Assessing robustness of radiomic features by image perturbation

Image features need to be robust against differences in positioning, acquisition and segmentation to ensure reproducibility. Radiomic models that only include robust features can be used to analyse new images, whereas models with non-robust features may fail to predict the outcome of interest accurately. Test-retest imaging is recommended to assess robustness, but may not be available for the phenotype of interest. We therefore investigated 18 combinations of image perturbations to determine feature robustness, based on noise addition (N), translation (T), rotation (R), volume growth/shrinkage (V) and supervoxel-based contour randomisation (C). Test-retest and perturbation robustness were compared for combined total of 4032 morphological, statistical and texture features that were computed from the gross tumour volume in two cohorts with computed tomography imaging: I) 31 non-small-cell lung cancer (NSCLC) patients; II): 19 head-and-neck squamous cell carcinoma (HNSCC) patients. Robustness was determined using the 95% confidence interval (CI) of the intraclass correlation coefficient (1, 1). Features with CI ≥ 0:90 were considered robust. The NTCV, TCV, RNCV and RCV perturbation chain produced similar results and identified the fewest false positive robust features (NSCLC: 0.2–0.9%; HNSCC: 1.7–1.9%). Thus, these perturbation chains may be used as an alternative to test-retest imaging to assess feature robustness.

Intensity-modulated proton therapy decreases dose to organs at risk in low-grade glioma patients: results of a multicentric in silico ROCOCO trial

BACKGROUND AND PURPOSE

Patients with low-grade glioma (LGG) have a prolonged survival expectancy due to better discriminative tumor classification and multimodal treatment. Consequently, long-term treatment toxicity gains importance. Contemporary radiotherapy techniques such as intensity-modulated radiotherapy (IMRT), volumetric modulated arc therapy (VMAT), tomotherapy (TOMO) and intensity-modulated proton therapy (IMPT) enable high-dose irradiation of the target but they differ regarding delivered dose to organs at risk (OARs). The aim of this comparative in silico study was to determine these dosimetric differences in delivered doses.

MATERIAL AND METHODS

Imaging datasets of 25 LGG patients having undergone postoperative radiotherapy were included. For each of these patients, in silico treatment plans to a total dose of 50.4 Gy to the target volume were generated for the four treatment modalities investigated (i.e., IMRT, VMAT, TOMO, IMPT). Resulting treatment plans were analyzed regarding dose to target and surrounding OARs comparing IMRT, TOMO and IMPT to VMAT.

RESULTS

In total, 100 treatment plans (four per patient) were analyzed. Compared to VMAT, the IMPT mean dose (D_mean) for nine out of 10 (90%) OARs was statistically significantly (p < .02) reduced, for TOMO this was true in 3/10 (30%) patients and for 1/10 (10%) patients for IMRT. IMPT was the prime modality reducing dose to the OARs followed by TOMO.

DISCUSSION

The low dose volume to the majority of OARs was significantly reduced when using IMPT compared to VMAT. Whether this will lead to a significant reduction in neurocognitive decline and improved quality of life is to be determined in carefully designed future clinical trials.

Comparison of different treatment planning approaches for intensity-modulated proton therapy with simultaneous integrated boost for pancreatic cancer

BACKGROUND

Neoadjuvant radio(chemo)therapy of non-metastasized, borderline resectable or unresectable locally advanced pancreatic cancer is complex and prone to cause side-effects, e.g., in gastrointestinal organs. Intensity-modulated proton therapy (IMPT) enables a high conformity to the targets while simultaneously sparing the normal tissue such that dose-escalation strategies come within reach. In this in silico feasibility study, we compared four IMPT planning strategies including robust multi-field optimization (rMFO) and a simultaneous integrated boost (SIB) for dose-escalation in pancreatic cancer patients.

METHODS

For six pancreatic cancer patients referred for adjuvant or primary radiochemotherapy, four rMFO-IMPT-SIB treatment plans each, consisting of two or three (non-)coplanar beam arrangements, were optimized. Dose values for both targets, i.e., the elective clinical target volume [CTV, prescribed dose D_pres = 51Gy(RBE)] and the boost target [D_pres = 66Gy(RBE)], for the organs at risk as well as target conformity and homogeneity indexes, derived from the dose volume histograms, were statistically compared.

RESULTS

All treatment plans of each strategy fulfilled the prescribed doses to the targets (D_{pres(GTV,CTV)} = 100%, D_{95%,(GTV,CTV)} ≥ 95%, D_2%,(GTV,CTV) ≤ 107%). No significant differences for the conformity index were found (p > 0.05), however, treatment plans with a three non-coplanar beam strategy were most homogenous to both targets (p < 0.045). The median value of all dosimetric results of the large and small bowel as well as for the liver and the spinal cord met the dose constraints with all beam arrangements. Irrespective of the planning strategies, the dose constraint for the duodenum and stomach were not met. Using the three-beam arrangements, the dose to the left kidney could be significant decreased when compared to a two-beam strategy (p < 0.045).

CONCLUSION

Based on our findings we recommend a three-beam configuration with at least one non-coplanar beam for dose-escalated SIB with rMFO-IMPT in advanced pancreatic cancer patients achieving a homogeneous dose distribution in the target while simultaneously minimizing the dose to the organs at risk. Further treatment planning studies on aspects of breathing and organ motion need to be performed.

CT imaging during treatment improves radiomic models for patients with locally advanced head and neck cancer

BACKGROUND AND PURPOSE

The development of radiomic risk models to predict clinical outcome is usually based on pre-treatment imaging, such as computed tomography (CT) scans used for radiation treatment planning. Imaging data acquired during the course of treatment may improve their prognostic performance. We compared the performance of radiomic risk models based on the pre-treatment CT and CT scans acquired in the second week of therapy.

MATERIAL AND METHODS

Treatment planning and second week CT scans of 78 head and neck squamous cell carcinoma patients treated with primary radiochemotherapy were collected. 1538 image features were extracted from each image. Prognostic models for loco-regional tumour control (LRC) and overall survival (OS) were built using 6 feature selection methods and 6 machine learning algorithms. Prognostic performance was assessed using the concordance index (C-Index). Furthermore, patients were stratified into risk groups and differences in LRC and OS were evaluated by log-rank tests.

RESULTS

The performance of radiomic risk model in predicting LRC was improved using the second week CT scans (C-Index: 0.79), in comparison to the pre-treatment CT scans (C-Index: 0.65). This was confirmed by Kaplan-Meier analyses, in which risk stratification based on the second week CT could be improved for LRC (p = 0.002) compared to pre-treatment CT (p = 0.063).

CONCLUSION

Incorporation of imaging during treatment may be a promising way to improve radiomic risk models for clinical treatment adaption, i.e., to select patients that may benefit from dose modification.

Melanoma Brain Metastases: Local Therapies, Targeted Therapies, Immune Checkpoint Inhibitors and Their Combinations-Chances and Challenges

Recent phase II trials have shown that BRAF/MEK inhibitors and immune checkpoint inhibitors are active in patients with melanoma brain metastases (MBM), reporting intracranial disease control rates of 50-75%. Furthermore, retrospective analyses suggest that combining stereotactic radiosurgery with immune checkpoint inhibitors or BRAF/MEK inhibitors prolongs overall survival. These data stress the need for inter- and multidisciplinary cooperation that takes into account the individual prognostic factors in order to establish the best treatment for each patient. Although the management of MBM has dramatically improved, a substantial number of patients still progress and die from brain metastases. Therefore, there is an urgent need for prospective studies in patients with MBM that focus on treatment combinations and sequences, new treatment strategies, and biomarkers of treatment response. Moreover, further research is needed to decipher brain-specific mechanisms of therapy resistance.

Applicability of a prognostic CT-based radiomic signature model trained on stage I-III non-small cell lung cancer in stage IV non-small cell lung cancer

OBJECTIVES

Recently it has been shown that radiomic features of computed tomography (CT) have prognostic information in stage I-III non-small cell lung cancer (NSCLC) patients. We aim to validate this prognostic radiomic signature in stage IV adenocarcinoma patients undergoing chemotherapy.

MATERIALS AND METHODS

Two datasets of chemo-naive stage IV adenocarcinoma patients were investigated, dataset 1: 285 patients with CTs performed in a single center; dataset 2: 223 patients included in a multicenter clinical trial. The main exclusion criteria were EGFR mutation or unknown mutation status and non-delineated primary tumor. Radiomic features were calculated for the primary tumor. The c-index of cox regression was calculated and compared to the signature performance for overall survival (OS).

RESULTS

In total CT scans from 195 patients were eligible for analysis. Patients having a prognostic index (PI) lower than the signature median (n = 92) had a significantly better OS than patients with a PI higher than the median (n = 103, HR 1.445, 95% CI 1.07-1.95, p = 0.02, c-index 0.576, 95% CI 0.527-0.624).

CONCLUSION

The radiomic signature, derived from daily practice CT scans, has prognostic value for stage IV NSCLC, however the signature performs less than previously described for stage I-III NSCLC stages. In the future, machine learning techniques can potentially lead to a better prognostic imaging based model for stage IV NSCLC.

Clinical Feasibility of Single-Source Dual-spiral 4D Dual-Energy CT for Proton Treatment Planning Within the Thoracic Region

PURPOSE

Single-source dual-spiral dual-energy computed tomography (DECT) provides additional patient information but is prone to motion between the 2 consecutively acquired computed tomography (CT) scans. Here, the clinical applicability of dual-spiral time-resolved DECT (4D-DECT) for proton treatment planning within the thoracic region was evaluated.

METHODS AND MATERIALS

Dual-spiral 4D-DECT scans of 3 patients with lung cancer were acquired. For time-averaged datasets and 4 breathing phases, the geometric conformity of 80 kVp and 140 kVp 4D-DECT scans before image post-processing was assessed by normalized cross correlation (NCC). Additionally, the conformity of the corresponding DECT-derived 58 keV and 79 keV pseudo-monoenergetic CT datasets after image post-processing, including deformable image registration (DIR), was determined. To analyze the reliability of proton dose calculation, clinical (Plan_Clin) and artificial worst-case (Plan_WorstCase, targeting the diaphragm) treatment plans were calculated on 140 kVp and 79 keV datasets and compared with gamma analyses (0.1% dose-difference and 1 mm distance-to-agreement criterion). The applicability of a patient-specific DECT-based prediction of stopping-power ratio (SPR) was investigated and proton range shifts compared with the clinical heuristic CT-number-to-SPR conversion were assessed. Finally, the delineation variability of an experienced radiation oncologist was quantified.

RESULTS

Dual-spiral 4D-DECT scans without DIR showed a high geometric conformity, with an average NCC ± standard deviation of 98.7% ± 1.0% when including all patient voxels or 88.2% ± 7.8% when considering only lung. DIR improved the conformity, leading to an average NCC of 99.9% ± 0.1% and 99.6% ± 0.5%, respectively. Plan_Clin dose distributions on 140 kVp and 79 keV datasets were similar, with an average gamma passing rate of 99.9% (99.2%-100%). The worst-case evaluation still revealed high passing rates (99.3% on average, 92.4% as minimum). Clinically relevant mean range shifts of 2.2% ± 1.2% were determined between patient-specific DECT-based SPR prediction and clinical heuristic CT-number-to-SPR conversion. The intra-observer delineation variability was slightly reduced using additional DECT-derived datasets.

CONCLUSIONS

The 79 keV pseudo-monoenergetic CT datasets can be consistently obtained from dual-spiral 4D-DECT and are applicable for dose calculation. Patient-specific DECT-based SPR prediction performed well and potentially reduces range uncertainty in proton therapy of patients with lung cancer.

Prospective data registration and clinical trials for particle therapy in Europe

The overarching aim of work package 1 of the European Proton Therapy Network (EPTN) is to create a firm basis for evidence-based particle therapy at the European level. To achieve this, this work package will set up a worldwide unique prospective data registration programme for nine different tumour sites. Such programme will provide more insights into the current practice across all European particle therapy centres and into the results of particle therapy with regard to radiation-induced toxicity and efficacy in terms of local control and survival. More importantly, prospective data registration provides major opportunities to continuously improve the quality of particle therapy, by defining bench marks, to identify best practices that may learn others to improve quality of particle therapy, to synchronize selection criteria and to create more homogeneous patient cohorts to evaluate results, which is particularly important in rare tumours. This will be supported by EORTC through existing and new IT-infrastructure for data collection in different formats next to QA-platforms. In addition, work package 1 will define the requirements for high quality clinical trials in order to enhance high quality clinical trial proposals and determine alternative methods for RCT, such as the model-based approach.

The EPTN consensus-based atlas for CT- and MR-based contouring in neuro-oncology

PURPOSE

To create a digital, online atlas for organs at risk (OAR) delineation in neuro-oncology based on high-quality computed tomography (CT) and magnetic resonance (MR) imaging.

METHODS

CT and 3 Tesla (3T) MR images (slice thickness 1 mm with intravenous contrast agent) were obtained from the same patient and subsequently fused. In addition, a 7T MR without intravenous contrast agent was obtained from a healthy volunteer. Based on discussion between experienced radiation oncologists, the clinically relevant organs at risk (OARs) to be included in the atlas for neuro-oncology were determined, excluding typical head and neck OARs previously published. The draft atlas was delineated by a senior radiation oncologist, 2 residents in radiation oncology, and a senior neuro-radiologist incorporating relevant available literature. The proposed atlas was then critically reviewed and discussed by European radiation oncologists until consensus was reached.

RESULTS

The online atlas includes one CT-scan at two different window settings and one MR scan (3T) showing the OARs in axial, coronal and sagittal view. This manuscript presents the three-dimensional descriptions of the fifteen consensus OARs for neuro-oncology. Among these is a new OAR relevant for neuro-cognition, the posterior cerebellum (illustrated on 7T MR images).

CONCLUSION

In order to decrease inter- and intra-observer variability in delineating OARs relevant for neuro-oncology and thus derive consistent dosimetric data, we propose this atlas to be used in photon and particle therapy. The atlas is available online at www.cancerdata.org and will be updated whenever required.

18F-fluorodeoxyglucose positron-emission tomography (FDG-PET)-Radiomics of metastatic lymph nodes and primary tumor in non-small cell lung cancer (NSCLC) - A prospective externally validated study

BACKGROUND

Lymph node stage prior to treatment is strongly related to disease progression and poor prognosis in non-small cell lung cancer (NSCLC). However, few studies have investigated metabolic imaging features derived from pre-radiotherapy 18F-fluorodeoxyglucose (FDG) positron-emission tomography (PET) of metastatic hilar/mediastinal lymph nodes (LNs). We hypothesized that these would provide complementary prognostic information to FDG-PET descriptors to only the primary tumor (tumor).

METHODS

Two independent cohorts of 262 and 50 node-positive NSCLC patients were used for model development and validation. Image features (i.e. Radiomics) including shape and size, first order statistics, texture, and intensity-volume histograms (IVH) (http://www.radiomics.io/) were evaluated by univariable Cox regression on the development cohort. Prognostic modeling was conducted with a 10-fold cross-validated least absolute shrinkage and selection operator (LASSO), automatically selecting amongst FDG-PET-Radiomics descriptors from (1) tumor, (2) LNs or (3) both structures. Performance was assessed with the concordance-index. Development data are publicly available at www.cancerdata.org and Dryad (doi:10.5061/dryad.752153b).

RESULTS

Common SUV descriptors (maximum, peak, and mean) were significantly related to overall survival when extracted from LNs, as were LN volume and tumor load (summed tumor and LNs' volumes), though this was not true for either SUV metrics or tumor's volume. Feature selection exclusively from imaging information based on FDG-PET-Radiomics, exhibited performances of (1) 0.53 -external 0.54, when derived from the tumor, (2) 0.62 -external 0.56 from LNs, and (3) 0.62 -external 0.59 from both structures, including at least one feature from each sub-category, except IVH.

CONCLUSION

Combining imaging information based on FDG-PET-Radiomics features from tumors and LNs is desirable to achieve a higher prognostic discriminative power for NSCLC.

Combined PET/MRI: Global Warming-Summary Report of the 6th International Workshop on PET/MRI, March 27-29, 2017, Tübingen, Germany

The 6th annual meeting to address key issues in positron emission tomography (PET)/magnetic resonance imaging (MRI) was held again in Tübingen, Germany, from March 27 to 29, 2017. Over three days of invited plenary lectures, round table discussions and dialogue board deliberations, participants critically assessed the current state of PET/MRI, both clinically and as a research tool, and attempted to chart future directions. The meeting addressed the use of PET/MRI and workflows in oncology, neurosciences, infection, inflammation and chronic pain syndromes, as well as deeper discussions about how best to characterise the tumour microenvironment, optimise the complementary information available from PET and MRI, and how advanced data mining and bioinformatics, as well as information from liquid biomarkers (circulating tumour cells and nucleic acids) and pathology, can be integrated to give a more complete characterisation of disease phenotype. Some issues that have dominated previous meetings, such as the accuracy of MR-based attenuation correction (AC) of the PET scan, were finally put to rest as having been adequately addressed for the majority of clinical situations. Likewise, the ability to standardise PET systems for use in multicentre trials was confirmed, thus removing a perceived barrier to larger clinical imaging trials. The meeting openly questioned whether PET/MRI should, in all cases, be used as a whole-body imaging modality or whether in many circumstances it would best be employed to give an in-depth study of previously identified disease in a single organ or region. The meeting concluded that there is still much work to be done in the integration of data from different fields and in developing a common language for all stakeholders involved. In addition, the participants advocated joint training and education for individuals who engage in routine PET/MRI. It was agreed that PET/MRI can enhance our understanding of normal and disrupted biology, and we are in a position to describe the in vivo nature of disease processes, metabolism, evolution of cancer and the monitoring of response to pharmacological interventions and therapies. As such, PET/MRI is a key to advancing medicine and patient care.

Photon vs. proton radiochemotherapy: Effects on brain tissue volume and perfusion

BACKGROUND AND PURPOSE

To compare the structural and hemodynamic changes of healthy brain tissue in the cerebral hemisphere contralateral to the tumor following photon and proton radiochemotherapy.

MATERIALS AND METHODS

Sixty-seven patients (54.9 ±14.0 years) diagnosed with glioblastoma undergoing adjuvant photon (n = 47) or proton (n = 19) radiochemotherapy with temozolomide after tumor resection underwent T1-weighted and arterial spin labeling MRI. Changes in volume and perfusion before and 3 to 6 months after were compared between therapies.

RESULTS

A decrease in gray matter (GM) (-2.2%, P<0.001) and white matter (WM) (-1.2%, P<0.001) volume was observed in photon-therapy patients compared to the pre-radiotherapy baseline. In contrast, for the proton-therapy group, no significant differences in GM (0.3%, P = 0.64) or WM (-0.4%, P = 0.58) volume were observed. GM volume decreased with 0.9% per 10 Gy dose increase (P<0.001) and differed between the radiation modalities (P<0.001). Perfusion decreased in photon-therapy patients (-10.1%, P = 0.002), whereas the decrease in proton-therapy patients, while comparable in magnitude, did not reach statistical significance (-9.1%, P = 0.12). There was no correlation between perfusion decrease and either dose (P = 0.64) or radiation modality (P = 0.94).

CONCLUSIONS

Our results show that the tissue volume decrease depends on radiation dose delivered to the healthy hemisphere and differs between treatment modalities. In contrast, the decrease in perfusion was comparable for both irradiation modalities. We conclude that proton therapy may reduce brain-volume loss when compared to photon therapy.

PORTAF - postoperative radiotherapy of non-small cell lung cancer: accelerated versus conventional fractionation - study protocol for a randomized controlled trial

BACKGROUND

In early-stage non-small cell lung cancer (NSCLC) without affected lymph nodes detected at staging, surgical resection is still the mainstay of treatment. However, in patients with metastatic mediastinal lymph nodes (pN2) or non-radically resected primary tumors (R1/R2), postoperative radiotherapy (possibly combined with chemotherapy) is indicated. So far, investigations about time factors affecting postoperative radiotherapy have only examined the waiting time defined as interval between surgery and start of radiotherapy, but not the overall treatment time (OTT) itself. Conversely, results from trials on primary radio(chemo)therapy in NSCLC show that longer OTT correlates with significantly worse local tumor control and overall survival rates. This time factor of primary radio(chemo)therapy is thought to mainly be based on repopulation of surviving tumor cells between irradiation fractions. It remains to be elucidated if such an effect also occurs when patients with NSCLC are treated with postoperative radiotherapy after surgery (and chemotherapy). Our own retrospective data suggest an advantage of shorter OTT also for postoperative radiotherapy in this patient group.

METHODS/DESIGN

This is a multicenter, prospective randomized trial investigating whether an accelerated course of postoperative radiotherapy with photons or protons (7 fractions per week, 2 Gy fractions) improves locoregional tumor control in NSCLC patients in comparison to conventional fractionation (5 fractions per week, 2 Gy fractions). Target volumes and total radiation doses will be stratified in both treatment arms based on individual risk factors.

DISCUSSION

For the primary endpoint of the study we postulate an increase in local tumor control from 70% to 85% after 36 months. Secondary endpoints are overall survival of patients; local recurrence-free and distant metastases-free survival after 36 months; acute and late toxicity and quality of life for both treatment methods.

TRIAL REGISTRATION

ClinicalTrials.gov, NCT02189967 . Registered on 22 May 2014.

Impact of robust treatment planning on single- and multi-field optimized plans for proton beam therapy of unilateral head and neck target volumes

BACKGROUND

Proton beam therapy is promising for the treatment of head and neck cancer (HNC), but it is sensitive to uncertainties in patient positioning and particle range. Studies have shown that the planning target volume (PTV) concept may not be sufficient to ensure robustness of the target coverage. A few planning studies have considered irradiation of unilateral HNC targets with protons, but they have only taken into account the dose on the nominal plan, without considering anatomy changes occurring during the treatment course.

METHODS

Four pencil beam scanning (PBS) proton therapy plans were calculated for 8 HNC patients with unilateral target volumes: single-field (SFO) and multi-field optimized (MFO) plans, either using the PTV concept or clinical target volume (CTV)-based robust optimization. The dose was recalculated on computed tomography (CT) scans acquired during the treatment course. Doses to target volumes and organs at risk (OARs) were compared for the nominal plans, cumulative doses considering anatomical changes, and additional setup and range errors in each fraction. If required, the treatment plan was adapted, and the dose was compared with the non-adapted plan.

RESULTS

All nominal plans fulfilled the clinical specifications for target coverage, but significantly higher doses on the ipsilateral parotid gland were found for both SFO approaches. MFO PTV-based plans had the lowest robustness against range and setup errors. During the treatment course, the influence of the anatomical variation on the dose has shown to be patient specific, mostly independent of the chosen planning approach. Nine plans in four patients required adaptation, which led to a significant improvement of the target coverage and a slight reduction in the OAR dose in comparison to the cumulative dose without adaptation.

CONCLUSIONS

The use of robust MFO optimization is recommended for ensuring plan robustness and reduced doses in the ipsilateral parotid gland. Anatomical changes occurring during the treatment course might degrade the target coverage and increase the dose in the OARs, independent of the chosen planning approach. For some patients, a plan adaptation may be required.

The posterior cerebellum, a new organ at risk?

Eekers et al. have recently proposed a neuro-oncology atlas, which was co-authored by most centers associated in the European Proton Therapy Network (EPTN; Figure 1). With the introduction of new treatment techniques, such as integrated magnetic resonance imaging and linear accelerators (MR-linac) or particle therapy, the prediction of clinical efficacy of these more costly treatment modalities becomes more relevant. One of the side-effects of brain irradiation, being cognitive decline, is one of the toxicities most difficult to measure and predict. In order to validly compare different treatment modalities, 1) a uniform nomenclature of the organs at risk (OARs), 2) uniform atlas-based delineation [e.g., Eekers et al.], 3) long-term follow-up data with standardized cognitive tests, 4) a large patient population, and 5) (thus derived) validated normal tissue complication probability (NTCP) models are mandatory. Apart from the Gondi model, in which the role of the dose to 40% of both hippocampi (HC) proves to be significantly related to cognition in 18 patients, no similar models are available. So there is a strong need for more NTCP models, on HC, brain tissue and possible other relevant brain structures. In this review we summarize the available evidence on the role of the posterior cerebellum as a possible new organ at risk for cognition, which is deemed relevant for irradiation of brain and head and neck tumors.

Sites of recurrent disease and prognostic factors in SCLC patients treated with radiochemotherapy

•

Most recurrences occurred within the primary tumor or initially affected lymph nodes.

•

All sites of loco-regional recurrence had received 92–106% of the prescribed dose.

•

No isolated nodal failure occurred, supporting the use of selective nodal irradiation.

Objectives

Concurrent radiochemotherapy (RCHT) is standard treatment in locally advanced small cell lung cancer (SCLC) patients. Due to conflicting results on elective nodal irradiation (ENI) or selective node irradiation (SNI) there is no clear evidence on optimal target volumes. Therefore, the purposes of this study were to assess the sites of recurrent disease in SCLC and to evaluate the feasibility of SNI versus ENI.

Methods

A retrospective single-institution study of 43 consecutive patients treated with RCHT was performed. After state-of-the-art staging including FDG-PET/CT, all patients underwent three-dimensional conformal radiotherapy to a total dose of 45 Gy in twice-daily fractions of 1.5 Gy starting concurrently with the first or second chemotherapy cycle. All sites of loco-regional recurrences were correlated to the initial tumor and dose delivered. The impact of potential prognostic variables on outcome was evaluated using the Cox-regression model.

Results

13 patients (30%) relapsed locally or regionally: six within the initial primary tumor volume, five within the initially affected lymph nodes, one metachronously within primary tumor and initially affected lymph nodes, and one both inside and outside of the initial nodal disease. All sites of loco-regional recurrence had received 92–106% of the prescribed dose.

Conclusion

In our study most recurrences occurred within the primary tumor or initially affected lymph nodes, or distantly. We did not register any case of isolated nodal failure, supporting the use of selective nodal irradiation, possibly with the addition of supraclavicular irradiation in patients with nodal disease in the upper mediastinum.

Residual tumour hypoxia in head-and-neck cancer patients undergoing primary radiochemotherapy, final results of a prospective trial on repeat FMISO-PET imaging

BACKGROUND

Hypoxia is a well recognised parameter of tumour resistance to radiotherapy, a number of anticancer drugs and potentially immunotherapy. In a previously published exploration cohort of 25 head and neck squamous cell carcinoma (HNSCC) patients on [¹⁸F]fluoromisonidazole positron emission tomography (FMISO-PET) we identified residual tumour hypoxia during radiochemotherapy, not before start of treatment, as the driving mechanism of hypoxia-mediated therapy resistance. Several quantitative FMISO-PET parameters were identified as potential prognostic biomarkers. Here we present the results of the prospective validation cohort, and the overall results of the study.

METHODS

FMISO-PET/CT images of further 25 HNSCC patients were acquired at four time-points before and during radiochemotherapy (RCHT). Peak standardised uptake value, tumour-to-background ratio, and hypoxic volume were analysed. The impact of the potential prognostic parameters on loco-regional tumour control (LRC) was validated by the concordance index (ci) using univariable and multivariable Cox models based on the exploration cohort. Log-rank tests were employed to compare the endpoint between risk groups.

RESULTS

The two cohorts differed significantly in several baseline parameters, e.g., tumour volume, hypoxic volume, HPV status, and intercurrent death. Validation was successful for several FMISO-PET parameters and showed the highest performance (ci=0.77-0.81) after weeks 1 and 2 of treatment. Cut-off values for the FMISO-PET parameters could be validated after week 2 of RCHT. Median values for the residual hypoxic volume, defined as the ratio of the hypoxic volume in week 2 of RCHT and at baseline, stratified patients into groups of significantly different LRC when applied to the respective other cohort.

CONCLUSION

Our study validates that residual tumour hypoxia during radiochemotherapy is a major driver of therapy resistance of HNSCC, and that hypoxia after the second week of treatment measured by FMISO-PET may serve as biomarker for selection of patients at high risk of loco-regional recurrence after state-of-the art radiochemotherapy.

The clinical target volume in lung, head-and-neck, and esophageal cancer: Lessons from pathological measurement and recurrence analysis

Radiotherapy research has achieved remarkable progress in target volume definition. Advances in medical imaging facilitate more precise localization of the gross tumor volume, alongside a more detailed understanding of the geometric uncertainties associated with treatment delivery that has enabled robust safety margins to be customized to the specific treatment scenario at hand. By contrast, the clinical target volume, meant to encompass gross tumor, as well as, adjacent sub-clinical disease, has evolved very little. It is more often defined by clinician experience and institutional convention than on a patient-specific basis. This disparity arises from the inherent invisibility of sub-clinical disease in current medical imaging. Its incidence and expanse can only be ascertained via indirect means. This article reviews two such strategies: histopathological measurements on resection specimen and analyses of locoregional recurrences after radiotherapy.

Individualized early death and long-term survival prediction after stereotactic radiosurgery for brain metastases of non-small cell lung cancer: Two externally validated nomograms

INTRODUCTION

Commonly used clinical models for survival prediction after stereotactic radiosurgery (SRS) for brain metastases (BMs) are limited by the lack of individual risk scores and disproportionate prognostic groups. In this study, two nomograms were developed to overcome these limitations.

METHODS

495 patients with BMs of NSCLC treated with SRS for a limited number of BMs in four Dutch radiation oncology centers were identified and divided in a training cohort (n=214, patients treated in one hospital) and an external validation cohort n=281, patients treated in three other hospitals). Using the training cohort, nomograms were developed for prediction of early death (<3months) and long-term survival (>12months) with prognostic factors for survival. Accuracy of prediction was defined as the area under the curve (AUC) by receiver operating characteristics analysis for prediction of early death and long term survival. The accuracy of the nomograms was also tested in the external validation cohort.

RESULTS

Prognostic factors for survival were: WHO performance status, presence of extracranial metastases, age, GTV largest BM, and gender. Number of brain metastases and primary tumor control were not prognostic factors for survival. In the external validation cohort, the nomogram predicted early death statistically significantly better (p<0.05) than the unfavorable groups of the RPA, DS-GPA, GGS, SIR, and Rades 2015 (AUC=0.70 versus range AUCs=0.51-0.60 respectively). With an AUC of 0.67, the other nomogram predicted 1year survival statistically significantly better (p<0.05) than the favorable groups of four models (range AUCs=0.57-0.61), except for the SIR (AUC=0.64, p=0.34). The models are available on www.predictcancer.org.

CONCLUSION

The nomograms predicted early death and long-term survival more accurately than commonly used prognostic scores after SRS for a limited number of BMs of NSCLC. Moreover these nomograms enable individualized probability assessment and are easy into use in routine clinical practice.

[18F]FDG PET/CT-based response assessment of stage IV non-small cell lung cancer treated with paclitaxel-carboplatin-bevacizumab with or without nitroglycerin patches

PURPOSE

Nitroglycerin (NTG) is a vasodilating drug, which increases tumor blood flow and consequently decreases hypoxia. Therefore, changes in [18F] fluorodeoxyglucose positron emission tomography ([18F]FDG PET) uptake pattern may occur. In this analysis, we investigated the feasibility of [18F]FDG PET for response assessment to paclitaxel-carboplatin-bevacizumab (PCB) treatment with and without NTG patches. And we compared the [18F]FDG PET response assessment to RECIST response assessment and survival.

METHODS

A total of 223 stage IV non-small cell lung cancer (NSCLC) patients were included in a phase II study (NCT01171170) randomizing between PCB treatment with or without NTG patches. For 60 participating patients, a baseline and a second [18F]FDG PET/computed tomography (CT) scan, performed between day 22 and 24 after the start of treatment, were available. Tumor response was defined as a 30 % decrease in CT and PET parameters, and was compared to RECIST response at week 6. The predictive value of these assessments for progression free survival (PFS) and overall survival (OS) was assessed with and without NTG.

RESULTS

A 30 % decrease in SUVpeak assessment identified more patients as responders compared to a 30 % decrease in CT diameter assessment (73 % vs. 18 %), however, this was not correlated to OS (SUVpeak30 p = 0.833; CTdiameter30 p = 0.557). Changes in PET parameters between the baseline and the second scan were not significantly different for the NTG group compared to the control group (p value range 0.159-0.634). The CT-based (part of the [18F]FDG PET/CT) parameters showed a significant difference between the baseline and the second scan for the NTG group compared to the control group (CT diameter decrease of 7 ± 23 % vs. 19 ± 14 %, p = 0.016, respectively).

CONCLUSIONS

The decrease in tumoral FDG uptake in advanced NSCLC patients treated with chemotherapy with and without NTG did not differ between both treatment arms. Early PET-based response assessment showed more tumor responders than CT-based response assessment (part of the [18F]FDG PET/CT); this was not correlated to survival. This might be due to timing of the [18F]FDG PET shortly after the bevacizumab infusion.