technique protecting the lungs and treating only the central mediastinal area were followed to assess the long-term effects of radiation. SOD2-PL-treated irradiated mice demonstrated a significant decrease in esophageal wall thickness at day 100 compared to irradiated controls. Mice with orthotopic thoracic tumors composed of 32D-v-abl cells that received intraesophageal SOD2-PL treatment showed transgenic mRNA in the esophagus at 24 h, but no detectable human SOD2 transgene mRNA in explanted tumors by nested RT-PCR. These data provide support for translation of this strategy of SOD2-PL gene therapy to studies leading to a clinical trial in fractionated irradiation to decrease the acute and chronic side effects of radiation-induced damage to the esophagus.

Optimum beam angles for the conformal treatment of lung cancer: a CT simulation study

Treatment of lung cancer is often performed with cone-down oblique beams to spare spinal cord and normal structures. However, there is no optimum technique to determine oblique beam angles when a CT simulation is not available. Impact of oblique beam angle was investigated in this study. Fifteen patients with centrally located lung tumors were immobilized and scanned using a CT simulator. The target volumes, left and right lungs, and spinal cord were delineated on each slice. Patients were simulated starting with anterior-posterior treatment beams and subsequently an oblique opposed pair beam from 0 degrees up to 60 degrees at an interval of 5 degrees to optimize the projection of target-to-cord distance and minimize the lung volume in the treatment fields. Analysis was performed with a dose volume histogram (DVH) in each beam orientation. The distance between the target volume and spinal cord was linearly related to the angle of the beam. A larger angle facilitated further sparing of the spinal cord; however, progressively more lung volume was exposed. The 50% DVH data for lung volume was used as an indicator of lung volume. Although, the minimum lung volume was irradiated with an angle of 30 degrees, the additional lung treated increased by only 8 +/- 7% of the total lung volume for 30-60 degrees beam angles and cord distance increased by 18.5 mm. A 30 degrees oblique parallel-opposed beam for the cone-down treatment of lung provided minimum lung volume in the irradiated field; however, the spinal cord distance increased linearly with beam angle. A CT simulator is ideally suited for simulation of lung cancer to maximize the clearance from the spinal cord and minimize the additional lung volume irradiated. Int. J. Cancer (Radiat. Oncol. Invest.) 90, 359-365 (2000).

Plasmid/liposome transfer of the human manganese superoxide dismutase transgene prevents ionizing irradiation-induced apoptosis in human esophagus organ explant culture

Esophagitis is a major limiting factor in the treatment of lung cancer by radiation alone or in combination with chemotherapy. We have previously demonstrated that intraesophageal injection of manganese superoxide dismutase-plasmid/liposome (MnSOD-PL) complex into C3H/HeNsd mice blocks irradiation-induced esophagitis. To determine whether the human esophagus can be similarly transfected, normal human esophageal sections obtained from the margins of esophagectomy specimens from esophageal cancer patients were transfected in vitro with alkaline phosphatase (AlkP)-PL complex and stained for AlkP activity, and the percent of cells expressing AlkP was calculated. At 24 hr after transfection with 20 or 200 microgram of AlkP-PL complex, 55.0% and 85.8% of esophageal epithelial cells expressed detectable AlkP, respectively. Other sections transfected with MnSOD-PL complex showed transgene mRNA by nested reverse transcriptase-polymerase chain reaction (RT-PCR) assay and increased MnSOD biochemical activity for at least 96 hr after transfection. Irradiated MnSOD-PL complex-transfected sections demonstrated a significantly decreased percentage of apoptotic cells when compared to irradiated control sections. Following 1,000 cGy, MnSOD-PL-treated samples showed 7.5 +/- 2.8% and 33.3 +/- 7.3% apoptotic cells at 24 and 48 hr compared to 53.6 +/- 6.9% and 59.0 +/- 13.8% for nontransfected controls (P < 0.0001 and P < 0.1175). After 2,000 cGy, results at 24 and 48 hr were 25.0 +/- 7.6% and 66.9 +/- 4.9% for MnSOD-transfected sections compared to 65.6 +/- 4.3% and 90.0 +/- 4.1% for control sections (P < 0.0001 and P = 0.0353), respectively. Thus, human esophageal sections can be transfected with MnSOD-PL complex in vitro and thereby protected against ionizing irradiation-induced apoptosis. Int. J. Cancer (Radiat. Oncol. Invest.) 90, 128-137 (2000).

Intratracheal injection of manganese superoxide dismutase (MnSOD) plasmid/liposomes protects normal lung but not orthotopic tumors from irradiation

To determine whether intratracheal (IT) lung protective manganese superoxide-plasmid/liposomes (MnSOD-PL) complex provided 'bystander' protection of thoracic tumors, mice with orthotopic Lewis lung carcinoma-bacterial beta-galactosidase gene (3LL-LacZ) were studied. There was no significant difference in irradiation survival of 3LL-LacZ cells irradiated, then cocultured with MnSOD-PL-treated compared with control lung cells (D0 2.022 and 2.153, respectively), or when irradiation was delivered 24 h after coculture (D0 0.934 and 0.907, respectively). Tumor-bearing control mice showed 50% survival at 18 days and 10% survival at 21 days. Mice receiving liposomes with no insert or LacZ-PL complex plus 18 Gy had 50% survival at 22 days, and a 20% and 30% survival at day 50, respectively. Mice receiving MnSOD-PL complex followed by 18 Gy showed prolonged survival of 45% at 50 days after irradiation (P < 0.001). Nested RT-PCR assay for the human MnSOD transgene demonstrated expression at 24 h in normal lung, but not in orthotopic tumors. Decreased irradiation induction of TGF-beta1, TGF-beta2, TGF-beta3, MIF, TNF-alpha, and IL-1 at 24 h was detected in lungs, but not orthotopic tumors from MnSOD-PL-injected mice (P < 0.001). Thus, pulmonary radioprotective MnSOD-PL therapy does not provide detectable 'bystander' protection to thoracic tumors.