Trial Information

The HIL-trial has been designed by the study initiators at the Clinical Cooperation Unit
Radiation Oncology at the German Cancer Research Center (DKFZ), the Clinical Cooperation
Unit Nuclear Medicine at DKFZ, the Department of Radiology at DKFZ and the Department of
Radiation Oncology at the University Clinic Heidelberg. The trial is carried out at DKFZ in
co-operation with the Department of Radiation Oncology at the University Clinic Heidelberg.

Aim of the study is to characterize the correlation of 18F-FMISO dPET-CT and functional MRI
for tumor hypoxia imaging in patients with stage III NSCLC, treated with intensity modulated
radiotherapy (IMRT) and evaluate changes in tumor oxygenation during radiation treatment.
Objectives include the evaluation of the prognostic value of multimodal hypoxia imaging for
locoregional control, progression free survival and overall survival of patients with NSCLC
treated with IMRT.

The study is designed as a single centre pilot trial with an accrual of 15 patients with
inoperable stage III NSCLC. Patients fulfilling the inclusion criteria are treated with
intensity modulated radiation therapy (IMRT). All patients undergo serial 18F-FMISO dPET-CT
and functional MRI before treatment, at week 5 of radiotherapy and 6 weeks post treatment.

Eligible patients are informed about participation in the trial with possible benefits and
risks, and written informed consent is obtained. Staging is completed through performance of
thoracic CT scan, abdominal ultrasound and 18F-FDG dPET-CT scan.

After immobilization in a vacuum mattress a contrast-enhanced thoracic CT scan including
four dimensional respiratory triggering is carried out. CT data are synchronized with the
recorded respiratory signal and four-dimensional reconstructions are performed to evaluate
the motion of the thoracic organs and the tumor during the breathing cycle. Based on the
4D-CT data set, radiation treatment planning is carried out as inverse planning.

Radiation therapy is performed as intensity modulated radiation therapy (IMRT). A dose of
50-54 Gy is applied to the primary tumor and mediastinal lymph nodes in daily fractions of 5
× 2 Gy. Subsequently, the primary tumor and involved lymph nodes are boosted to a total dose
of 60-72 Gy in daily fractions of 5 × 2 Gy. Tolerance doses of thoracic organs at risk are
not exceeded.

Treatment is carried out on an out-patient basis unless the condition of the patient
requires hospital administration.

18F-FMISO is provided by Iason (Graz, Austria). 18F-FMISO dPET-CT investigations are
performed prior to radiotherapy, at week 5 of radiation therapy and at 6 weeks post
treatment. dPET-CT examinations are performed after the i.v. injection of 18F-FMISO using an
Biograph mCT S128(Siemens Medical Solutions Co., Erlangen, Germany). The dynamic studies are
acquired with a 28-frame protocol for one hour. Quantification is performed following the
iterative reconstruction of the dPET-CT data using a dedicated software package. Generally,
volumes-of-interest (VOIs) are placed over the tumor and reference regions, followed by a
compartment and non-compartment analysis. A two-tissue compartment model is the model of
choice and five parameters will be obtained. The quantification includes the calculation of
the fractional blood volume, also named as vessel density (VB), the parameters k1 and k2,
which reflect the influx and efflux of FMISO into and out of the cells, and k3 and k4, which
are related to the trapping and re-oxygenation of FMISO. For the input function the mean
value of the VOI data obtained from a large arterial vessel like the descending aorta is
used. Besides the VOI based analysis, parametric images are calculated to assess dedicated
parameters of the FMISO kinetics.

Besides the compartment analysis a non-compartment model based on the fractal dimension is
used. The fractal dimension (FD) is a parameter for the heterogeneity and is calculated for
the time activity data of each individual VOI. The values of the fractal dimension vary from
0 to 2 showing the deterministic or chaotic distribution of the tracer activity. We use a
subdivision of 7x7 and a maximal SUV of 20 for the calculation of FD.

Functional MRI investigations are performed prior to radiation therapy, at week 5 of
radiation therapy and at 6 weeks post treatment. All examinations are performed using a
clinical 1.5-T MRI scanner (Magnetom Avanto, Siemens Medical Solutions, Erlangen, Germany).

The standard protocol comprises a coronal and a transversal breath-hold TrueFISP, T2w
single-shot half-Fourier TSE (HASTE) and T1w 3D-GRE (VIBE) sequence. Afterwards, a navigator
triggered transversal T2w-FatSat sequence (T2-FS BLADE) is performed.

Diffusion weighted imaging is performed using an axial single shot echoplanar (EPI) sequence
with and without flow-compensation. A total of ten b-values (0, 10, 25, 50, 100, 200, 300,
400, 500 and 800 s/mm2) is acquired enabling extraction of diffusion and perfusion
parameters. DWI parameters are evaluated based on the Intravoxel Incoherent motion (IVIM)
model, yielding the parameters perfusion fraction f and diffusion constant D, using in house
developed open-source software MITK Diffusion, Version 2011 (downloadable at www.mitk.org).
The parameter estimation is based on the assumption that the diffusion measurement is
influenced by mainly two effects, a perfusion related effect introduced by the molecules
moving in the capillary network (pseudodiffusion coefficient, D*) and extravascular effects
of passive diffusion (D). Since a simultaneous nonlinear fit for all parameters D, D*, and
the weighting coefficient f can be instable, measurement at b-values greater than 200 s/mm²
are used in a first step to estimate f and D as described. D* is then calculated in a second
step by using exhaustive search.

Lung cancer perfusion is assessed using a spoiled 3D gradient echo sequence after bolus
injection of 0.07 mmol/kg body weight of Gd-DTPA. Ten acquisitions in one expiratory breath
hold (10 x 2.25 s = 22 s) are followed by 50 navigator-triggered acquisitions under free
breathing. After a co-registration of the 3D data sets, a ROI-based visualization of the
signal-time curves is performed.

Furthermore, a time-resolved echoshared gradient echo sequence (TWIST) is performed for
assessment of three-dimensional tumor motion (240 x 0.5s = TA 2 min). A dynamic 2D-TrueFISP
sequence acquired in coronal orientation crossing the centre of the tumor provides
additional information about lung and tumor motion during the breathing cycle.

Contrast-enhanced sequences breath hold 3D-GRE sequence (VIBE) complete this protocol.

The first follow-up is planned 6 weeks post treatment and includes study-related 18F-FMISO
dPET-CT and functional MRI examinations. Further regular follow-up visits are scheduled
every 3 months for the first 2 years, every 6 months for the following 3 years and
thereafter yearly. Individual trial participation is completed three years after patient
enrolment or death of the patient.

The therapeutic efficacy will be evaluated through thoracic CT-scan at follow up visits.

The study is a prospective and non-randomized trial with inclusion of 15 patients. Repeated
examinations with 18F-FMISO dPET-CT and functional MRI lead to longitudinal data for every
patient. The data consists of maps obtained from both measurement devices. Data are
quantitative measurements but may be dichotomized or categorized into more than two classes.
For all parameters, differences between the site of local relapse and a selected control
region are derived and compaired by paired tests at 5% level.

All analyses are exploratory. A sample size calculation cannot be performed because neither
standard deviation of the differences has been estimated before, nor the relevant difference
is known. Therefore, the study is a pilot study to generate hypotheses for future research.

In the frame of the radiation therapy planning study, virtual radiation therapy strategies
are compared to the radiation therapy administered to the patient.