Module 8: Clinical Outcomes by Disease Site - The Use of Proton Therapy in the Treatment of Cancers of the Lung
Introduction
Lung cancer represents the leading cause of cancer death in the United States with an estimated 158,000 deaths in 2016. Roughly 90% of new lung cancers are non-small cell lung cancers’ (NSCLC). While several general principles discussed here do apply to small cell lung cancer as well, specifics in treatment approaches differ between these two subtypes and most major studies discussed here are limited to patients with NSCLC. In general, the use of the term lung cancer in this module will refer to NSCLC.
Broadly, lung cancers characterized as AJCC Stages IA-IIIB are considered curable with local treatment. Stage I disease is often best treated by lobectomy or stereotactic body radiation therapy (SBRT) in inoperable patients or those wishing to forego surgery. Stage II lung cancers are also often treated surgically in operable patients. Conventionally-fractionated radiation, ideally with concurrent chemotherapy, represents the mainstay of treatment for most Stage III lung cancers or in inoperable patients with Stage II disease. Currently, proton therapy may offer the largest potential benefit in definitive treatment of this latter cohort.
Evidence supporting proton therapy relative to photon therapy in locally-advanced lung cancer is derived from dosimetric studies, retrospective studies, and small early prospective experiences. There are no large published randomized controlled trials investigating patient outcomes between proton and photon radiation in lung cancer, though such studies are currently in progress. These will be discussed in more detail in the discussion that follows.
Technical considerations in proton radiation for locally-advanced lung cancer
From a technical perspective, x-ray (photon)-based radiation for lung cancer involves one of two techniques: 3D-conformal radiotherapy, or more often, intensity-modulated radiation therapy (IMRT). A primary advantage to these modalities, particularly IMRT, is high conformality. Photon-based radiation additionally provides plans are that are more robust to internal organ motion and tissue heterogeneities. A major challenge in proton treatment of lung cancers in general is the sensitivity of protons to changes in tissue density (specifically electron density). Put simply, the sharp contrast in density between a moving solid lung tumor and the surrounding aerated lung complicates our ability to control the radiation dose when treating lung cancer with proton radiation. While organ motion is still a challenge with photon radiation, it does not compromise the plans to the same degree as it does proton radiation. However, the drawback of photon radiation is the larger degree of medium and low dose spread to healthy lung, heart, and esophagus, which can cause toxicity that limits our ability to treat lung cancer more effectively.
Proton radiation is administered in one of two ways: passive scattering or pencil beam scanning (see Modules 1-2 for detailed discussion on these techniques). Because of the characteristic dose distribution of protons, either technique often better limits spread of low dose radiation to surrounding organs. Passive scattering is less sensitive to organ motion compared to pencil beam scanning, but both are still generally inferior to IMRT in this regard. However, passive scattering is less conformal than other radiation modalities. While pencil beam scanning is more conformal than passive scattering, the dose distribution is less robust in highly mobile targets and its complexity renders it more costly from the standpoint of planning, optimization, and quality assurance. Preliminary data indicate that pencil beam scanning is safe in lung cancer, but formal guidelines on its use are still pending.
Another technical limitation of proton therapy in treating lung cancer includes lack of daily three dimensional image guided radiation therapy. While cone beam CT is starting to emerge at select proton centers, this technology is largely unavailable in most proton facilities. Therefore, confidence in daily patient positioning in certain situations may be compromised compared with proton radiation compared to photons.
Clinical rationale for proton radiation in locally-advanced lung cancer
Given the ability of proton radiation to limit excessive radiation dose to organs at risk, it offers two strategies to improve outcomes in lung cancer. The first is to deliver a similar dose of radiation while lowering radiation exposure to surrounding organs. For example on RTOG 0617, a trial that compared chemotherapy and radiation to a dose of 60 Gy vs. 74 Gy in locally advanced lung cancer, grade 3 toxicities were seen in over 75% of patients. The primary outcome of this study was survival. Contrary to the authors’ hypothesis, the high dose arm experienced worse overall survival compared to the low dose (median overall survival 20.3 months [17.7,25.0] vs. 28.7 months [24.1, 36.9]). A principle explanation for this finding was a higher dose of radiation received by the heart in the high dose arm contributing to treatment-related death. In particular, the heart V5 and V30 were both significant independent predictors for overall survival on multivariable Cox proportional hazards. Proton radiation offers a theoretical benefit of improving this toxicity profile by lowering radiation exposure to thoracic organs such as heart, lung, and esophagus that may contribute to treatment toxicity.
However, our outcomes in locally-advanced lung cancer are poor and lowering dose to healthy organs alone may not be sufficient. To this end, protons may allow for safer dose escalation in lung cancer. Historically, chemoradiation at current dose levels controls the primary tumor only 50% of the time and the best estimates of 2 year survival for locally-advanced lung cancer with contemporary treatment, such as on RTOG 0617, is just over 50%. Therefore, escalating radiation dose using the right technology and with prudent organ-sparing may offer a chance for improvement. Proton therapy may provide a means to do this safely while limiting dose to surrounding organs at risk. For example, a prominent dosimetric study from MD Anderson (Zhang et al) demonstrated that dose escalation up to 84.4 Gy with proton radiation is feasible while maintaining doses to lung, spinal cord, heart, and esophagus to safe levels. They cite this accomplishment as particularly noteworthy given the conventionally held limit of 74 Gy as maximum esophageal tolerance. Furthermore, a Phase II trial at MD Anderson treating patients to 74 Gy with proton therapy demonstrated the best survival outcomes ever reported in the locally-advanced population treated with definitive chemoradiation with a median survival of 29.4 months. Moreover, only 11% of patients experienced grade 3 esophagus toxicity and 2% experienced grade 3 lung toxicity. The local failure on that trial was only 20.5%, which favorably compares with 50% local failure described historically in patients with locally-advanced NSCLC treated with definitive chemoradiation. It is with this logic that the ongoing RTOG 1308, a multicenter randomized controlled trial of proton versus photon radiation to 70 Gy, was proposed and is currently under accrual. The trial is designed to detect an improvement in overall survival from 21 months to 28 months with proton therapy versus photon radiation, and is also examining progression-free survival and toxicity as secondary endpoints. In a similar vein, MD Anderson Cancer Center and Harvard are collaborating on a randomized study of protons versus photons treating to 74 Gy. Early results of this study were presented in abstract form at the ASCO 2016 Annual Meeting indicating equivalent outcomes in proton versus proton radiation. However, as of the writing of this module, full published results are pending.
Other applications for proton re-irradiation for lung cancer: re-irradiation and early stage disease
While most efforts in proton radiation for lung cancer are directed at treatment of patients with locally-advanced disease, proton radiation is additionally being investigated in Stage I disease and also in the setting of recurrence. Proton therapy had initially been explored in hypofractionated therapy for Stage I lung cancer at Loma Linda around 20 years ago. In the final report of their single arm trial, doses of 70 Gy in 10 fractions were shown to be safe and offered local control rates of 96% in peripheral T1 tumors with no increase in toxicity for centrally treated lesions. While results were initially promising, this technique was eclipsed by the development of photon-based SBRT for early stage disease as discussed previously. This latter approach became more popular as it was more widely available than proton therapy and allowed the additional advantage of daily image guidance. In general, there is not a clear dosimetric advantage of proton therapy in Stage I lung cancer, though some work has demonstrated that in larger tumors, central tumors, tumors at the lung apex, and in treatment of multiple tumors, proton-based SBRT may have a role.
Additionally, proton therapy is being used in the re-irradiation setting in lung cancer. Due to high local recurrence rates, patients with previously irradiated lung cancer may recur within or near a previously treated field. Proton therapy can accomplish reirradiation more safely by limiting dose to previously-irradiated tissues. Previous reports from the University of Pennsylvania, who presented results of 24 patients treated on a prospective trial of lung reirradiation to a median dose of 66.6 Gy with concurrent chemotherapy, have demonstrated this can be done safely in well-selected patients. Deaths possibly related to treatment were noted in the subgroup of patients with particularly high volume disease in this trial, underscoring the need for appropriate patient selection for reirradiation. In general, proton re-irradiation is better tolerated than photon re-irradiation due to the improved dosimetry of proton plans.
Conclusions
In summary, proton therapy is showing early promise in the treatment of locally-advanced non small-cell lung cancer, and likely is useful in the retreatment setting and in certain early stage patients as well. Advantages to proton therapy hinge on its ability to limit dose to the healthy lung, esophagus, and heart and possibly to further escalate radiation doses beyond what has traditionally been feasible. Mature prospective data are pending, but will hopefully soon shed light on the true relative benefits of costs of this technology. Because the relative advantages and disadvantages of each radiation modality depend on the patient’s unique anatomy, the decision of which plan is best suited to treat should be made on an individual basis.