Hypofractionated stereotactic radiotherapy in Spetzler Martin grades 4 and 5 arteriovenous malformations in the pediatric population: Is it a viable alternative?

OBJECTIVE

To analyze the outcomes of hypofractionated stereotactic radiotherapy (HFSRT) for Spetzler Martin grades 4 and 5 arteriovenous malformations (AVMs) in a pediatric population.

METHODS

Fourteen patients with Spetzler Martin (SM) grades IV and V large AVMs who underwent HFSRT between January 2013 and July 2019 were retrospectively reviewed. All patients received HFSRT at a dose of 30-36 Gy in 5 to 6 fractions. They were followed up annually with clinical and imaging assessments to evaluate obliteration rates.

RESULTS

The median age at presentation was 15 years (range 8-21 years). Ten (71%) were SM grade 4 AVMs and the rest were SM grade 5 AVMs. The majority presented with headache (8 [57%]), and 3 (21%) presented with bleeding. The median nidus volume was 39.4 cc (IQR, 31.4-52.4). Two (14%) patients had infratentorial AVMs. All of them had deep venous drainage. The median clinical follow-up duration was 75 months (range 31-107 months). There was complete obliteration of the nidus in 3 (21%) patients with a median time to obliteration of 39 months. HFSRT resulted in a reduction of the AVM volume to 12 cc or less in nearly 70% of patients. None of the patients experienced re-bleeding. 79% reported an improvement in their symptoms.

CONCLUSION

HFSRT is a highly effective treatment for high-grade AVMs in children, which can result in either complete elimination or significant reduction of the nidus volume or make it suitable for additional treatment, such as single-session stereotactic radiosurgery (SRS).

Correlation Between Post-Radiosurgery Perinidal Hyperintensity and AVM Obliteration Following LINAC-Based Stereotactic Radiosurgery

OBJECTIVE

We studied the correlation between new-onset perinidal hyperintensity (PH) on T2-weighted magnetic resonance imaging and obliteration of intracranial arteriovenous malformation (AVM) after stereotactic radiosurgery (SRS).

METHODS

A retrospective study of 148 patients with an intracranial AVM who underwent SRS between September 2005 and June 2018 and had ≥1 radiological follow-up (early magnetic resonance imaging) 12-18 months after SRS was performed to analyze the correlation between PH (graded from 0 to 2) and AVM obliteration.

RESULTS

Of the 148 patients, 95 were male. The mean patient age was 27.7 ± 12.4 years. Of the 148 AVMs, 105 (70.9%) were obliterated at a median follow-up of 27 months (interquartile range, 14-48 months). The cumulative 3-, 5-, 10-year obliteration rate was 51.8%, 70.8%, and 91.8%, respectively. New-onset PH was observed in 58 AVMs (39.2%; 50 obliterated and 8 not obliterated). No association was found between the pretreatment variables or dose delivered and the development of PH. Grade 2 PH was associated with the risk of symptoms developing compared with grade 1 PH (37.5% vs. 4%; P = 0.002). Symptomatic PH was more likely to develop in patients with a larger AVM (P = 0.05). On multivariate analysis, the presence of a single draining vein (odds ratio [OR], 2.9; 95% confidence interval [CI], 1.3-6.8), a lower median AVM volume (OR, 0.97; 95% CI, 0.6-0.89), a mean marginal radiation dose (OR, 1.29; 95% CI, 1.02-1.64), and the presence of PH (OR, 3.16; 95% CI, 1.29-7.71) were independent predictors of AVM obliteration.

CONCLUSIONS

The incidence of PH after SRS for AVM was 39.2%. PH was an independent predictor of AVM obliteration after SRS. Grade 2 PH and a larger AVM volume were associated with symptomatic PH.

Feasibility study of total marrow lymphoid irradiation with volumetric modulated arc therapy: clinical implementation in a tertiary care center

PURPOSE

Total marrow lymphoid irradiation (TMLI) with volumetric modulated arc therapy (VMAT) is challenging due to large treatment fields with multiple isocenters, field matching at junctions, and targets being surrounded by many organs at risk. This study aimed to describe our methodology for safe dose escalation and accurate dose delivery of TMLI treatment with the VMAT technique based on early experience at our center.

MATERIALS AND METHODS

Computed tomography (CT) scans were acquired in head-first supine and feet-first supine orientations for each patient with an overlap at mid-thigh. VMAT plans were generated for 20 patients on the head-first CT images with either three or four isocenters in the Eclipse treatment planning system (Varian Medical Systems Inc., Palo Alto, CA) and the treatment was delivered in a Clinac 2100 C/D linear accelerator (Varian Medical Systems Inc., Palo Alto, CA).

RESULTS

Five patients were treated with a prescription dose of 13.5 Gy in 9 fractions and 15 patients were treated with an escalated dose of 15 Gy in 10 fractions. The mean doses to 95% of the clinical target volume (CTV) and planning target volume (PTV) were 14.3 ± 0.3 Gy and 13.6 ± 0.7 Gy for the prescription doses of 15 Gy, and 13 ± 0.2 Gy and 12.3 ± 0.3 Gy for the prescription doses of 13.5 Gy, respectively. Mean dose to the lung in both schedules was 8.7 ± 0.6 Gy. The overall time taken to execute the treatment plans was approximately 2 h for the first fraction and 1.5 h for subsequent fractions. The average in-room time of 15.5 h per patient over 5 days leads to potential changes in the regular treatment schedules for other patients.

CONCLUSION

This feasibility study highlights the methodology adopted for safe implementation of TMLI with the VMAT technique at our institution. Escalation of dose to the target with adequate coverage and sparing of critical structures was achieved with the adopted treatment technique. Clinical implementation of this methodology at our center could serve as a practical guide to start the VMAT-based TMLI program safely by others who are keen to start this service.

Total marrow lymphoid irradiation and cyclophosphamide is associated with low toxicity and good outcomes in patients undergoing hematopoietic stem cell transplantation for acute lymphoblastic leukemia and chronic myeloid leukemia in lymphoid blast crises - A phase I study

INTRODUCTION

Total marrow lymphoid irradiation (TMLI) can deliver higher doses of irradiation without increasing toxicity compared to Total body irradiation (TBI).

METHODS

Twenty adult patients undergoing hematopoietic stem cell transplantation (HSCT) for acute lymphoblastic leukemia (ALL) and chronic myeloid leukemia with lymphoid blast crises (CML-LBC) received TMLI and cyclophosphamide for conditioning. Ten patients each received 13.5 or 15 Gy of TMLI. The graft source was peripheral blood stem cells in all, and donors included matched related (n = 15), haplo-identical (n = 3) or matched unrelated donors (n = 2).

RESULTS

The median cell dose infused was 9 × 106 CD34/kg (range 4.8-12.4). Engraftment occurred in all (100%) at a median of 15 days (range: 14-17). Toxicity was low with hemorrhagic cystitis seen in two but no sinusoidal obstruction syndrome. Acute GVHD occurred in 40% while chronic GVHD was seen in 70.5%. Viral infections were seen in 55% while blood stream bacterial infections occurred in 20% and invasive fungal disease (IFD) in 10%. The Day 100 non-relapse mortality (NRM) was 10%. At a median follow up of 25 months (range 2-48), two patients have relapsed. Overall survival at 2 years is 80% while the disease-free survival is 75%.

CONCLUSIONS

The combination of TMLI and cyclophosphamide for myeloablative conditioning is associated with low toxicity and favorable early outcomes in patients undergoing HSCT for ALL and CML-LBC.

Multi-centre radiomics for prediction of recurrence following radical radiotherapy for head and neck cancers: Consequences of feature selection, machine learning classifiers and batch-effect harmonization

Background and purpose

Radiomics models trained with limited single institution data are often not reproducible and generalisable. We developed radiomics models that predict loco-regional recurrence within two years of radiotherapy with private and public datasets and their combinations, to simulate small and multi-institutional studies and study the responsiveness of the models to feature selection, machine learning algorithms, centre-effect harmonization and increased dataset sizes.

Materials and methods

562 patients histologically confirmed and treated for locally advanced head-and-neck cancer (LA-HNC) from two public and two private datasets; one private dataset exclusively reserved for validation. Clinical contours of primary tumours were not recontoured and were used for Pyradiomics based feature extraction. ComBat harmonization was applied, and LASSO-Logistic Regression (LR) and Support Vector Machine (SVM) models were built. 95% confidence interval (CI) of 1000 bootstrapped area-under-the-Receiver-operating-curves (AUC) provided predictive performance. Responsiveness of the models' performance to the choice of feature selection methods, ComBat harmonization, machine learning classifier, single and pooled data was evaluated.

Results

LASSO and SelectKBest selected 14 and 16 features, respectively; three were overlapping. Without ComBat, the LR and SVM models for three institutional data showed AUCs (CI) of 0.513 (0.481-0.559) and 0.632 (0.586-0.665), respectively. Performances following ComBat revealed AUCs of 0.559 (0.536-0.590) and 0.662 (0.606-0.690), respectively. Compared to single cohort AUCs (0.562-0.629), SVM models from pooled data performed significantly better at AUC = 0.680.

Conclusions

Multi-institutional retrospective data accentuates the existing variabilities that affect radiomics. Carefully designed prospective, multi-institutional studies and data sharing are necessary for clinically relevant head-and-neck cancer prognostication models.

Influence of Jaw Setting in the Determination of Stereotactic Small-Field Output Factors with Different Detectors

Background

The experimental determination of relative output factors presents the greatest challenge, especially for small fields with different detectors. The aim of this study is to evaluate the influence of jaw positions on small-field output factors for the fields defined by micro-multileaf collimator and circular cones with different detectors.

Materials and Methods

The stereotactic output factors were measured on Primus linear accelerator with BrainLab micro-multileaf collimator (mMLC) and circular cones as add-on tertiary collimators. Square field sizes ranging from 0.6 cm × 0.6 cm to 9.8 cm × 9.8 cm and circular fields of diameter ranging from 1.0 cm to 4.0 cm were defined by mMLC and circular cones, respectively. The influence of jaw position on output factor was assessed for different geometric configurations with three different detectors.

Results

The values obtained with PinPoint ion chamber were consistent with microDiamond detector for fields greater than 24 mm × 24 mm, but an underestimation of 23.9% was noticed in 6 mm x 6 mm field size. For the mMLC defined field size of 6 mm × 6 mm, when the X-Y jaw was moved from 8 mm × 8 mm to 80 mm × 80 mm, an increase in the output by a factor of 1.7 was observed with both microDiamond and stereotactic radiosurgery diode, whereas an increase in output by a factor of 1.9 was noticed with PinPoint ion chamber.

Conclusion

Output factors obtained with different detectors show high differences in the smallest field size for all collimating systems. This study confirms that the position of X and Y jaw above the tertiary collimator significantly influences the small-field output factor.

Output factor measurements with multiple detectors in CyberKnife® Robotic Radiosurgery System

Aim

The aim of this study was to measure and compare the output factor (OF) of a CyberKnife Robotic Radiosurgery System with eight different small field detectors and validate with Technical Report Series (TRS) report 483.

Background

Accurate dosimetry of CyberKnife system is limited due to the challenges in small field dosimetry. OF is a vital dosimetric parameter used in the photon beam modeling and any error would affect the dose calculation accuracy.

Materials and Methods

In this study, the OF was measured with eight different small-field detectors for the 12 IRIS collimators at 800 mm SAD setup at 15 mm depth. The detectors used were PTW 31016 PinPoint 3D, IBA PFD shielded diode, IBA EFD unshielded diode, IBA SFD unshielded diode (stereotactic), PTW 60008 shielded diode, PTW 60012 unshielded diode, PTW 60018 unshielded diode (stereotactic), and PTW 60019 CVD diamond detector. OF was obtained after correcting for field output correction factors from IAEA TRS No. 483.

Results

The field OFs in CyberKnife are derived from the measured data by applying the correction factors from Table 23 in TRS 483 for the eight small field detectors. These field OFs matched within 2% of peer-reviewed published values. The range and standard deviation showed a decreasing trend with collimator diameter.

Conclusion

The field OF obtained after applying the appropriate correction factor from TRS 483 matched well with the peer-reviewed published OFs. The inter-detector variation showed a decreasing trend with increasing collimator field size. This study gives physicists confidence in measuring field OFs while using small field detectors mentioned in this work.

In-vivo dose measurements with MOSFET dosimeters during MV portal imaging

Background

The purpose of this study was to investigate the feasibility of MOSFET dosimeter in measuring eye dose during 2D MV portal imaging for setup verification in radiotherapy.

Materials and methods

The in-vivo dose measurements were performed by placing the dosimeters over the eyes of 30 brain patients during the acquisition of portal images in linear accelerator by delivering 1 MU with the field sizes of 10 × 10 cm² and 15 × 15 cm².

Results

The mean doses received by the left and right eyes of 10 out of 30 patients when both eyes were completely inside the anterior portal field were found to be 2.56 ± 0.2 cGy and 2.75 ± 0.2, respectively. Similarly, for next 10 patients out of the same 30 patients the mean doses to left and right eyes when both eyes were completely out of the anterior portal fields were found to be 0.13 ± 0.02 cGy and 0.17 ± 0.02 cGy, respectively. The mean doses to ipsilateral and contralateral eye for the last 10 patients when one eye was inside the anterior portal field were found to be 3.28 ± 0.2 cGy and 0.36 ± 0.1 cGy, respectively.

Conclusion

The promising results obtained during 2D MV portal imaging using MOSFET have shown that this dosimeter is well suitable for assessing low doses during imaging thereby enabling to optimize the imaging procedure using the dosimetric data obtained. In addition, the documentation of the dose received by the patient during imaging procedure is possible with the help of an in-built software in conjunction with the MOSFET reader module.

Response of Nanodot Optically Stimulated Luminescence Dosimeters to Therapeutic Electron Beams

Response of Al₂O₃:C-based nanoDot optically stimulated luminescence (OSL) dosimeter was studied for the dosimetry of 6, 9, 12, 16, and 20 MeV therapeutic electron beams. With reference to ionization chamber, no change in the response was observed with the change in the energy of electron beams for the field size from 6 cm × 6 cm to 25 cm × 25 cm, dose rates from 100 MU/min to 600 MU/min, and the linearity in the response up to 300 cGy. The fading of the transient signal was higher for 20 MeV electron beam than that of 6 MeV electron beam by about 5% as compared to value at 20 min after irradiation. The depletion of OSL signal per readout in 200 successive readouts was also found to change with dose and energy of electron beam from 6 MeV (9% and 12% per readout at 2 and 10 Gy, respectively) to 20 MeV (9% and 16% at 2 and 10 Gy, respectively). The OSL sensitivity changed in the range from 2% to 6% with accumulated doses from 2 to 8 Gy and with electron energy from 6 to 20 MeV, but the sensitivity could be reset using an optical annealing treatment. Although negligible fading for postirradiation storage from 20 min to several months, acceptable precision and linearity in the desired range, and high reproducibility makes nanoDot dosimeters very attractive for the dosimetry of therapeutic electron beams, a note should be made for changes in sensitivity at doses beyond 2 Gy and electron beams energy dependence in reuse, short-term fading, and signal depletion on repeated readout.

Dosimetric Performance of Newly Developed Farmer-Type Ionization Chamber in Radiotherapy Practice

Dose measurement with ionization chamber is essential to deliver accurate dose to the tumor in radiotherapy. The cylindrical Farmer-type ionization chamber is recommended by various dosimetry protocols for dose measurement of radiotherapy beams. The air-equivalent graphite wall Farmer-type ionization chamber (FAR 65 GB) of active volume 0.65 cm³ with aluminum as the central electrode material was fabricated. Various dosimetric parameters were studied for the newly developed ionization chamber in cobalt-60, 6 and 18 MV photon beams. The preirradiation and postirradiation leakage of the chamber was within 0.08%. The long-term stability and the stem effect of the chamber were within 0.07% and 0.3%, respectively. The sensitivity of the ionization chamber was found to be 22.15 nC/Gy. The chamber shows linear response with dose for cobalt-60, 6 and 18 MV photon beams. The ion recombination correction factor increases with increase in bias voltage. For all energies and field sizes, the polarity correction factor is almost closer to unity. The ion recombination and polarity correction measurements show that the polarizing potential and polarity recommended during the calibration of ionization chamber should be used for routine measurement to avoid the uncertainty. The chamber response is independent of dose rate and energy. The chamber is cost-effective and shows precise and reproducible response. The study carried out confirms that the newly fabricated ion chamber can be used in the measurement of absolute dose for high-energy photon beams.

Analysis of small field percent depth dose and profiles: Comparison of measurements with various detectors and effects of detector orientation with different jaw settings

The advent of modern technologies in radiotherapy poses an increased challenge in the determination of dosimetric parameters of small fields that exhibit a high degree of uncertainty. Percent depth dose and beam profiles were acquired using different detectors in two different orientations. The parameters such as relative surface dose (D_S), depth of dose maximum (D_max), percentage dose at 10 cm (D₁₀), penumbral width, flatness, and symmetry were evaluated with different detectors. The dosimetric data were acquired for fields defined by jaws alone, multileaf collimator (MLC) alone, and by MLC while the jaws were positioned at 0, 0.25, 0.5, and 1.0 cm away from MLC leaf-end using a Varian linear accelerator with 6 MV photon beam. The accuracy in the measurement of dosimetric parameters with various detectors for three different field definitions was evaluated. The relative D_S(38.1%) with photon field diode in parallel orientation was higher than electron field diode (EFD) (27.9%) values for 1 cm ×1 cm field. An overestimation of 5.7% and 8.6% in D₁₀ depth were observed for 1 cm ×1 cm field with RK ion chamber in parallel and perpendicular orientation, respectively, for the fields defined by MLC while jaw positioned at the edge of the field when compared to EFD values in parallel orientation. For this field definition, the in-plane penumbral widths obtained with ion chamber in parallel and perpendicular orientation were 3.9 mm, 5.6 mm for 1 cm ×1 cm field, respectively. Among all detectors used in the study, the unshielded diodes were found to be an appropriate choice of detector for the measurement of beam parameters in small fields.