Patient-centered use of stereotactic body radiation therapy in lung cancer

Lung cancer is still the leading cause of cancer-related death in Denmark. Only about 50% of patients are diagnosed at a localized or locally advanced stage, and even after curative-intent treatment, relapse remains common. Stereotactic body radiation therapy (SBRT) is the primary treatment for patients with localized non-small cell lung cancer not eligible for or not wanting surgery. While the treatment is usually well tolerated, dosing is tailored to the expected risk of side effects and to whether the target is a primary tumor or a metastatic lesion. The patients are rarely involved in the decision of the dose level. The infrequent patient involvement, the desire for more efficient radiotherapy causing fewer relapses and the need for earlier detection of relapse led us to plan and conduct the studies in this PhD thesis. The thesis is based on three studies all concerned with patients receiving curative SBRT for tumors in the lung but dealing with different points of impact during the treatment trajectory. All patients were included at the Department of Oncology at Lillebaelt Hospital in Vejle. 

Study I is an ongoing randomized controlled trial concerning shared decisionmaking (SDM) and patient involvement when determining the treatment dose with SBRT. Twenty-eight patients have been included and randomized to have their primary consultation with or without a specifically developed patient decision aid. Eligible patients had a tumor in the lung located ≤ 1 cm from the thoracic wall, where the expected risk of side effects from the thoracic wall was higher. Patients were informed about the possibility of reducing the doseand hereby reducing the risk of side effects but also risking a decrease in local tumor control. The decision of receiving radiotherapy with a high dose of 66Gray (Gy) or a lower dose of 45 Gy were made in all cases. The consultations were audiotaped for later scoring with the OPTION-12 tool, the most reliable tool to evaluate the extent of SDM in a consultation. A range of secondary scores and evaluation of side effects will be made during the 5-year follow-up. 

Study II was a retrospective radiotherapy dose planning study involving 39 patients planned with SBRT for a tumor or metastasis in the lung. The patients were treated using the department’s normal planning target volume-basedmethod, but later an extra plan using robust optimization was created for each patient. The two plans were then compared using robust evaluation using the maximum amplitudes of the gross tumor volume (GTV) respiratory motion seen for each patient on the 4D planning CT and the daily cone-beam CTs during each fraction. Comparisons included isodose volumes, organ-at-risk (OAR)doses, target doses in both the mid-position CT phase and accumulated for all10 respiratory phases, monitor units, and plan creation time. The study showed statistically significant reductions in almost all isodose volumes and for most OAR doses when using robust optimization. The mean dose to the GTV was also increased, and monitor units was slightly higher. On average, plan creation required five times longer, suggesting that robust optimization may be impractical for routine use but could be valuable in selected cases.

Study III was a prospective cohort study that collected blood samples from 68patients treated with SBRT or long-course radiotherapy for a stage I-III lung cancer. The samples were collected before, during and for one year following the radiotherapy treatment, and were analyzed for natural killer (NK) cell activity, soluble programmed death-ligand 1 and methylated circulating tumor DNA (ctDNA). ctDNA was analyzed with droplet digital polymerase chain reaction(ddPCR) near the four genes TFAP2B, MCIDAS, HOXA9 and SP9. The blood marker levels were compared among patients with and without a disease relapse following the first 12 months after the treatment. A reduced baseline level of NK activity or having a positive ctDNA status at baseline or at 6 months were all significantly prognostic of developing a relapse before 12 months following treatment. Receiver operating characteristic analyses were used in a stepwise fashion to establish the ability of the combinations of the six biomarkers to predict a relapse during the 12 months follow-up. The combination of all six biomarkers had an area under the curve of 0.86 [0.74-0.98], revealing a very good discriminative capability of the combined biomarkers. 

These studies provide new insights that contribute to further improvements in SBRT treatment, as well as a better understanding of patient preferences and approaches to post-treatment monitoring in lung cancer.