Is oligometastatic disease an applicable and useful concept in haematologic malignancies? A narrative review of radiation therapy standards, modern techniques, and innovations

PURPOSE

Haematologic malignancies are particular in that they can generally be cured, even when distant metastases are present at diagnosis, unlike solid malignancies. Systemic treatments, including chemotherapy, targeted therapies, and immunotherapy, are the standard of care with excellent results. The considerable progress made in the management of these diseases in the last 20years has redefined the role of radiation therapy as minor in many clinical situations. We propose a literature review of data, showing that radiation therapy still has a role in curative, salvage, and palliative therapy situations.

MATERIAL AND METHODS

A document and literature search was carried out in the following databases: Medline and ClinicalTrial.gov, for the terms "radiotherapy", "haematologic malignancies", "Hodgkin lymphoma", "non-Hodgkin lymphoma", "CAR T cells", "multiple myeloma", "solitary plasmocytoma", "intensity-modulated radiotherapy", "extracranial stereotactic body radiation therapy" and "proton therapy references".

RESULTS

Haemopathological malignancies include a wide range of diseases and radiation therapy indications have been assessed over the past 20years. Currently, radiation therapy is indicated for localized disease (solitary plasmocytoma), as an adjuvant (Hodgkin lymphoma), in palliative settings, or after systemic treatment in relapsed patients (chimeric antigen receptor [CAR] T-cells) with a low recurrence burden, which can therefore be considered "oligorecurrence". Radiation therapy, through total body irradiation, has important indications, thanks to its immunomodulatory and/or myeloablative effects. Moreover, recent technological developments have made possible significant improvement in safety, contributing to radiation therapy being positioned in the treatment strategy of several indications.

CONCLUSIONS

Given the effectiveness of systemic treatments in hematologic malignancies, the oligometastasis stage is of little importance. A curative intent after local radiation therapy, even advanced stage, is possible, both with residual disease for advanced Hodgkin lymphoma, aggressive non-Hodgkin lymphoma, or solitary plasmocytoma, and even without evidence of disease after chemotherapy for Hodgkin or non-Hodgkin lymphoma. The role of new treatments, such as CAR T cells, allows us to consider radiation therapy after systemic treatment of relapsed diseases with low volume recurrence, which can be considered oligorecurrence.

Image-guided radiotherapy

We present the updated recommendations of the French society for oncological radiotherapy on image-guided radiotherapy (IGRT). The objective of the IGRT is to take into account the anatomical variations of the target volume occurring between or during the irradiation fractions, such as displacements and/or deformations, so that the delivered dose corresponds to the planned dose. This article presents the different IGRT devices, their use and quality control, and quantify the possible additional dose generated by each of them. The practical implementation of IGRT in various tumour locations is summarised, from the different "RecoRad™" guideline articles. Adaptive radiotherapy is then detailed, due to its complexity and its probable development in the next years. The place of radiation technologist in the practice of IGRT is then specified. Finally, a brief update is proposed on the delicate question of the additional dose linked to the in-room imaging, which must be estimated and documented at a minimum, as long as it is difficult to integrate it into the calculation of the dose distribution.

P04.15 Radiomics features in prediction of pseudo-progression and response to hypofractionated stereotactic radiotherapy and anti-PDL1 Durvalumab combination in recurrent glioblastoma from STERIMGLI phase I trial

BACKGROUND

Due to the risk of pseudo-progression, evaluating the response of such combined treatment with anti-PDL1 immunotherapy and hypofractionated stereotactic radiotherapy (hFSRT) in patients with recurrent glioblastoma (GBM) is difficult even with iRANO criteria. We aim to analyze multi-modal MRI radiomic features in recurrent GBM patients included in the phase I STERIMGLI (NCT02866747) clinical trial.

MATERIAL AND METHODS

In phase I trial STERIMGLI, six patients received hFSRT 24 Gy in 3 fractions of 8 Gy prescribed to the 80% covering isodose in combination with Durvalumab IV 1500mg the same day following the third fraction of radiation then every 4 weeks for a maximum of 12 months. All patients underwent multi-modal MRI acquisitions. Contrast-enhancement on post-gadolinium T1 MRI (Gd-T1) and hyper-T2 signal on FLAIR MRI were manually segmented as volume-of-interest (VOIs) from baseline at week 0 (W0) and every eight weeks until weeks sixteen (W16). For patient 05, segmentation was performed until progression (W40). First order radiomics were extracted and normalized from these longitudinal VOIs in a total of 42 MRI: volume (in cm3), mean and median intensity, entropy, skewness and kurtosis were computed for each VOI. In total 504 (2 x 42 x 6) features were extracted.

RESULTS

Two patients underwent pseudoprogression (patients 04 and 05).

All patients but one (patient 05) presented an early relapse with an increase of FLAIR and Gd-T1 volumes and respective entropies from the first evaluation without decrease during follow up (until W16).

Patient 05 presented a Dose Limiting Toxicities and had the longest time to progression after Durvalumab and hFSRT. Contrary to the other patients, his FLAIR radiomics features presented a decrease of volume, intensity and entropy between W0 and W16 then, interestingly an increase between W24 and W40, date of progression. Moreover, patient 05 presented the lowest tumor volume and the lowest Gd-T1entropy. The intensity of the Gd-T1 followed the same trend as the hyper intense signal on the FLAIR in all patients, in particular for patient 05.

CONCLUSION

In our study, entropy and tumor volume for both FLAIR and Gd-T1, and FLAIR intensity seems to be the most interesting parameters to access the response of combined treatment. FLAIR signal may be more specific to the tumor microenvironment.

First order radiomics allowed us to follow tumor heterogeneity and identify the patient with the longest time to progression.

Performance comparison of modified ComBat for harmonization of radiomic features for multicenter studies

Multicenter studies are needed to demonstrate the clinical potential value of radiomics as a prognostic tool. However, variability in scanner models, acquisition protocols and reconstruction settings are unavoidable and radiomic features are notoriously sensitive to these factors, which hinders pooling them in a statistical analysis. A statistical harmonization method called ComBat was developed to deal with the "batch effect" in gene expression microarray data and was used in radiomics studies to deal with the "center-effect". Our goal was to evaluate modifications in ComBat allowing for more flexibility in choosing a reference and improving robustness of the estimation. Two modified ComBat versions were evaluated: M-ComBat allows to transform all features distributions to a chosen reference, instead of the overall mean, providing more flexibility. B-ComBat adds bootstrap and Monte Carlo for improved robustness in the estimation. BM-ComBat combines both modifications. The four versions were compared regarding their ability to harmonize features in a multicenter context in two different clinical datasets. The first contains 119 locally advanced cervical cancer patients from 3 centers, with magnetic resonance imaging and positron emission tomography imaging. In that case ComBat was applied with 3 labels corresponding to each center. The second one contains 98 locally advanced laryngeal cancer patients from 5 centers with contrast-enhanced computed tomography. In that specific case, because imaging settings were highly heterogeneous even within each of the five centers, unsupervised clustering was used to determine two labels for applying ComBat. The impact of each harmonization was evaluated through three different machine learning pipelines for the modelling step in predicting the clinical outcomes, across two performance metrics (balanced accuracy and Matthews correlation coefficient). Before harmonization, almost all radiomic features had significantly different distributions between labels. These differences were successfully removed with all ComBat versions. The predictive ability of the radiomic models was always improved with harmonization and the improved ComBat provided the best results. This was observed consistently in both datasets, through all machine learning pipelines and performance metrics. The proposed modifications allow for more flexibility and robustness in the estimation. They also slightly but consistently improve the predictive power of resulting radiomic models.

P14.24 Bevacizumab treatment in atypical disseminated choroid plexus papilloma in adult patients

BACKGROUND

Choroid plexus tumours represent less than 1% of brain tumours. Low-grade papilloma may be treated with gross total surgical resection, while in disseminated progressive atypical choroid plexus papilloma (APP), there is no standard treatment: various chemotherapy regimens have been reported. Since this tumour is characterized by a rich vascular component, antiangiogenic therapy is an attractive treatment. The use of Bevacizumab has already been reported in three patients. Authors expand this experience with 5 further patients diagnosed with progressive APP treated with bevacizumab.

MATERIAL AND METHODS

Patients were recruited through the weekly Adolescent Young Adult French web conference. They have been treated with bevacizumab 10 mg/kg by intravenous injection each 2 to 3 weeks. Their clinical status and radiological response are reported: Karnofsky Index (KI), Pain Scale (PS) and RANO criteria were used.

RESULTS

All our patients had a progressive disease prior to bevacizumab. Pt 1: 41 years old; APP with cranio-spinal dissemination; Previous treatments: local and cranio-spinal irradiation in 2012 and in 2014; surgery: complete and partial resection in 2010, 2014, and VP shunt in 2018; Bevacizumab: total of 33 cures in 18 months. Result: radiological and clinical stabilization. Pt 2: 58 years old; APP with cranio-spinal dissemination; Previous treatments: surgery: VP shunt and gross total resection, in 2006; VP shunt, in 2011. Bevacizumab: total of 4 cures, in 2 months. Result: radiological and clinical stabilization. Pt 3: 34 years old; APP with cranio-spinal dissemination; Previous treatments: surgery: gross total resection, in 1992, and shunt in 2008. Bevacizumab: total of 21 cures, in 21 months. Result: radiological and clinical stabilization. Pt 4: 63 years old; APP with cranio-spinal dissemination; Previous treatments: surgery: surgical resection and VP shunt in 2013; chemotherapy: temozolomide. Bevacizumab: 29 months (still treated). Result: radiological and clinical stabilization. Pt 5: 62 years old; APP with cranio-spinal dissemination; Previous treatments: surgery: gross total resection in 1999 and VP shunt in 2009; chemotherapy: Carboplatin Vespesid. Bevacizumab: 23 cures, in 14 months. Result: radiological stabilization and clinical amelioration.

CONCLUSION

Despite their previous worsening disease, all patients obtained a stabilization or amelioration of their IK and PS under bevacizumab. Bevacizumab should be evaluated in a multicentric trial as standard therapy for disseminated metastasized progressive choroid plexus tumours.

OS4.4 Hypofractionated stereotactic radiotherapy and anti-PDL1 Durvalumab combination in recurrent glioblastoma: Results of the phase I part of the phase I/II STERIMGLI trial

BACKGROUND

Glioblastoma (GBM) remains a lethal disease with inevitable local relapse and no standard treatment. Re-irradiation by hypofractionated stereotactic radiotherapy (hFSRT) is an option of treatment with tolerable safety, but needs improvement in term of efficacy. Radiotherapy (RT) causes immunogenic tumor cell death but also induces PDL1 and PD1 expression on tumors and immune cells, potentially evoking resistance to RT. Pre-clinical studies combining hFSRT with an anti-PD-1 antibody in GBM have shown increased efficacy of the combination. Clinical studies also show encouraging results when checkpoint inhibitors have been combined with high dose RT. We hypothesized that combining the anti PD-L1 Durvalumab (Durva) with hFSRT will be an effective regimen for patients with recurrent GBM. We designed a phase I/II clinical trial studying the combination of hFSRT with Durva for recurrent GBM≤35 mm diameter. Results of the phase I are presented.

MATERIAL AND METHODS

Patients were included from February 2017 to October 2017.

A standard 3 + 3 de-escalation design was used. Patients were treated by hFSRT 24 Gy, 8 Gy/fraction at 80% isodose, every other day, combined with Durva infusion 1500mg first dose (Level 1) or 750 mg (Level -1) delivered on the last hFSRT day followed by 1500 mg Durva infusion every four weeks until relapse. The schema was defined as safe if one patient or less among 6 presents a dose limiting toxicity (DLT). Brain MRI was performed before RT and then every 8 weeks until relapse. Tumor assessment was performed according to RANO criteria.

RESULTS

Among the 6 patients (3 methylated MGMT, 3 unmethylated MGMT; all wild type IDH) included at the level 1, all completed the hFSRT course, only one had a DLT which was an immune related grade 3 vestibular neuritis. At the time of analysis (24/01/19), all the patients had a local tumor progression, 4 were still alive. Local progression free interval (LPFI) ranges from 2.1 to 8.1 months. Interestingly the 2 patients who presented a pseudo-progression had a prolonged LPFI (5.7 and 8.1 months) compared to the other patients. All the patients except these 2 patients had a lymphopenia at inclusion. PDL-1 expression varied from 0 to 70% in the primary tumor.

CONCLUSION

Combining three 8 Gy fractions of hFSRT with 1500 mg Durvalumab on the 3rd fraction hFSRT and every 4 weeks for recurrent GBM is well tolerated justifying exploration of its efficacy in the phase II which is currently in interim analysis

[Radiotherapy for retroperitoneal sarcomas]

The management of retroperitoneal sarcoma can be very challenging, and the quality of initial treatment strategy appears to be a crucial prognostic factor. En bloc surgery is currently the standard of care for these rare tumours and perioperative treatments such as chemotherapy or radiotherapy have not been validated yet. However, local-regional relapse constitutes the most common disease course. While adjuvant radiotherapy is less and less common due to gastrointestinal toxicities, preoperative radiation therapy offers numerous advantages and is being evaluated as part of a national multicentre phase II study (TOMOREP trial) and is the subject of a European randomized phase III study (STRASS trial). The objective of this article is to present data on preoperative irradiation in terms of dose, volumes and optimal radiotherapy techniques for the treatment of this rare disease.