Individualized Dose Escalation for non-small cell Lung cancer (NSCLC) using volumetric modulated arc therapy (VMAT)

In the Netherlands, approximately 10.000 new patients are diagnosed with lung
cancer every year. Of these, 80% present with non-small-cell lung cancer.
Between 1995 and 2008, the national incidence has risen with 16% caused by an
impressive increase of 53% in women suffering from this disease. The aggressive
nature of this disease leads to a one-year survival rate of 45% and a 5-year
survival rate of only 14%.
It is widely accepted that surgery provides the best chance of cure in patients
with resectable NSCLC (www.oncoline.nl). In practice, only 20% of patients are
amenable for tumor resection with curative intent. Alternatively, stereotactic
body radiation therapy (SBRT) results in excellent local control in localized
early stage disease (1).
In locally advanced, irresectable disease, combined chemotherapy and
external-beam radiotherapy (EBRT) are increasingly being used. Evidence
suggests that concurrent schedules are more effective than sequential
treatments despite increased toxicity, although the true magnitude of the
additional benefit remains uncertain (2). However, a large number of patients
with locally advanced NSCLC is not suitable for concurrent chemoradiotherapy
due to their general condition, age, comorbidity or tumor-related factors (3).
Therefore, there is a need to increase effectiveness of treatment for all
patients with advanced stage NSCLC undergoing either radiotherapy alone,
neoadjuvant chemotherapy followed by radiotherapy, or concurrent treatment.

Apart from the addition of chemotherapy, treatment modification by
intensification of the radiotherapy schedule or by dose escalation has been
proven beneficial (4). Several phase I/II trials explored altered EBRT
fractionation schedules that increased the biological effective dose to the
primary tumor and reduced local relapse rate (5-7). Thereby, two main
principles were pursued: reduction of the dose per fraction (* 1.8 Gy), giving
two or three fractions per day (so-called hyperfractionation), aimed at sparing
normal tissues while increasing the dose to the primary tumor; increase of the
fraction dose (* 2 Gy), combined with a reduction in the total number of
fractions (so-called hypofractionation) aimed at increasing the effective tumor
dose in less radiation-sensitive primary tumors. On the one hand,
hyperfractionation limits the treatment-related side-effects, on the other
hypofractionation is attractive for the patient and radiation department as the
number of treatment fractions can be reduced.

Intensification of the irradiation schedule by continuous, hyperfractionated
radiotherapy (CHART) delivered in 12 consecutive days showed an absolute
improvement in two-year survival of 9% comparing the CHART schedule with a
conventional scheme in six weeks (6). Due to its heavy logistic load (i.e., 3
fractions per day with an 6-8 hour interval allowing for normal tissue repair)
it is only available in a few centers. With the advent of highly conformal dose
planning and delivery techniques during the last decade (i.e., 3-dimensional
conformal radiation therapy, 3D-CRT; intensity-modulated radiation therapy,
IMRT; volumetric-modulated arc therapy, VMAT/RapidArc; Tomotherapy),
organ-sparing technology became widely available. Recently, van Baardwijk and
collaborators published an individualized dose prescription study in 166
stage-III NSCLC patients (7). Using a hyperfractionated 3D-CRT technique,
patients were treated to the maximally tolerable dose (MTD) by increasing the
fraction number until normal tissue constraints for the healthy lung tissue and
the spinal cord were met (7). One-year and two-year overall survival was 69%
and 45%, respectively, with acceptable toxicity. Already in 2006, Belderbos et
al. reported favorable toxicity data and an encouraging failure-free interval
in 88 inoperable NSCLC patients treated with intensified, hypofractionated
3D-CRT based on the MTD to the lung (5). With a median total tumor dose of 80.1
Gy (range 49.5 Gy to 94.5 Gy), dose-limiting toxicity was only observed in 9
patients.

Apart from these reported studies, there are three hypofractionation trials
being conducted elsewhere. In the UK, two 3D-CRT based phase I/II trials have
been approved investigating individualized dose escalation based on normal
tissue dose constraints in patients with stage II or stage III NSCLC. In the
IDEAL-CRT trial (ISRCTN12155469), concurrent chemotherapy is combined with a
total radiation dose of 63*73 Gy given in 30 fractions over 6 weeks. The dose
is prescribed such that each patient has the same risk of grade *2 radiation
pneumonitis, but it may be limited by the tolerance dose of the spinal cord or
esophagus. Based on this, patients are allocated to one of seven total dose
levels ranging from 63 Gy to 73 Gy in 30 fractions. The second trial, IsoToxic
Accelerated RadioTherapy (I-START; CRUK/10/005), has recently been approved for
patients with locally advanced NSCLC to be treated with radiation only.
According to this protocol, the dose to an individual patient is prescribed as
the maximum achievable dose within the range of 58*65 Gy in 20 fractions over 4
weeks, by observing established dose constraints to organs at risk. The primary
endpoint of both trials is to establish the maximally tolerable dose to the
esophagus and lung during and shortly after treatment. In the US, the
University of Wisconsin is conducting a helical tomotherapy-based
hypofractionation study (NCT00214123) with pulmonary toxicity (pneumonitis
grade 3 lasting for more than 2 weeks) as primary endpoint. The purpose of
this trial is to pilot reducing the duration of radiation treatment for lung
cancer patients from 6 to 5 weeks using tomotherapy.

The reported hypo- and hyperfractionation studies have a *trial-and-error*
approach for dose-level estimation in common. In a recent in silico trial in 26
stage III NSCLC patients, we have investigated the use of a dedicated software
tool for individual dose escalation by hypofractionation (8). Based on an
existing, clinical IMRT treatment plan (66 Gy in 33 fractions), radiation dose
was escalated by scaling the radiation dose until the maximum tolerated dose
constraints for the healthy lung, the esophagus, spinal cord, brachial plexus
or heart was met. By doing so, the physical total tumor dose could be escalated
by 2.3 to 21.1 % (range 67.5 Gy to 79.9 Gy). Further dose escalation was
hampered by the maximally tolerated esophageal dose in 58% of the patients, and
by the mean pulmonary dose in 23% of the patients.

The aim of this present study is to test the feasilibity and toxicity of
individualized hypofractionated radiotherapy, and to report outcome data. In
case this phase II trial has favorable results, a phase II/III trial on
maximally tolerable, individualized, hypofractionated radiotherapy within a
shorter overall-treatment time is aimed for.

i. Primary endpoint
Treatment toxicity in terms of acute or late grade 2-4 esophageal and pulmonary
adverse events, or other grade 2-4 adverse events (RTOG Acute Radiation
Morbidity Scoring Criteria and RTOG/ESTRO Late Radiation Morbidity Scoring
Schema); (maximum) expected increase in normal tissue complication probability
of the lung (NTCPlung) is 18.5% [see 3.c.].

ii. Secondary endpoint
Detection of increase in tumor control probability (TCP) of (expected) 20% [see
3.c.]
Local-regional failure
Progression-free survival
Overall survival
Death during or within 30 days of discontinuation of radiation treatment
Quality of life (EORTC questionnaire QLQ-C13 and QCQ-LC13, H.A.D.S.
questionnaire)

Predictive value of 18F-fluoroglucose (18FDG-)PET of the primary tumor and
metastatic mediastinal lymph nodes performed at three time-points: before,
during (2nd week) and (3 months) after treatment; findings on 18FDG-PET will
not influence the treatment.

For patients undergoing surgery after (chemo)radiation (in accordance with the
local protocol):
Rate of pathological complete responses of the primary tumor and rate of
non-viable tumor cells in
treated mediastinal lymph nodes
Correlation of immunohistochemical findings with alterations of the 18FDG-PET
signal

iii. Final aim:
Ultimately, this phase II and the possible consecutive phase III trial aim at
improving treatment outcome for patients with stage IIIA/B NSCLC without
increasing toxicity to unacceptable limits.

This is a non-randomized single-centre open label study.
a. Patients with stage IIIA/B NSCLC scheduled to undergo radiation treatment
(with sequential or concurrent chemotherapy) with curative intent (33 fractions
in 6 * weeks) are asked to participate in this study.
b. Before initiation of treatment, all patients undergo a routine 18FDG-PET-CT
scan at the Department of Nuclear Medicine for tumor delineation and treatment
planning purposes.
c. For all patients participating in this study, an individualized,
hypofractionated treatment plan is generated on the basis of an original,
clinically acceptable plan. No randomization is performed.
d. Based on the radiation doses to the organs at risk (lung, esophagus, spinal
cord, brachial plexus and heart), individual dose escalation in keeping with
the maximally tolerable constraints is performed. This is achieved by scaling
the prescribed dose of the existing treatment plan using a dedicated software
routine.
e. During the treatment, the treating physician weekly scores the patients*
toxicity (RTOG Acute Radiation Morbidity Scoring Criteria).
f. At three-weekly intervals (during radiotherapy or sequential/concurrent
chemoradiation), the patients are asked to complete a quality of life
questionnaires (EORTC questionnaire QLQ-C13 and QCQ-LC13, H.A.D.S.
questionnaire). This is repeated when patients are seen for routine follow-up.
g. After completion of treatment, patients will regularly be followed at the
Department for Radiation Oncology/Pneumology to assess late toxicity
(RTOG/ESTRO Late Radiation Morbidity Scoring Schema).

i. Patients included in this study will be subjected to the following
procedures:
Standard 18FDG-PET-CT-scan with intravenous contrast agent for radiotherapy
planning purposes prior to treatment
18FDG-PET-CT-scan with intravenous contrast agent in the end of the 2nd week of
treatment
18FDG-PET-CT-scan with intravenous contrast agent 3 months after the end of
treatment
Completion of quality of life questionnaires (QLQ-C13, QCQ-LC13, and H.A.D.S.)
three-weekly during treatment and at regular follow-up visits thereafter.

ii. Procedures not affecting the patient
A clinically acceptable radiation treatment plan will be generated using the
institute*s standard treatment planning system (Pinnacle3). Subsequently, the
radiation dose of the standard treatment plans is rescaled using
special-purpose MATLAB routines.
The treating physician will assess acute radiation related morbidity on a
weekly basis using the RTOG/ESTRO Acute Radiation Morbidity Scoring Schema.

i. Patients included in this study will be subjected to the following
procedures:
Standard 18FDG-PET-CT-scan with intravenous contrast agent for radiotherapy
planning purposes prior to treatment
18FDG-PET-CT-scan with intravenous contrast agent at the end of the 2nd week of
treatment
18FDG-PET-CT-scan with intravenous contrast agent 3 months after the end of
treatment
Completion of quality of life questionnaires (QLQ-C13, QCQ-LC13, and H.A.D.S.)
three-weekly during treatment and at regular follow-up visits thereafter.

ii. Regular follow-up visits after the end of treatment
After the end of treatment, regular follow-up appointments are scheduled at the
Department of Radiation Oncology/Pneumology:
first 6 months: monthly;
until end of 2nd year: every 3 months;
3rd year: every 6 months;
annually until death thereafter.

At these appointments, the relevant medical history will be taken and a general
physical examination performed. Furthermore, late radiation induced toxicity is
scored by the treating physician using the RTOG/ESTRO Late Radiation Morbidity
Scoring Schema.

The potential risks include increased treatment-related toxicity, e.g.,
pulmonal complaints and dysphagia on the basis of esophagitis. The additional
radiation exposure caused by two additional 18FDG-PET-CT scans is in the light
of the external-beam radiation dose negligible.