External validation of the lung-molGPA to predict survival in patients treated with stereotactic radiotherapy for brain metastases of non-small cell lung cancer.

BACKGROUND: In the era of personalized medicine, individualized prognostic models with tumor characteristics are needed to inform patients about survival. Before clinical use, external validation of such models by an independent group is needed. An updated version of the graded prognostic assessment (GPA) estimates survival in patients with brain metastases (BMs) of non-small cell lung cancer (NSCLC). This is the first external validation of the updated Lung-molGPA in patients treated with stereotactic radiotherapy (SRT) for one or more BMs. MATERIALS AND METHODS: Patients treated with SRT for BMs from NSCLC adenocarcinoma were retrospectively included. GPA score was calculated for each patient based on six prognostic factors including age, Karnofsky Performance Status, number of BMs, extracranial metastases, EGFR/ALK status, and PD-L1 expression. Kaplan-Meier analysis evaluated survival probability. Impact of individual prognostic factors on survival was assessed by univariate and multivariate analyses using the Cox proportional hazard model. Predictive performance was evaluated using discrimination (C-statistic) and calibration (Brier test). RESULTS: The cohort (n = 241) was divided into four prognostic groups. Overall median survival was 15 months. Predicted and observed median survival were similar between the original and validation cohorts, apart from the most favorable prognostic group. With adequate C-statistics and Brier scores, the Lung-molGPA provided accurate survival predictions. CONCLUSION: The Lung-molGPA accurately predicted survival in our European population, except for an overestimation of survival in the small most favorable prognostic group. This prognostic model was externally validated and is therefore useful for counseling of patients with BMs of NSCLC adenocarcinoma.

Development and evaluation of an automated EPTN-consensus based organ at risk atlas in the brain on MRI.

BACKGROUND AND PURPOSE: During radiotherapy treatment planning, avoidance of organs at risk (OARs) is important. An international consensus-based delineation guideline was recently published with 34 OARs in the brain. We developed an MR-based OAR autosegmentation atlas and evaluated its performance compared to manual delineation. MATERIALS AND METHODS: Anonymized cerebral T1-weighted MR scans (voxel size 0.9 × 0.9 × 0.9 mm3) were available. OARs were manually delineated according to international consensus. Fifty MR scans were used to develop the autosegmentation atlas in a commercially available treatment planning system (Raystation®). The performance of this atlas was tested on another 40 MR scans by automatically delineating 34 OARs, as defined by the 2018 EPTN consensus. Spatial overlap between manual and automated delineations was determined by calculating the Dice similarity coefficient (DSC). Two radiation oncologists determined the quality of each automatically delineated OAR. The time needed to delineate all OARs manually or to adjust automatically delineated OARs was determined. RESULTS: DSC was ≥ 0.75 in 31 (91 %) out of 34 automated OAR delineations. Delineations were rated by radiation oncologists as excellent or good in 29 (85 %) out 34 OAR delineations, while 4 were rated fair (12 %) and 1 was rated poor (3 %). Interobserver agreement between the radiation oncologists ranged from 77-100 % per OAR. The time to manually delineate all OARs was 88.5 minutes, while the time needed to adjust automatically delineated OARs was 15.8 minutes. CONCLUSION: Autosegmentation of OARs enables high-quality contouring within a limited time. Accurate OAR delineation helps to define OAR constraints to mitigate serious complications and helps with the development of NTCP models.

Accelerated Partial Breast Irradiation Using External Beam or Intraoperative Electron Radiation Therapy: 5-Year Oncological Outcomes of a Prospective Cohort Study.

PURPOSE: To evaluate the ipsilateral breast tumor recurrence (IBTR) after 2 accelerated partial breast irradiation (APBI) techniques (intraoperative electron radiation therapy [IOERT] and external beam APBI [EB-APBI]) in patients with early-stage breast cancer. METHODS AND MATERIALS: Between 2011 and 2016, women ≥60 years of age with breast carcinoma or Ductal Carcinoma In Situ (DCIS) of ≤30 mm and cN0 undergoing breast-conserving therapy were included in a 2-armed prospective multicenter cohort study. IOERT (1 × 23.3 Gy prescribed at the 100% isodose line) was applied in 1 hospital and EB-APBI (10 × 3.85 Gy daily) in 2 other hospitals. The primary endpoint was IBTR (all recurrences in the ipsilateral breast irrespective of localization) at 5 years after lumpectomy. A competing risk model was used to estimate the cumulative incidences of IBTR, which were compared using Fine and Gray's test. Secondary endpoints were locoregional recurrence rate, distant recurrence, disease-specific survival and overall survival. Univariate Cox regression models were estimated to identify risk factors for IBTR. Analyses were performed of the intention to treat (ITT) population (IOERT n = 305; EB-APBI n = 295), and sensitivity analyses were done of the per-protocol population (IOERT n = 270; EB-APBI n = 207). RESULTS: The median follow-up was 5.2 years (IOERT) and 5 years (EB-APBI). Cumulative incidence of IBTR in the ITT population at 5 years after lumpectomy was 10.6% (95% confidence interval, 7.0%-14.2%) after IOERT and 3.7% (95% confidence interval, 1.2%-5.9%) after EB-APBI (P = .002). The locoregional recurrence rate was significantly higher after IOERT than EB-APBI (12.1% vs 4.5%, P = .001). There were no differences between groups in other endpoints. Sensitivity analysis showed similar results. For both groups, no significant risk factors for IBTR were identified in the ITT population. In the per-protocol population, surgical margin status of the DCIS was the only significant risk factor for developing IBTR in both treatment groups. CONCLUSIONS: Ipsilateral breast tumor recurrences and locoregional recurrence rates were unexpectedly high in patients treated with IOERT, and acceptable in patients treated with EB-APBI.

Health-related quality of life of early-stage breast cancer patients after different radiotherapy regimens.

PURPOSE: To evaluate and compare health-related quality of life (HRQL) of women with early-stage breast cancer (BC) treated with different radiotherapy (RT) regimens. METHODS: Data were collected from five prospective cohorts of BC patients treated with breast-conserving surgery and different RT regimens: intraoperative RT (IORT, 1 × 23.3 Gy; n = 267), external beam accelerated partial breast irradiation (EB-APBI, 10 × 3.85 Gy; n = 206), hypofractionated whole breast irradiation(hypo-WBI, 16 × 2.67 Gy; n = 375), hypo-WBI + boost(hypo-WBI-B, 21-26 × 2.67 Gy; n = 189), and simultaneous WBI + boost(WBI-B, 28 × 2.3 Gy; n = 475). Women ≥ 60 years with invasive/in situ carcinoma ≤ 30 mm, cN0 and pN0-1a were included. Validated EORTC QLQ-C30/BR23 questionnaires were used to asses HRQL. Multivariable linear regression models adjusted for confounding (age, comorbidity, pT, locoregional treatment, systemic therapy) were used to compare the impact of the RT regimens on HRQL at 12 and 24 months. Differences in HRQL over time (3-24 months) were evaluated using linear mixed models. RESULTS: There were no significant differences in HRQL at 12 months between groups except for breast symptoms which were better after IORT and EB-APBI compared to hypo-WBI at 12 months (p < 0.001). Over time, breast symptoms, fatigue, global health status and role functioning were significantly better after IORT and EB-APBI than hypo-WBI. At 24 months, HRQL was comparable in all groups. CONCLUSION: In women with early-stage breast cancer, the radiotherapy regimen did not substantially influence long-term HRQL with the exception of breast symptoms. Breast symptoms are more common after WBI than after IORT or EB-APBI and improve slowly until no significant difference remains at 2 years posttreatment.

Lower doses to hippocampi and other brain structures for skull-base meningiomas with intensity modulated proton therapy compared to photon therapy.

BACKGROUND AND PURPOSE: Radiotherapy of skull-base meningiomas is challenging due to the close proximity of multiple sensitive organs at risk (OARs). This study systematically compared intensity modulated proton therapy (IMPT), non-coplanar volumetric modulated arc therapy (VMAT) and intensity modulated radiotherapy (IMRT) based on automated treatment planning. Differences in OARs sparing, with specific focus on the hippocampi, and low-dose delivery were quantified. MATERIALS AND METHODS: Twenty patients, target diameter >3 cm, were included. Automated plan generation was used to calculate a VMAT plan with three non-coplanar arcs, an IMRT plan with nine non-coplanar beams with optimized gantry and couch angles, and an IMPT plan with three patient-specific selected non-coplanar beams. A prescription dose of 50.4 GyRBE in 28 fractions was used. The same set of constraints and prioritized objectives was used. All plans were rescaled to the same target coverage. Repeated measures ANOVA was used to assess the statistical significance of differences in OAR dose parameters between planning techniques. RESULTS: Compared to VMAT and IMRT, IMPT significantly improved dose conformity to the target volume. Consequently, large dose reductions in OARs were observed. With respect to VMAT, the mean dose and D40% in the bilateral hippocampus were on average reduced by 48% and 74%, respectively (p ≤ 0.005). With IMPT, the mean dose in the normal brain and volumes receiving 20-30 Gy were up to 47% lower (p ≤ 0.01). When comparing IMPT and IMRT, even larger dose differences in those OARs were observed. CONCLUSION: For skull-base meningiomas IMPT allows for a considerable dose reduction in the hippocampi, normal brain and other OARs compared to both non-coplanar VMAT and IMRT, which may lead to a clinically relevant reduction of late neurocognitive side effects.

Whole breast proton irradiation for maximal reduction of heart dose in breast cancer patients.

PURPOSE: In left-sided breast cancer radiotherapy, tangential intensity modulated radiotherapy combined with breath-hold enables a dose reduction to the heart and left anterior descending (LAD) coronary artery. Aim of this study was to investigate the added value of intensity modulated proton therapy (IMPT) with regard to decreasing the radiation dose to these structures. METHODS: In this comparative planning study, four treatment plans were generated in 20 patients: an IMPT plan and a tangential IMRT plan, both with breath-hold and free-breathing. At least 97 % of the target volume had to be covered by at least 95 % of the prescribed dose in all cases. Specifically with respect to the heart, the LAD, and the target volumes, we analyzed the maximum doses, the mean doses, and the volumes receiving 5-30 Gy. RESULTS: As compared to IMRT, IMPT resulted in significant dose reductions to the heart and LAD-region even without breath-hold. In the majority of the IMPT cases, a reduction to almost zero to the heart and LAD-region was obtained. IMPT treatment plans yielded the lowest dose to the lungs. CONCLUSIONS: With IMPT the dose to the heart and LAD-region could be significantly decreased compared to tangential IMRT with breath-hold. The clinical relevance should be assessed individually based on the baseline risk of cardiac complications in combination with the dose to organs at risk. However, as IMPT for breast cancer is currently not widely available, IMPT should be reserved for patients remaining at high risk for major coronary events.

Left-sided breast cancer radiotherapy with and without breath-hold: does IMRT reduce the cardiac dose even further?

PURPOSE: In radiotherapy for left-sided breast cancer, Active Breathing Control enables a decrease of cardiac and Left Anterior Descending (LAD) coronary artery dose. We compared 3D-Conformal (3D-CRT) to Intensity Modulated Radiotherapy (IMRT) treatment plans based on free-breathing (FB) and breath-hold (BH). We investigated whether IMRT enables an additional decrease of cardiac dose in radiotherapy plans with and without BH. METHODS AND MATERIALS: Twenty patients referred for whole breast irradiation were included. The whole breast, heart and LAD-region were contoured. Four treatment plans were generated: FB_3D-CRT; FB_IMRT; BH_3D-CRT; BH_IMRT. Several doses were obtained from Dose Volume Histograms and compared. Results were compared statistically using the Wilcoxin Signed Rank Test. For heart and LAD-region, a significant dose reduction was found in BH (p<0.01). For both BH and FB, a significant dose reduction was found using IMRT (p<0.01). By using IMRT an average reduction of 5% was noted in the LAD-region for the volume receiving 20Gy. In 5 cases, the LAD-region remained situated in the vicinity of the radiation portals even in BH. Nevertheless, with IMRT the LAD dose was reduced in these cases. CONCLUSION: IMRT results in a significant additional decrease of dose in the heart and LAD-region in both breath-hold and free-breathing.

MRI- versus CT-based volume delineation of lumpectomy cavity in supine position in breast-conserving therapy: an exploratory study.

PURPOSE: To examine magnetic resonance imaging (MRI) and computed tomography (CT) for lumpectomy cavity (LC) volume delineation in supine radiotherapy treatment position and to assess the interobserver variability. METHODS AND MATERIALS: A total of 15 breast cancer patients underwent a planning CT and directly afterward MRI in supine radiotherapy treatment position. Then, 4 observers (2 radiation oncologists and 2 radiologists) delineated the LC on the CT and MRI scans and assessed the cavity visualization score (CVS). The CVS, LC volume, conformity index (CI), mean shift of the center of mass (COM), with the standard deviation, were quantified for both CT and MRI. RESULTS: The CVS showed that MRI and CT provide about equal optimal visibility of the LC. If the CVS was high, magnetic resonance imaging provided more detail of the interfaces of the LC seroma with the unaffected GBT. MRI also pictured in more detail the interfaces of axillary seromas (if present) with their surroundings and their relationship to the LC. Three observers delineated smaller, and one observer larger, LC volumes comparing the MRI- and CT-derived delineations. The mean ± standard deviation CI was 32% ± 25% for MRI and 52% ± 21% for CT. The mean ± standard deviation COM shift was 11 ± 10 mm (range 1-36) for MRI and 4 ± 3 mm (range 1-10) for CT. CONCLUSIONS: MRI does not add additional information to CT in cases in which the CVS is assessed as low. The conformity (CI) is lower for MRI than for CT, especially at a low CVS owing to greater COM shifts for MRI, probably caused by inadequate visibility of the surgical clips on magnetic resonance (MR) images. The COM shifts seriously dictate a decline in the CI more than the variability of the LC volumes does. In cases in which MRI provides additional information, MRI must be combined with the CT/surgical clip data.

Preradiotherapy calcium scores of the coronary arteries in a cohort of women with early-stage breast cancer: a comparison with a cohort of healthy women.

PURPOSE: Breast cancer radiotherapy has been associated with an increased risk of cardiac toxicity. However, no data are available on the probability of developing coronary artery disease (CAD) in breast cancer patients when compared with healthy women. Therefore, baseline coronary artery calcium (CAC) scores, as an accepted tool to predict CAD, were determined and compared with the CAC scores of a healthy, asymptomatic cohort, the Multi-Ethnic Study of Atherosclerosis (MESA) cohort. METHODS AND MATERIALS: Eighty consecutive patients with ductal carcinoma in situ or infiltrative breast cancer referred for radiotherapy after breast-conserving surgery were included in our study. Their cardiovascular risk profile was registered, and a 64 multislice CT scan was performed. The CAC scores of an unselected (Caucasian only) Radiotherapy Centre West (RCWEST) cohort, as well as of those of a selected (comorbidity and race adjusted) RCWEST cohort, were determined. The scores of both cohorts were compared with those of the female (Caucasian only) MESA cohort. RESULTS: For the unselected RCWEST cohort (n = 62) we found significant (p < .01) higher scores for women in the 55-64 age category compared with those of the MESA cohort. In the selected cohort (n = 55) the CAC scores of the women in the age category 55-64 were significantly (p = .02) higher compared with the MESA cohort. No significant differences were noted in the other age categories. CONCLUSION: Both cohorts revealed that CAC scores in the 55-64 age category were significantly higher than the CAC scores in the asymptomatic (female) MESA population. These data suggest that breast cancer patients bear a higher risk of developing coronary heart disease before the start of radiotherapy. Therefore, measures to decrease cardiac dose further in breast cancer radiotherapy are even more important.

Magnetic resonance imaging- versus computed tomography-based target volume delineation of the glandular breast tissue (clinical target volume breast) in breast-conserving therapy: an exploratory study.

PURPOSE: To examine MRI and CT for glandular breast tissue (GBT) volume delineation and to assess interobserver variability. METHODS AND MATERIALS: Fifteen breast cancer patients underwent a planning CT and MRI, consecutively, in the treatment position. Four observers (two radiation oncologists and two radiologists) delineated the GBT according to the CT and separately to the MR images. Volumes, centers of mass, maximum extensions with standard deviations (SD), and interobserver variability were quantified. Observers viewed delineation differences between MRI and CT and delineation differences among observers. RESULTS: In cranio-lateral and cranio-medial directions, GBT volumes were delineated larger using MRI when compared with those delineated with CT. Center of mass on MRI shifted a mean (SD) 17% (4%) into the cranial direction and a mean 3% (4%) into the dorsal direction when compared with that on the planning CT. Only small variations between observers were noted. The GBT volumes were approximately 4% larger on MRI (mean [SD] ratio MRI to CT GBT volumes, 1.04 [0.06]). Findings were concordant with viewed MRI and CT images and contours. Conformity indices were only slightly different; mean conformity index was 77% (3%) for MRI and 79% (4%) for CT. Delineation differences arising from personal preferences remained recognizable irrespective of the imaging modality used. CONCLUSIONS: Contoured GBT extends substantially further into the cranio-lateral and cranio-medial directions on MRI when compared with CT. Interobserver variability is comparable for both imaging modalities. Observers should be aware of existing personal delineation preferences. Institutions are recommended to review and discuss target volume delineations and to design supplementary guidelines if necessary.

Stereotactic radiotherapy of intracranial tumors: a comparison of intensity-modulated radiotherapy and dynamic conformal arc.

PURPOSE: Intensity-modulated radiotherapy (IMRT) and dynamic conformal arc (DCA) are two state-of-the-art techniques for linac-based stereotactic radiotherapy (SRT) using the micromultileaf collimator. The purpose of this planning study is to examine the relative merits of these techniques in the treatment of intracranial tumors. MATERIALS AND METHODS: SRT treatment plans were made for 25 patients with a glioma or meningioma. For all patients, we made an IMRT and a DCA plan. Plans were evaluated using: target coverage, conformity index (CI), homogeneity index (HI), doses in critical structures, number of monitor units needed, and equivalent uniform dose (EUD) in planning target volume (PTV) and critical structures. RESULTS: In the overall comparison of both techniques, we found adequate target coverage in all cases; a better mean CI with IMRT in concave tumors (p = 0.027); a better mean HI with DCA in meningiomas, complex tumors, and small (< 92 mL) tumors (p = 0.000, p = 0.005, and p = 0.005, respectively); and a higher EUD in the PTV with DCA in convex tumors (gliomas) and large tumors (p = 0.000 and p = 0.003, respectively). In all patients, significantly more monitor units were needed with IMRT. The results of the overall comparison did not enable us to predict the preference for one of the techniques in individual patients. The DCA plan was acceptable in 23 patients and the IMRT plan in 19 patients. DCA was preferred in 18 of 25 patients. CONCLUSIONS: DCA is our preferred SRT technique for most intracranial tumors. Tumor type, size, or shape do not predict a preference for DCA or IMRT.

Does 4 MV perform better compared to 6 MV in the presence of air cavities in the head and neck region?

BACKGROUND AND PURPOSE: The underdose near air cavities in the head and neck region at photon energies of 4 MV and 6 MV was studied in search for clinical advantages of the 4 MV over 6 MV treatments. MATERIALS AND METHODS: The on-axis and off-axis dose distributions were measured with a parallel-plate ionization chamber and films in polystyrene phantoms containing an air cavity of appropriate size based on the results of computed tomography scans. RESULTS: Although most results are similar for both energies, the 4 MV photon beams give a somewhat smaller underdose effect and a faster re-build up than the 6 MV. For both energies a significant underdose effect was observed at the edge of the field in the larynx phantom. This proved to be true for small and large fields, for smaller and larger cavities, for one-beam as well as parallel-opposed beams. CONCLUSION: For most clinically relevant situations there is no remarkable benefit in the use of either of the two energies.

How to assess post-occlusive reactive hyperaemia by means of laser Doppler perfusion monitoring: application of a standardised protocol to patients with peripheral arterial obstructive disease.

The standardisation of manoeuvres to perform clinically discriminative microvascular flow reserve tests is still poorly developed, as well as the response analysis. The aim of this study was to establish a reproducible analysis method for the post-occlusive reactive hyperaemia (PORH) test measured using laser Doppler perfusion monitoring (LDPM). LDPM data were measured from the PORH response of 24 Fontaine class II-III peripheral atherosclerotic/arterial obstructive disease (PAOD) patients and 30 healthy subjects. The PORH response was recorded from the dorsum of the foot after 3 min of arterial occlusion at the thigh. The resulting tracings were analysed by describing their morphology through five defined parameters: resting flux (RF), time to RF level (tRF), maximum flux (MF) during reactive hyperaemia, time to maximum flux (tMF), and time to half recovery (tHR). While the time parameters were discriminative between patients and controls, flux parameters were not. The time to resting flux (tRF) led to the most discriminative model that correctly predicted 88.5% of the cases. Hence, we concluded that obtaining t(RF) with the presented procedures provides an optimal model to quantify the patient's microvascular condition from the PORH response.

Suppression of dynamic laser speckle signals in multimode fibers of various lengths.

The effects of fiber coupling and fiber length on photocurrent fluctuations are studied when the light of a laser diode transmitted to and from a dynamic turbid medium by a step-index multimode fiber is studied. When the laser light is coupled asymmetrically, filling only the higher-order modes, the photocurrent fluctuations are suppressed significantly when fiber lengths of as much as 16 m are added between the laser and the medium. Addition of as much as 16 m of detection fiber, or any fiber in the case of symmetric light coupling, leads to much less or no suppression of the photocurrent fluctuations.

Glass-fiber self-mixing intra-arterial laser Doppler velocimetry: signal stability and feedback analysis.

We have developed a blood velocimeter based on the principle of self-mixing in a semiconductor laser diode through an optical fiber. The intensity of the light is modulated by feedback from moving scattering particles that contain the Doppler-shift frequency. Upon feedback the characteristics of the laser diode change. The threshold current decreases, and an instable region may become present above the new threshold. The amplitude of the Doppler signal turns out to be related to the difference in intensity between situations with and without feedback. This amplitude is highest just above feedback. The suppression of reflection from the glass-fiber facets is of paramount importance in the obtaining of a higher signal-to-noise ratio. Using an optical stabilization of the feedback, we optimized the performance of the laser-fiber system and the Doppler modulation depth and clarified its behavior with a suitable physical model. We also investigated the effect of the finite coherence length of the laser. We tested the efficiency of the self-mixing velocimeter in vivo with the optical glass fiber inserted in the artery with endoscopic catheters, both in upstream and in downstream blood flow conditions. For the latter we used a special side-reflecting device solution for the fiber facet to allow downstream measurements.