Lung Cancer and Proton Therapy

Reviewer: Eric Shinohara MD, MSCI
Abramson Cancer Center of the University of Pennsylvania
Last Modified: May 26, 2008

Presenter: Joe Chang, MD, PhD
Presenter's Affiliation: MD Anderson Cancer Center
Type of Session: Reporting


Dr. Chang’s presentation focused on the use of image guided proton therapy in the treatment of non-small cell lung cancer. He discussed ways to decrease inter and intra-fraction tumor motion as well as how changes in anatomy and tumor volume affect treatment. Dr. Chang began the seminar by discussing the incidence and outcomes associated with lung cancer. Lung cancer is responsible for more cancer deaths in the United States than any other cancer and an estimated 161,840 people will die of lung cancer in 2008. This is higher than the mortality associated with prostate, breast, rectal and pancreatic cancer combined. This number averages to one patient dying every three minutes of lung cancer. The 5 year overall survival for patients with lung cancer is estimated to be 15%. Currently, the local control is less than 50% using current radiation techniques and changes are needed to improve on this number.

Dr. Chang notes that there is an increasing amount of interest in using proton therapy to improve the therapeutic ratio in lung cancer, which may allow dose escalation. The spread out Bragg peak seen with protons may allow dose escalation while sparing normal tissues due to decreased exit dose compared with photons. Dr. Chang discussed two ways to deliver proton therapy, either with a passive scatter technique or a spot scanning beam, which allows the use of Intensity Modulated Proton Therapy (IMPT). However, organ motion presents a major problem in the treatment of lung cancer with proton therapy. Dr. Chang then presented data regarding the improvements in normal tissue sparing seen when IMRT and 3D conformal radiation therapy are compared. Prior studies have found a 10-20% improvement in V5 and V10 dose in both stage I and stage III lung cancers when they were planned with Intensity Modulated Radiation Therapy (IMRT) versus 3D conformal radiation therapy. Studies have also shown that when passive scatter protons plans were compared with IMRT plans there was greater normal tissue sparing with protons despite using a higher total tumor dose in the proton plans. IMRT plans have also been compared with IMPT plans in patients with stage III non-small cell lung cancer. Results demonstrated a 13-22% improvement in V5 and V10 dose with the IMPT plans compared with the IMRT plans. IMPT plans have also been compared with passive scatter proton plans and results demonstrated a 5-10% improvement in V5 and V10 dose with IMPT versus passive scatter.

Dr. Chang also points out that while there is greater sparing of normal tissues with protons, lung motion must be accounted for. He presented data which demonstrated that 50% of lung cancer tumors move between 0.5 to 1 cm with respiration and 10% move greater than 1 cm. He then presented data on two techniques to account for this breathing motion, by either using an internal target volume (ITV) or a gating technique. He presented data from Massachusetts General Hospital which demonstrating intra-fraction motion of lung tumors on 4D CT scanning. He also showed data demonstrating that the proton beam’s distal edge moves more distally as the tumor moves out of the beam. Based on this motion the tumor could be under-dosed if the margins used are not adequate. He then presented data on using 4D CT scanning to construct an ITV. By combining the path of motion of the tumor seen on the 4D CT into one volume, an ITV can be created. This study then demonstrated that when standard margins were used the tumor is under-dosed where as when an ITV is used there was greater sparing of normal tissues (due to better control of the distal margin of the proton beam) and better target coverage. Gating can also be used to spare normal lung tissue and Dr. Chang presented data demonstrating a 6% improvement in V5, V10, and V20 dose with gating.

Dr. Chang then discussed inter-fraction variation in proton therapy of lung cancer. He presented data which examined dose volume histograms generated during treatment of a lung cancer. This study found that as the tumor shrank the CTV coverage remianed intact. However, the cord, heart and contralateral lung dose increased during the course of treatment. Dr. Chang also presented data demonstrating that the density of a lung tumor changes over the course of radiation therapy based on CT data. They found that the dose to the contralateral lung was inversely proportional to the density of the CTV. This change in density was thought to be responsible for a decrease in CTV coverage from 99% to 92.3% with the use of protons over the course of treatment. This change in CTV dose was not seen in IMRT plans. In order to account for this, Dr. Chang presented data on adaptive proton therapy. In one patient that he had been treating, an early skin reaction was noted. On rescanning it was noted that the tumor had shrunk substantially and the patient was replanned to account for this shrinkage. He believes that rescanning lung cancer patients at 5 weeks into therapy would allow plans to be adapted in response to changes in tumor density and size.

Dr. Chang then discussed recent studies of the use of protons in the treatment of non-small cell lung cancer. He noted five series which have studied proton therapy for non-small cell lung cancer with the majority of patients having stage I cancer. The doses and fractionations used in these studies varied greatly with doses from 45-94 CGE used in seven to 32 fractions. Dr. Chang stated that the various stages, radiation doses and fractionation schemes included in these studies makes it difficult to interpret their results. Furthermore, he noted that 4D CT scans were not used and that the doses used in some of these studies may not have been adequate. Nonetheless, he believes these studies suggest that patients treated with doses of >100 CGE may have comparable outcomes compared with those treated with surgery and that protons may decrease treatment related toxicity. Dr. Chang then goes on to discuss trials which are currently accruing. He describes a trial examining the use of concurrent chemoradiation for treatment of stage III non-small cell lung cancer using either protons or photons. The retrospective trials which were the basis for this new prospective trial were presented by Dr. Ritsuko Komaki at this meeting and are discussed elsewhere. He also mentions a dose escalation study using protons which is accruing at MD Anderson. He then noted two adaptive studies being run jointly by MD Anderson and Harvard. The first study compares proton (87.5 CGE in 2.5 CGE fractions) to photon therapy (84 Gy in 2.15 Gy fractions) in the treatment of Stage I non-small cell lung cancer. The second study compares treatment with protons (74 CGE in 2 CGE fractions) to photons (74 Gy in 2 Gy fractions) with concurrent chemotherapy in stage III non-small cell lung cancer.

Author's Conclusions

Dr. Chang finished with the following conclusions:

  1. Proton therapy may allow dose escalation and decreased toxicity in the treatment of non-small cell lung cancer
  2. Proton therapy may enable greater use of hypofractionation and accelerated radiation therapy in lung cancer
  3. The use of 4D CT in treatment planning is important
  4. Further studies are need to investigate the use of protons in lung cancer and these studies are ongoing

Clinical/Scientific Implications

As more data has become available it has become increasingly clear that target delineation and accounting for tumor motion will be even more critical with protons than with IMRT. The range of protons can be affected by changes in the tumor size, tumor/normal tissue density, size of the patient, and organ motion. Though most of these factors do affect photons beams to a certain extent, because these factors also affect the range of protons, there can be a dramatic affect on tumor and normal tissue dose. Adaptive therapy will be a key component in ensuring that are consistently delivering dose to the tumor over the course of treatment, particularly if we are going to attempt to tighten margins. By rescanning we can adjust for changes in weight, tumor size and density. 4D CT will help greatly in defining volumes which take respiratory motion into account adequately. Breath holding may be the optimal option for treating lung cancers as it would prevent high density organs (such as ribs) from moving in and out of the field as well as limiting tumor motion. If breath holding can be used effectively, ITV’s may be reduced even further. However, many lung cancers patients may not be able to adequately hold their breaths during treatment. Clearly further studies are needed regarding the use of protons in lung cancer with a focus on high quality and reproducible patient positioning and motion control. If they are of high quality we can then think about progressing to tighter margins and dose escalation.