The British Journal of Radiology, 77 (2004), 8896 E 2004 The British Institute of Radiology

DOI: 10.1259/bjr/84246820

Review article
Clinical use of intensity-modulated radiotherapy: part I
M T GUERRERO URBANO, MRCPI, FRCR and C M NUTTING, MRCP, FRCR, MD
Radiotherapy Department and Head and Neck Unit, Institute of Cancer Research and Royal Marsden NHS Trust,
London and Surrey, UK

Abstract. Intensity-modulated radiotherapy (IMRT) is a novel conformal radiotherapy technique which is

gaining increasing clinical use worldwide. This article aims to summarize the published data pertaining to
clinical indications of this therapy for head and neck, central nervous system, and lung tumours. The main
indications in head and neck cancer are parotid gland sparing and dose escalation to tumours close to organs at
risk. For central nervous system tumours, IMRT has been used to reduce normal tissue radiation by more
conformal dose distributions. To date, the majority of reports concern patients treated in the context of clinical
trials, and for most tumour types longer term follow up of treated patients will be required to confirm the
clinical benefits of IMRT.

Intensity-modulated radiotherapy (IMRT) is a new all present in complicated arrangements, in the relatively
development in three-dimensional conformal radiotherapy small volume. The organs at risk (ORs) include spinal cord,
(3DCRT) combining several intensity-modulated beams to brain stem, optic nerves, oesophagus and salivary glands
provide improved dose homogeneity and highly conformal that often lie very close to the target volume which com-
dose distributions. This allows improved sparing of normal monly has an irregular concave shape. Partial reductions
tissues in many tumour sites. Radiotherapy planning studies of the volume of normal tissue irradiated, such as those
have confirmed the dosimetric advantages of IMRT over offered by 3DCRT, often do not reduce the risk of late
conventional or conformal techniques, and recently some toxicity. This is because the most critical OR (optic nerve,
studies evaluating its clinical impact have been published. spinal cord, brain stem, optic chiasm) have in-series
These have mostly been reports of single-centre experience organization of functional subunits, where partial volume
and some Phase I/II clinical studies that have reported high reductions do not significantly reduce the risk of radio-
levels of tumour control and/or a reduction in normal therapy-induced damage. Because of this, the dose to the
tissue radiation toxicity. There is to date no randomized planning target volume (PTV) sometimes has to be
clinical trial data to prove conclusively an advantage of compromised. IMRT allows more conformal dose dis-
IMRT over conventional radiotherapy. Despite this, IMRT tributions, and plans can be produced with the aim of
is rapidly becoming part of the standard treatment of conformal avoidance of critical OR or dose escalation of
patients with prostate and head and neck cancer, parti- the PTV.
cularly in centres in the USA. This paper aims to discuss the The head and neck region can be readily immobilized,
clinical use of IMRT and to summarize the available clinical and accurate assessment of set up uncertainties can be
data. Due to the wealth of data available this review has been made. This makes head and neck cancer an ideal model for
split into two parts. Part I covers tumours of the head and IMRT because the tight dose gradients that can be
neck region, central nervous system, and lung. Part II achieved with IMRT can be used to avoid OR located
discusses IMRT use for prostate, gynaecological, breast and close to the PTV.
gastrointestinal malignancies as well as other issues related
to the clinical use of this new technique.
Techniques
For locally advanced head and neck cancer, it has
Head and neck cancer become popular to treat with simultaneous integrated
boost (SIB) [1, 2] or simultaneous modulated accelerated
Head and neck squamous cell carcinoma displays a clear radiotherapy (SMART) techniques [3]. These are char-
radiation doseresponse relationship, with both the prob- acterized by the delivery of a different dose-per-fraction to
ability of tumour control and the risk of radiation-induced different targets within the head and neck region. For
normal tissue damage increasing with radiation dose. example in the Cancer Research UK Parotid Sparing
Treatment with radiotherapy is curative for many patients IMRT trial (PARSPORT), a dose of 2.17 Gy per fraction
with localized disease, but with current radiation techniques, is delivered to the primary tumour site and involved
dose is limited by both acute and late side effects. lymph nodes, and 1.8 Gy per fraction to elective lymph
The anatomy of the head and neck region is very node groups. After 30 fractions the primary tumour and
complex, with bony structures, soft tissues and air cavities involved lymph nodes have received a total of 65 Gy, and
Received 2 October 2003 and accepted 1 December 2003. the elective lymph nodes 54 Gy (Figure 1).
Address correspondence to Dr C Nutting, Head and Neck Unit, The advantage of the SIB or SMART techniques is that
Royal Marsden NHS Trust, Fulham Road, London SW3 6JJ, UK. the whole treatment course is planned only once, with

88 The British Journal of Radiology, February 2004

Review article: Clinical use of IMRT: part I

xerostomia has been the subject of most head and neck

trials (Figure 2). Parotid sparing IMRT was first used in
Phase I/II studies performed at the University of Michigan
with a forward-planned step and shoot IMRT technique
using multiple non-coplanar photon beams and low-
weighted electron fields to avoid the parotid gland [16, 17].
In 15 patients treated with this technique, IMRT improved
the dose distribution and reduced dose inhomogeneity to
the primary tumour and lymph node regions compared
with standard three-field conformal plans. IMRT reduced
the radiation dose to the contralateral parotid gland to
32% compared with 93% for the standard plans, the spared
Figure 1. SMART boost technique used in PARSPORT trial parotid glands receiving a mean dose of 19.9 Gy and
showing a higher total dose (65 Gy) and dose per fraction recovering 63% of their pre-treatment stimulated salivary
(2.17 Gy) delivered to the primary tumour and involved nodes flow rates at 1 year. This compared with only a 3% recovery
(red in 3D reconstruction and green colour wash) and lower total for treated parotid glands, which received 57.5 Gy [18, 19]
(54 Gy) dose and dose per fraction (1.8 Gy) to the elective nodes (Table 1).
(purple in 3D reconstruction and orange colour wash). An analysis of 88 patients treated with parotid-sparing
IMRT allowed the correlation of radiotherapy dose with
savings in simulation, planning, delivery and verification salivary flow measurements to produce doseresponse
time compared with conventional multiphase plans [4]. curves for parotid gland function. A mean dose threshold
Radiobiologically, SIB and SMART techniques repre- was found for both stimulated (26 Gy) and unstimulated
sent accelerated fractionation schedules that may reduce (24 Gy) saliva flow rates, such that glands receiving a
accelerated repopulation of tumour clonogens [5] and mean dose below or equal to the threshold showed sub-
have shown improved tumour control [6]. Theoretically, stantial preservation of the saliva flow following radio-
the use of larger doses per fraction may be associated with therapy, which may continue to improve over time. By
increased late normal tissue radiation toxicity to structures contrast, most glands receiving mean doses above the
with a low a/b (e.g. nerves) within the high dose PTV. The threshold produced little saliva and had no recovery over
long term follow up of patients will indicate if this is a time [20].
significant clinical problem. A retrospective analysis of 28 patients with a spectrum
IMRT plans can produce concave dose distributions that of head and neck tumours treated with the MIMiC tomo-
include a midline primary tumour (e.g. larynx or thyroid)
therapy apparatus (NOMOS Corporation, Sewickley, PA)
and lymph nodes on both sides of the neck. IMRT therefore
has been published from the Baylor College of Medicine
can eliminate the use of electron fields to treat lymph nodes in
the posterior triangle (level V). This reduces dose inhomo-
geneity in the PTV, and higher minimum doses offer the
potential for improved tumour control [7].

Clinical results
Planning studies have shown potential dosimetric
advantages of IMRT over conventional radiotherapy in
head and neck tumours [8, 9]. Site specific planning studies
have been performed for maxillary sinus [1012], thyroid
[13], parotid [14, 15] and other sites that will be discussed.
Butler et al [3] reported early follow up on 20 head and
neck patients treated with a SMART boost technique
delivering 2.4 Gy per fraction to the primary disease and
metastatic lymph nodes and 2.0 Gy to the elective nodal
regions (total doses of 60 Gy and 50 Gy, respectively, in
25 daily fractions). IMRT was delivered using 3 to 5 arcs
(NOMOS PeacockTM system; MIMiC) matched to con-
ventional anterior neck fields. They reported a 95% com-
plete response rate and a 10% local recurrence rate. Radiation
Therapy Oncology Group (RTOG) grade 3 acute mucositis
was reported in 80% of patients; 45% reported moderate
xerostomia with significant relief reported within 6 months
(mean ipsilateral parotid gland dose was 23 Gy and con-
tralateral was 21 Gy).

Parotid sparing IMRT Figure 2. Parotid gland sparing intensity-modulated radio-

therapy (IMRT): a dose distribution to deliver a high dose to the
The ability of IMRT to produce dose distributions that target volume (blue contour and red colour wash) whilst sparing
may allow preservation of salivary tissue and reduction of the parotid gland (pink contours) can be achieved with IMRT.

The British Journal of Radiology, February 2004 89

 M T Guerrero Urbano and C M Nutting

Table 1. Mean stimulated salivary flow ( standard deviation) after parotid sparing intensity-modulated radiotherapy (IMRT)

Spared parotid Treated parotid p value

21 21
Pre-radiotherapy 0.400.22 ml min 0.360.31 ml min N/A
Completion of RT 0.120.07 ml min21 0.0080.02 ml min21 0.0004
3 months 0.200.21 ml min21 0.0030.01 ml min21 0.05
6 months 0.240.17 ml min21 0.0060.02 ml min21 0.001
1 year 0.250.02 ml min21 0.0110.03 ml min21 0.006

N/A, not applicable.

[21]. 10 patients were treated for tumour recurrence after post-operatively recurrent squamous cell carcinoma (SCC)
previous conventional radiotherapy, and in 18 patients of the left alveolar ridge. Four patients developed in field
IMRT was part of the primary treatment. A high degree of recurrences in the jugulodigastric nodes with superior exten-
parotid sparing was demonstrated in these patients, with sions in the ipsilateral neck. Recently consensus guidelines
less than 20% of the total parotid volume receiving greater have been published suggesting the superior limit of level 2
than 20 Gy. Long-term results are awaited. nodes to be at the caudal edge of the transverse process of
The ability to spare the parotid gland is largely affected C1 [26]. In this study all target volumes were treated with
by its proximity to the PTV, and target volume definition IMRT. The median dose and range of RT delivered to
is of critical importance. Van Asselen et al [22] showed a the PTV for gross tumour, operative bed and subclinical
linear increase in the mean parotid dose with increasing disease were 70.4 Gy (6676 Gy), 61.2 Gy (57.664 Gy),
clinical target volume (CTV)PTV margin, emphasising 50.4 Gy (4654 Gy) and after a median follow up period
the importance of immobilization and its effect on parotid of 27 months (660 months) locoregional control rate was
gland sparing. 17 patients from the Mallinckrodt Institute 86%.
received parotid sparing IMRT without a significant Chao et al [25] reported a 2 year actuarial locoregional
compromise of the dose delivered to the target volume. control rate of 85% and an ultimate locoregional control
27% (8%) of parotid gland volumes were treated to more rate after surgical salvage of 89%. In this study there was
than 30 Gy and an average of 3.3% (0.6%) of the target only one marginal recurrence, located in the region adja-
volume received less than 95% of the prescribed dose. This cent to the spared parotid gland. This was in a patient with
is mainly related to the steep dose gradient in the region a T3N0 piriform fossa tumour treated post-operatively
where the target abuts the parotid gland [23]. where recurrence was in the level II nodal region adjacent
Parotid sparing IMRT requires the mean dose to the to the spared parotid gland. Of the 11 in field recurrences,
spared gland to be less than 2426 Gy. This has raised 9 were within the high dose CTV and 2 within the low
concern about a potential risk of tumour recurrence in the dose CTV. It is interesting to note that five patients
spared area. Two studies have evaluated the risk and recurred in the lower neck, which had been treated in four
location of locoregional recurrences in patients treated patients with a matched anterior neck field and not treated
with parotid sparing IMRT [24, 25]. Locoregional recur- in one patient. Matching of multiple IMRT fields to an
rences were classified as within (in field), marginal or anteroposterior (AP) lower-neck field is inherently asso-
outside the IMRT treated volume. Out of a total of 184 ciated with dose inhomogeneities as a result of the beams
evaluated patients, there were 3 marginal failures, only one divergence and set up inaccuracies. In addition, dose
of which was in the region adjacent to the spared parotid distributions of AP neck fields are very inhomogeneous
gland (Table 2). In both studies most recurrences occurred [27]. Areas of under-dosage using this technique could
in areas of previous disease, within the radiotherapy potentially be associated with the observed increased risk
treated area. of recurrence in this area. Overall most of the recurrences
In the Dawson et al [24] study there were no failures occur within the high dose region, in agreement with
in the tissue adjacent to the spared parotid gland. One conventional radiotherapy [28]. This suggests the existence
patient recurred in the contralateral retropharyngeal nodes within this volume of a subpopulation of cells resistant to
at the base of skull following post-operative radiotherapy radiation and/or chemotherapy. Tumour hypoxia has been
for a T4N0 left tonsil/soft palate tumour. These nodes shown to be associated with radio-resistance. The use of
had not been defined as PTV, and their PTV definition IMRT to selectively dose escalate hypoxic areas, in con-
protocol has subsequently been changed to include this junction with hypoxic modifiers, is an interesting area of
area. The other marginal recurrence occurred in the research that may lead to an improvement in the thera-
ipsilateral submandibular nodes in a patient treated for a peutic ratio.

Table 2. Locoregional recurrences following parotid sparing intensity-modulated radiotherapy (IMRT)

Locoregional recurrences 28 (15.2%) Distant Locoregional and distant

In field (IF) Marginal Outside IMRT field
Dawson et al n558 9 2 0 7 1 (IF)
Chao et al n5126 11 1 5 0 0
Total n5184 20 (10.8%) 3 (1.6%) 5 (2.7%) 7 (3.8%) 1 (0.54%)

90 The British Journal of Radiology, February 2004

Review article: Clinical use of IMRT: part I

Currently, a phase I/II study of conformal and Intensity Nasopharynx carcinoma

Modulated Radiotherapy for oropharyngeal cancer is being
conducted under the RTOG (RTOG H-0022), and an Tumours of the nasopharynx, which lie in close prox-
international randomized study of parotid sparing IMRT imity to many ORs, commonly present with lymph node
versus conventional radiotherapy (PARSPORT) is under- involvement requiring bilateral nodal irradiation. Local
way in the UK, Belgium and the Netherlands. recurrence is the most common pattern of treatment
failure, with rates between 32% and 60% for advanced
tumours [3032]. IMRT provides good coverage of the
primary tumour and nodes, with the potential advantage
Larynx/hypopharynx carcinoma of sparing the parotid glands and other ORs, and dose
For patients with advanced cancer of the larynx and escalation.
hypopharynx, the anatomical position of the tumour and 23 patients with nasopharyngeal cancer were treated
regional lymph nodes relative to the spinal cord precludes with IMRT at MSKCC. Plans provided higher PTV mean
the delivery of radiotherapy in a single phase. This requires dose (77.3 Gy) compared with 3D CRT (74.6 Gy), improved
the matching of photon and electron fields around the PTV coverage and reduced dose to ORs (mean spinal cord
spinal cord, which leads to dose inhomogeneities close to dose reduced from 44 Gy to 34.5 Gy) [33].
the tumour or lymph nodes in the neck. With conventional The UCSF group reported 35 patients treated with IMRT
radiotherapy, doses as low as 38 Gy have been observed with the goal to deliver a prescribed dose of 70 Gy to 95% or
within the nodal PTV, which is considerably less than more of the gross target volume (GTV) (nasopharyngeal
those required to achieve tumour cell kill, and likely to primary and retropharyngeal nodes) and 60 Gy to 95% or
contribute to local recurrence. Using IMRT, this treatment more of the CTV (entire nasopharynx, retropharyngeal
volume can be delivered in a single phase, without the need lymph nodes, clivus, skull base, pterygoid fossae, paraphar-
to match photons and electrons, resulting in more homo- yngeal spaces, inferior sphenoid sinus and posterior third of
geneous dose distributions and spinal cord sparing to the nasal cavity and maxillary sinuses). Concurrent che-
below 40 Gy [7] (Figure 3). This may give a higher prob- motherapy was used in 91% and intracavitary brachytherapy
ability of tumour control and allow the design of dose in 32% of patients. Improved target coverage and dose
escalation studies. One such study is currently underway at escalation to a mean GTV dose of 75.8 Gy, with an average
the Royal Marsden Hospital where a SMART boost of 3% or less of the GTV receiving less than 95% of the
chemo-IMRT technique is used to treat locally advanced prescribed dose were possible, and 100% local control rate at
SCC of the larynx and hypopharynx. An initial report of 2 years was reported. The average mean dose to the CTV was
acute toxicity showed a maximum incidence of grade 3 71.2 Gy, with an average of 2% or less of the CTV
skin toxicity in 40% patients and grade 3 dysphagia and receiving less than 95% of the prescribed dose. Serial ORs
mucositis in 30% [29]. (brain stem, spinal cord, chiasm and optic nerves) were
kept within tolerance and the average doses to 50% of the
right and left parotids were 43.2 Gy and 41.0 Gy. All
patients (100%) reported RTOG Grade 0 or 1 xerostomia
2 years after therapy [34]. In an update of their experience
[35], with a total of 67 patients treated, they reported 4
year estimates of local, locoregional progression-free and
metastasis-free survival of 97%, 98% and 66%, respec-
tively. Overall dosevolume histogram (DVH) statistics
showed a mean GTV dose of 74.5 Gy, a mean CTV dose
of 68.7 Gy and average doses to 50% of the right and left
parotid glands of 34.8 Gy and 33.9 Gy. It is important to
note that in this study the authors stated that the GTV
and CTV had an in-built PTV so as to account for patient
set-up errors, and that the parapharyngeal spaces were
included in the CTV. These spaces are in continuity with
the deep lobe of the parotid gland and in our experience,
in cases where the parapharyngeal space is target volume,
and in particular the high dose target volume, achieving a
mean parotid dose below 24 Gy is very difficult without
compromising the dose to the target volume (Figure 4).
Local control rates in this update remained very good,
with one local recurrence at the primary site and one neck
recurrence. There were, however, 17 distant metastases.
Late grade 2 xerostomia was observed in 2% patients,
but increased skin toxicity was observed. This was attri-
buted to a bolus effect of the thermoplastic mask used for
Figure 3. The primary planning target volume (PTV) (in red)
is shown encompassed by the 95% (green) isodose curve (63 Gy
immobilization and the use of multiple tangential fields
in 28 fractions). The elective PTV (neck nodes in pink) is treated and the authors suggested the skin should be considered a
to a lower dose (78% isodose curve in orange; 51.8 Gy in 28 frac- sensitive structure for the inverse planning process [36].
tions), whilst the spinal cord is kept below 40 Gy (60% isodose This increase in toxicity has not been confirmed by other
curve in pale blue). groups [37]. Levendag et al [38] suggested that stereotactic

The British Journal of Radiology, February 2004 91

 M T Guerrero Urbano and C M Nutting

Figure 5. Primary planning target volume (PTV) (thyroid bed

and high risk nodal areas) encompassed by the 95% isodose
curve (green), elective nodes by the 78% isodose and the spinal
cord is kept below 40 Gy (60% isodose curve). Dose prescribed
58.8 Gy in 28 fractions to the primary PTV and 50 Gy in 28
fractions to the elective neck nodes.

thyroid bed and/or nodal areas with IMRT may be able to

improve local control and currently a phase I/II dose
escalation study is underway at the Royal Marsden
Hospital; patients with thyroid cancer, where external
beam radiotherapy is indicated, receive 58.8 Gy in 28
fractions to the primary tumour bed and 50 Gy in 28
Figure 4. The left parapharyngeal space is shown within the
high dose planning target volume (PTV), in close proximity to fractions to the elective nodal areas.
the left parotid gland, treated beyond tolerance (mean dose
above 24 Gy). On the right side, where the parapharyngeal
space is not part of the high dose PTV, sparing can be more Paranasal sinuses, aesthesioneuroblastoma
readily achieved.
Tumours of this region are often advanced at presenta-
tion and invade adjacent structures, although lymph node
radiotherapy or IMRT provided better target coverage involvement is generally uncommon. They lie in close prox-
and sparing than brachytherapy for nasopharyngeal boost imity to many ORs (lens, retina, optic nerve and chiasm,
in patients with extensive residual disease after 46 Gy pituitary gland) and irradiation to radical doses is often
external beam radiotherapy (EBRT) and/or advanced compromised due to this. Dose distribution with conven-
(T3,4) tumours. Hsiung et al [39] found that increased tional techniques is often rather inhomogeneous, with areas
vertical length and overlap of the target volume and the of low dose that may contribute to local recurrence.
brain stem, spinal cord and/or eyes predicted greater For maxillary sinus tumours, Adams et al [10] showed
dosimetric benefits with IMRT. improved target coverage and a reduction in the doses to ORs
with step and shoot IMRT when compared with 3DCRT
and conventional RT (Table 3). A report of 11 patients
Thyroid cancer
with ethmoid sinus tumours treated post-operatively at the
For patients with thyroid cancer considered at high risk University of Ghent showed that dose escalation to 70 Gy
of recurrence after thyroidectomy, external beam radio- with an optic nerve Dmax of 60 Gy was possible in 3
therapy is used sometimes in addition to radioiodine. With patients using IMRT [11]. A further report by this group
present radiotherapy techniques 32% of patients do not [40] of 47 patients treated with post-operative radiotherapy
obtain a complete response (CR), and of those attaining a for ethmoid sinus tumours, with median follow up of 32
CR, 39% relapse within the radiation portals, especially in months, revealed 3 year overall and disease free survival of
the thyroid bed. IMRT has been shown to reduce the 71% and 62%. 17 patients were treated with IMRT and
maximal spinal cord dose to 41.0 Gy and to improve sig- prescribed doses ranged between 60 Gy and 70 Gy.
nificantly the coverage of the thyroid bed and nodal target Radiation induced severe dry eye syndrome and optic
volume (dose range 18.3%) using a five-equispaced field neuropathy were observed in 7 and 2, respectively, of the
IMRT class solution when compared with conven- 47 cases, but none were in the group treated with IMRT.
tional and 3DCRT (Figure 5) [13]. Dose escalation to the Zabel et al [41] reported increased conformality and

Table 3. Comparative dosevolume histogram statistics for maxillary sinus tumours [10]

Conventional RT 3DRT 15-segment IMRT p value

Inhomogeneity (%) Volume , 95%+volume . 105% 30.1% 30.9% 17.6% p,0.001
Contralateral optic nerve (maximum dose) 65.7 Gy 64.2 Gy 56.2 Gy p,0.001
Ipsilateral optic nerve (maximum dose) 65.7 Gy 65.7 Gy 58.7 Gy p50.02

RT, radiotherapy; IMRT, intensity-modulated radiotherapy.

92 The British Journal of Radiology, February 2004

Review article: Clinical use of IMRT: part I

statistically significant sparing of the orbit and optic nerves which had the potential to reduce mucositis and ipsilateral
(p,0.05) in 13 patients with aesthesioneuroblastoma treated hearing loss. IMRT was also found to reduce the mean
to a dose of 60 Gy with inverse-planned IMRT when dose to the contralateral parotid gland and maximum
compared with CRT (Table 4). The authors reported an doses to the brain and spinal cord [47].
increased benefit of IMRT with larger and more complex
shaped volumes.
Optic nerves, chiasm, lens and retina show a positive Re-treatment
correlation between maximum dose and complication
rate [42, 43] and IMRT can potentially reduce the dose Locoregional relapse following high dose irradiation
delivered to these organs, particularly in complex volumes remains the most common form of treatment failure in
that wrap around ORs, where a radical curative dose head and neck cancer and it is often difficult to treat.
cannot be achieved without delivering a dose beyond Some cases can be salvaged by surgery, but for some sites,
tolerance to these. such as the nasopharynx, curative surgery is difficult or
impossible, and re-irradiation may be preferable. Often
this can be done with brachytherapy, which delivers highly
Skull-base tumours conformal dose distributions with steep dose-gradients,
IMRT can improve target coverage of complex-shaped but in some cases, access to position the sources is difficult
skull-base tumours, with a reduction in the dose delivered and IMRT could play a role in treatment of these patients.
to the ORs, thereby allowing the delivery of higher doses, De Neve [48] reported three patients with nasopharynx,
while the same ORs constraints can be met. oropharynx and hypopharynx recurrences following radi-
Uy et al [44] reported 40 patients with intracranial cal radiotherapy, where IMRT allowed re-treatment of the
meningioma treated using IMRT with the NOMOS system. tumour while avoiding overdose of the mandible, brain-
The median dose to the target volume was 53 Gy and stem and spinal cord. Mean PTV doses were 6373 Gy
mean dose to the optic nerve/chiasm 47 Gy with maxi- and maximum doses to the brain stem were 6067 Gy and
mum doses up to 55 Gy. Cumulative 5 year local control 2134 Gy to the spinal cord. Re-irradiation of head and neck
was 93% and 2 patients progressed, one locally and one tumours with IMRT to doses between 30 Gy and 70 Gy
distally. Two patients experienced Grade 3 or higher late with improved normal tissue sparing was has also been
CNS toxicity with one possible treatment-related death. described by Chen et al [49].
Pirzkall et al [45] demonstrated an improvement in target
conformality and target coverage in 20 patients with
benign skull-base meningiomas treated with IMRT by an Central nervous system tumours
average 10% and 36%, respectively, using 57 equispaced Several studies have compared stereotactic radiotherapy
coplanar beams. At a median follow up of 36 months, they with IMRT. Carol et al [50] reported a group of 13 (6
reported improvement of pre-existing neurological symp- previously irradiated) patients treated using the Peacock
toms in 60% of patients and 2 patients developed late system and showed an improved conformality index with
toxicity (pituitary dysfunction and visual loss). 05% of ORs exceeding dose limits. Stereotactic radio-
Kuppersmith [46] reported the use of IMRT for the therapy was shown to be preferable for small targets but
treatment of extensive and/or recurrent juvenile angiofi- IMRT allowed more sparing of normal brain tissue in
broma in three patients. Doses delivered to the tumour large (.4 cm) and moderately sized (23 cm) irregularly
ranged from 34 Gy to 45 Gy and good conformality and shaped targets [51, 52]. A small improvement in PTV
sparing of normal tissues was achieved. Good radiological coverage of convex tumours using the IMRT-tomotherapy
response was observed in all 3 cases with no endoscopic method (Peacock system; Nomos Corporation) and a
evidence of disease in two cases at 15 months and 40 transaxial method of arc delivery was shown by Khoo et al
months. No acute toxicity was reported and late toxicity [53]. However, due to the delivery method, there were
was limited to one episode of epistaxis and persistent higher doses to the optic nerves (11.211.6% higher) and
rhinitis in one patient. lens (10.315.2%). Cardinale et al [54] evaluated different
targets (ellipsoid, hemisphere and irregularly shaped with
sizes 25.3 cm) planned with 5-arc linac stereotactic radio-
Parotid tumours therapy, 6-fixed non-coplanar custom blocked fields (3D)
A planning study of IMRT for parotid tumours [14] and intensity modulation using 6 non-coplanar beams and
showed reduction of the radiation dose to the cochlea and a mini-multileaf collimator. Arc stereotactic radiotherapy
oral cavity. Beam direction optimization software gene- spared more normal brain tissue for ellipsoid lesions, but
rated a novel 4 field ipsilateral coplanar anterior and for the hemisphere and irregular tumour targets, dose
posterior paired oblique fields (15 , 45 , 145 and 170 ) [15] conformity and high/low isodose normal brain volumes
were more favourable with the IMRT technique.
Table 4. Dosevolume histogram dose statistics for aesthesio- Grant et al [55] reported one optic sheath meningioma
neuroblastoma [41] treated to 50 Gy in 25 fractions and a craniopharyngioma
treated to 50.4 Gy in 28 fractions, with the dose to the
OR CRT max. dose IMRT max. dose optic chiasm limited to 45 Gy. Fuss et al [56] reported
Left orbit 45 Gy 28.8 Gy 100% local control and hearing preservation rate (median
Right optic nerve 52.2 Gy 48.6 Gy follow up of 18.5 months) in 8 patients with acoustic
neuromas treated with fractionated stereotactic IMRT.
OR, organ at risk; CRT, conformal radiotherapy; IMRT, Stereotactic IMRT was also used in the treatment of 10 pre-
intensity-modulated radiotherapy. viously irradiated recurrent malignant gliomas by Voynov

The British Journal of Radiology, February 2004 93

 M T Guerrero Urbano and C M Nutting

et al [57]. A median dose of 30 Gy at the 71% to 93% Conclusion

median isodose line was delivered and a median overall
survival time of 10.1 months was reported. The IMRT Collaborative working group suggested that
Thilmann et al [58] reported improved target coverage direct testing of IMRT versus no IMRT is too reductionist
and a 45% reduction in the volume of bowel receiving [69]. This may be partly true, but a better dose distribution
more than 40 Gy in a case of a partially resected sacral does not necessarily correlate with better clinical outcome
chordoma. IMRT has also been used to spare the spinal or improved sparing associated with improved side effect
cord in the treatment of 8 patients with primary and profile and/or improvements in quality of life. Because of
metastatic tumours of the spine (6 re-irradiation cases) this, IMRT should be tested head to head with conven-
with no reported spinal cord complications [59]. Similar tional radiotherapy techniques where possible. Some sites
results were observed by Milker-Zabel et al [60] in 14 where an adequate dose is not achievable with conformal
patients re-irradiated for recurrent spinal metastases. radiotherapy may be suitable for IMRT outside the context
of a trial, as well as rare tumours where clinical trials are
not feasible. The most exciting application of IMRT is to
try and improve the therapeutic ratio by using its ability to
Lung cancer spare normal tissues to evaluate modified fractionation
schedules, dose escalation and chemoradiation.
In locally advanced lung cancer, the use of high
dose radiotherapy (RT) and/or concurrent chemo-RT is
associated with significant pulmonary and oesophageal
toxicity. References
The Rotterdam Oncological Study Group [61] showed 1. Wu Q, Manning M, Schmidt-Ullrich R, Mohan R. The
a reduction of 20.3% in the mean lung dose using 3D potential for sparing of parotids and escalation of biologically
missing tissue compensators as well as a reduction in the effective dose with intensity-modulated radiation treatments
total lung volume exceeding 20 Gy (V20). Derycke et al of head and neck cancers: a treatment design study. Int
[62] compared a conventional three- or four-beam 3DCRT J Radiat Oncol Biol Phys 2000;46:195205.
technique and two techniques involving, respectively, seven 2. Mohan R, Wu Q, Manning M, Schmidt-Ullrich R.
and five non-coplanar beam incidences with intensity Radiobiological considerations in the design of fractionation
strategies for intensity-modulated radiation therapy of head
modulation and showed an improvement both in TCP and
and neck cancers. Int J Radiat Oncol Biol Phys 2000;46:
lung NTCP for the IMRT plans, with no improvement 61930.
with an increasing number of fields. Brugmans et al [63] 3. Butler EB, Teh BS, Grant WH 3rd, Uhl BM, Kuppersmith RB,
showed that for lung cancer, a beam energy of 8 MV is Chiu JK, et al. Smart (simultaneous modulated accelerated
more suitable than 18 MV and that the mean lung dose radiation therapy) boost: a new accelerated fractionation
can be significantly reduced by decreasing the field sizes of schedule for the treatment of head and neck cancer with
conformal fields. Marnitz et al [64] showed a reduction of intensity modulated radiotherapy. Int J Radiat Oncol Biol
the irradiated lung volume using non-coplanar IMRT Phys 1999;45:2132.
fields. 4. Miles EA, Clark C, Guerrero Urbano MT, Dearnaley DP,
IMRT was initially reported, following radical pleurectomy/ Nutting CM. How routine can IMRT become in daily clinical
practice? Radiother Oncol 2003;68:S121.
decortication for malignant mesothelioma, by Lee et al
5. Withers HR. Inherent acceleration of tumor dose-rate in
[65]. Tobler et al [66] reported an intensity modulated hyperfractionated regimens. Int J Radiat Oncol Biol Phys
photon arc therapy technique that allows reduction of the 1988;14:400.
lung dose and Forster et al and Ahamad et al [67, 68] 6. Fu KK, Clery M, Ang KK, Byhardt RW, Maor MH,
reported 7 patients treated with adjuvant post-operative Beitler JJ. Randomized phase I/II trial of two variants of
IMRT where good CTV coverage (50 Gy to 9298% of the accelerated fractionated radiotherapy regimens for advanced
CTV) and normal tissue sparing were achieved. The most head and neck cancer: results of RTOG 88-09. Int J Radiat
severe acute side effects reported were anorexia, N/V and Oncol Biol Phys 1995;32:58997.
dyspnoea (Figure 6). 7. Clark C, Bidmead M, Mubata CD, Harrington KJ,
Nutting CM. Intensity modulated radiotherapy improves
target coverage, spinal cord sparing and allows dose
escalation in patients with locally advanced cancer of the
larynx. Radiother Oncol (In press).
8. Boyer AL, Geis P, Grant W, Carol M. Modulated beam
conformal therapy for head and neck tumours. Int J Radiat
Oncol Biol Phys 1997;39:22736.
9. Meeks SL, Buatti JM, Bova FJ, Friedman WA,
Mendenhall WM, Zlotecki RA. Potential clinical efficacy of
intensity-modulated conformal therapy. Int J Radiat Oncol
Biol Phys 1998;40:48395.
10. Adams EJ, Nutting CM, Convery DJ, Cosgrove VP,
Henk JM, Dearnaley DP, et al. Potential role of intensity-
modulated radiotherapy in the treatment of tumors of the
maxillary sinus. Int J Radiat Oncol Biol Phys 2001;51:57988.
11. Claus F, De Gersem W, De Wagter C, Van Severen R,
Vanhoutte I, Duthoy W, et al. An implementation strategy
Figure 6. Intensity-modulated radiotherapy planned dose distri- for IMRT of ethmoid sinus cancer with bilateral sparing of
bution for treatment of right pleura (reproduced with permis- the optic pathways. Int J Radiat Oncol Biol Phys
sion of Dr C Scrace). 2001;51:31831.

94 The British Journal of Radiology, February 2004

Review article: Clinical use of IMRT: part I

12. Claus F, Vakaet L, De Gersem W, Lemmerling M, Vanhoutte I, 28. Pigott K, Dische S, Saunders MI. Where exactly does failure
Vermael S, et al. Postoperative radiotherapy of paranasal sinus occur after radiation in head and neck cancer? Radiother
tumours: a challenge for intensity modulated radiotherapy. Oncol 1995;37:179.
Acta Otorhinolaryngol Belg 1999;53:2639. 29. Guerrero Urbano T, Clark C, Miles C, Bidmead M, Harmer C,
13. Nutting CM, Convery DJ, Cosgrove VP, Rowbottom C, Vini L, Harrington K, et al. Phase I/II dose escalation study of intensity
Harmer C, et al. Improvements in target coverage and reduced modulated radiotherapy (IMRT) in cancer of the thyroid,
spinal cord irradiation using intensity-modulated radiotherapy larynx and hypopharynx: report of acute toxicity. Eur J Cancer
(IMRT) in patients with carcinoma of the thyroid gland. 2003;S1(5):S39.
Radiother Oncol 2001;60:17380. 30. Vikram B, Mishra UB, Strong EW, Manolatos S. Patterns of
14. Nutting CM, Rowbottom CG, Cosgrove VP, Henk JM, failure in carcinoma of the nasopharynx: I. Failure at
Dearnaley DP, Robinson MH, et al. Optimisation of the primary site. Int J Radiat Oncol Biol Phys 1985;11:
radiotherapy for carcinoma of the parotid gland: a compar- 14559.
ison of conventional, three-dimensional conformal, and 31. Perez CA, Ackerman LV, Mill WB, Ogura JH, Powers WE.
intensity-modulated techniques. Radiother Oncol 2001;60: Cancer of the nasopharynx. Factors influencing prognosis.
16372. Cancer 196l;24:117.
15. Rowbottom CG, Nutting CM, Webb S. Beam-orientation 32. Sanguineti G, Geara FB, Garden AS, Tucker SL, Ang KK,
optimization of intensity-modulated radiotherapy: clinical Morrison WH, et al. Carcinoma of the nasopharynx treated
application to parotid gland tumours. Radiother Oncol 2001; by radiotherapy alone: determinants of local and regional
59:16977. control. Int J Radiat Oncol Biol Phys 1997;37:98596.
16. Eisbruch A, Ship JA, Martel MK, Ten Haken RK, 33. Hunt MA, Zelefsky MJ, Wolden S, Chui CS, LoSasso T,
Marsh LH, Wolf GT, et al. Parotid gland sparing in patients Rosenzweig K, et al. Treatment planning and delivery of
undergoing bilateral head and neck irradiation: techniques and intensity-modulated radiation therapy for primary naso-
early results. Int J Radiat Oncol Biol Phys 1996;36:46980. pharynx cancer. Int J Radiat Oncol Biol Phys 2001;49:
17. Eisbruch A, Marsh LH, Martel MK, Ship JA, Ten Haken R, 62332.
Pu AT, et al. Comprehensive irradiation of head and neck 34. Sultanem K, Shu H, Xia P. Three-dimensional intensity
cancer using conformal multisegmental fields: assessment of modulated radiotherapy in the treatment of nasopharyngeal
target coverage and non-involved tissue sparing. Int J Radiat carcinoma: the University of Calafornia-San Francisco
Oncol Biol Phys 1998;41:55968. experience. Int J Radiat Oncol Biol Phys 2000;47:71122.
18. Ship JA, Eisbruch A, DHondt E, Jones RE. Parotid sparing 35. Lee N, Xia P, Quivey JM, Sultanem K, Poon I, Akazawa C,
study in head and neck cancer patients receiving bilateral radia- et al. Intensity-modulated radiotherapy in the treatment of
tion therapy: one year results. J Dent Res 1997;76:80713. nasopharyngeal carcinoma: an update of the UCSF experi-
19. DHondt E, Eisbruch A, Ship JA. The influence of pre- ence. Int J Radiat Oncol Biol Phys 2002;53:1222.
radiation salivary flow rates and radiation dose on parotid 36. Lee N, Chuang C, Quivey JM, Phillips TL, Akazawa P,
salivary gland dysfunction in patients receiving radio- Verhey LJ, et al. Skin toxicity due to intensity-modulated
therapy for head and neck cancers. Spec Care Dentist 1998; radiotherapy for head-and-neck carcinoma. Int J Radiat
18:1028. Oncol Biol Phys 2002;53:6307.
20. Eisbruch A, Ten Haken RK, Kim HM, Marsh LH, Ship JA. 37. Clark CH, Miles EA, Bidmead AM, Mubata CD,
Dose, volume, and function relationships in parotid salivary Harrington KJ, Nutting CM. In regard to Lee et al.,
glands following conformal and intensity-modulated irradia- IJROBP 2002;53:630637. Int J Radiat Oncol Biol Phys
tion of head and neck cancer. Int J Radiat Oncol Biol Phys 2003;55:1150.
1999;45:57787. 38. Levendag PC, Lagerwaard FJ, de Pan C, Noever I, van
21. Kuppersmith RB, Greco SC, Teh BS. Intensity-modulated Nimwegen A, Wijers O, et al. High-dose, high-precision
radiotherapy: first results with this new technology on neo- treatment options for boosting cancer of the nasopharynx.
plasms of the head and neck. Ear Nose Throat J 1999;78: Radiother Oncol 2002;63:6774.
23851. 39. Hsiung CY, Yorke ED, Chui CS, Hunt MA, Ling CC,
22. van Asselen B, Dehnad H, Raaijmakers CP, Roesink JM, Huang EY, et al. Intensity-modulated radiotherapy versus
Lagendijk JJ, Terhaard CH. The dose to the parotid glands conventional three-dimensional conformal radiotherapy for
with IMRT for oropharyngeal tumors: the effect of reduction boost or salvage treatment of nasopharyngeal carcinoma. Int
of positioning margins. Radiother Oncol 2002;64:197204. J Radiat Oncol Biol Phys 2002;53:63847.
23. Chao KS, Low DA, Perez CA, Purdy JA. Intensity- 40. Claus F, Boterberg T, Ost P, Huys J, Vermeersch H,
modulated radiation therapy in head and neck cancers: the Braems S, et al. Postoperative radiotherapy for adenocarci-
Mallinckrodt experience. Int J Cancer 2000;90:92103. noma of the ethmoid sinuses: treatment results for 47
24. Dawson LA, Anzai Y, Marsh L, Martel MK, Paulino A, Ship patients. Int J Radiat Oncol Biol Phys 2002;54:108994.
JA, et al. Patterns of loco-regional recurrence following 41. Zabel A, Thilmann C, Milker-Zabel S, Schlegel W, Zuna I,
parotid-sparing conformal and segmental intensity-modulated Wannenmacher M, et al. The role of stereotactically guided
radiotherapy for head and neck cancer. Int J Radiat Oncol conformal radiotherapy for local tumor control of esthesio-
Biol Phys 2000;46:111726. neuroblastoma. Strahlenther Onkol 2002;178:18791.
25. Chao KS, Ozyigit G, Tran BN, Cengiz M, Dempsey JF, 42. Martel MK, Sandler HM, Cornblath WT, Marsh LH,
Low DA. Patterns of failure in patients receiving definitive Hazuka MB, Roa WH, et al. Dose-volume complication
and postoperative IMRT for head-and-neck cancer. Int analysis for visual pathway structures of patients with
J Radiat Oncol Biol Phys 2003;55:31221. advanced paranasal sinus tumors. Int J Radiat Oncol Biol
26. Gregoire V, Levendag P, Ang KK, Bernier J, Braaksma M, Phys 1997;38:27384.
Budach V, et al. CT-based delineation of lymph node levels 43. Takeda A, Shigematsu N, Suzuki S, Fujii M, Kawata T,
and related CTVs in the node-negative neck: DAHANCA, Kawaguchi O, et al. Late retinal complications of radiation
EORTC, GORTEC, NCIC, RTOG consensus guidelines. therapy for nasal and paranasal malignancies: relationship
Radiother Oncol 2003;69:22736. between irradiated-dose area and severity. Int J Radiat Oncol
27. Nutting CM, Normile PS, Bedford JL, Harrington KJ, Biol Phys 1999;44:599605.
Webb S. A systematic study of techniques for elective cervical 44. Uy NW, Woo SY, Teh BS, Mai WY, Carpenter LS, Chiu JK,
nodal irradiation with anterior or opposed and posterior et al. Intensity-modulated radiation therapy (IMRT) for
beams. Radiother Oncol 2003;69:4351. meningioma. Int J Radiat Oncol Biol Phys 2002;53:126570.

The British Journal of Radiology, February 2004 95

 M T Guerrero Urbano and C M Nutting

45. Pirzkall A, Debus J, Haering P, Rhein B, Grosser KH, Hoss 59. Kuo JV, Cabebe E, Al-Ghazi M, Yakoob I, Ramsinghani NS,
A, et al. Intensity modulated radiotherapy (IMRT) for Sanford R. Intensity-modulated radiation therapy for the spine
recurrent, residual, or untreated skull-base meningiomas: at the University of California, Irvine. Med Dosim 2002;27:
preliminary clinical experience. Int J Radiat Oncol Biol Phys 13745.
2003;55:36272. 60. Milker-Zabel S, Zabel A, Thilmann C, Schlegel W,
46. Kuppersmith RB, Teh BS, Donovan DT, Mai WY, Chiu JK, Wannenmacher M, Debus J. Clinical results of retreatment
Woo SY, et al. The use of intensity modulated radiotherapy of vertebral bone metastases by stereotactic conformal
for the treatment of extensive and recurrent juvenile radiotherapy and intensity-modulated radiotherapy. Int
angiofibroma. Int J Pediatr Otorhinolaryngol 2000;52:2618. J Radiat Oncol Biol Phys 2003;55:1627.
47. Bragg CM, Conway J, Robinson MH. The role of intensity- 61. van Sornsen de Koste J, Voet P, Dirkx M, van Meerbeeck J,
modulated radiotherapy in the treatment of parotid tumors. Senan S. Rotterdam Oncological Thoracic Study Group. An
Int J Radiat Oncol Biol Phys 2002;52:72938. evaluation of two techniques for beam intensity modulation
48. De Neve W, De Gersem W, Derycke S, De Meerleer G, in patients irradiated for stage III non-small cell lung cancer.
Moerman M, Bate MT, et al. Clinical delivery of intensity Lung Cancer 2001;32:14553.
modulated conformal radiotherapy for relapsed or second- 62. Derycke S, De Gersem WR, Van Duyse BB, De Neve WC.
primary head and neck cancer using a multileaf collimator Conformal radiotherapy of Stage III non-small cell lung
with dynamic control. Radiother Oncol 1999;50:30114. cancer: a class solution involving non-coplanar intensity-
49. Chen YJ, Kuo JV, Ramsinghani NS, Al-Ghazi MS. Intensity- modulated beams. Int J Radiat Oncol Biol Phys 1998;41:
modulated radiotherapy for previously irradiated, recurrent 7717.
head-and-neck cancer. Med Dosim 2002;27:1716. 63. Brugmans MJ, van der Horst A, Lebesque JV, Mijnheer BJ.
50. Carol M, Grant WH 3rd, Pavord D, Eddy P, Targovnik HS, Beam intensity modulation to reduce the field sizes for
Butler B, et al. Initial clinical experience with the Peacock
conformal irradiation of lung tumors: a dosimetric study. Int
intensity modulation of a 3-D conformal radiation therapy
J Radiat Oncol Biol Phys 1999;43:893904.
system. Stereotact Funct Neurosurg 1996;66:304.
64. Marnitz S, Stuschke M, Bohsung J, Moys A, Reng I,
51. Woo SY, Grant WH 3rd, Bellezza D, Grossman R,
Wurm R, et al. Intraindividual comparison of conventional
Gildenberg P, Carpentar LS, et al. A comparison of intensity
three-dimensional radiotherapy and intensity modulated
modulated conformal therapy with a conventional external
radiotherapy in the therapy of locally advanced non-small
beam stereotactic radiosurgery system for the treatment of
cell lung cancer a planning study. Strahlenther Onkol 2002;
single and multiple intracranial lesions. Int J Radiat Oncol
Biol Phys 1996;35:5937. 178:6518.
65. Lee TT, Everett DL, Shu HK, Jahan TM, Roach M 3rd,
52. Kramer BA, Wazer DE, Engler MJ, Tsai JS, Ling MN.
Dosimetric comparison of stereotactic radiosurgery to Speight JL, et al. Radical pleurectomy/decortication and
intensity modulated radiotherapy. Radiat Oncol Investig intraoperative radiotherapy followed by conformal radiation
1998;6:1825. with or without chemotherapy for malignant pleural meso-
53. Khoo VS, Oldham M, Adams EJ, Bedford JL, Webb S, thelioma. J Thorac Cardiovasc Surg 2002;124:11839.
Brada M. Comparison of intensity-modulated tomotherapy 66. Tobler M, Watson G, Leavitt DD. Intensity-modulated
with stereotactically guided conformal radiotherapy for brain photon arc therapy for treatment of pleural mesothelioma.
tumors. Int J Radiat Oncol Biol Phys 1999;45:41525. Med Dosim 2002;27:2559.
54. Cardinale RM, Benedict SH, Wu Q, Zwicker RD, 67. Forster KM, Smythe WR, Starkschall G, Liao Z, Takanaka
Gaballa HE, Mohan R. A comparison of three stereotactic T, Kelly JF, et al. Intensity-modulated radiotherapy following
radiotherapy techniques; ARCS vs. noncoplanar fixed fields extrapleural pneumonectomy for the treatment of malignant
vs. intensity modulation. Int J Radiat Oncol Biol Phys mesothelioma: clinical implementation. Int J Radiat Oncol
1998;42:4316. Biol Phys 2003;55:60616.
55. Grant W 3rd, Cain RB. Intensity modulated conformal 68. Ahamad A, Stevens CW, Smythe WR, Vaporciyan AA,
therapy for intracranial lesions. Med Dosim 1998;23:23741. Komaki R, Kelly JF, et al. Intensity-modulated radiation
56. Fuss M, Salter BJ, Sadeghi A, Vollmer DG, Hevezi JM, therapy: a novel approach to the management of malignant
Herman TS. Fractionated stereotactic intensity-modulated pleural mesothelioma. Int J Radiat Oncol Biol Phys 2003;
radiotherapy (FS-IMRT) for small acoustic neuromas. Med 55:76875.
Dosim 2002;27:14754. 69. Intensity Modulated Radiation Therapy Collaborative
57. Voynov G, Kaufman S, Hong T, Pinkerton A, Simon R, Working Group. Intensity-modulated radiotherapy: current
Dowsett R. Treatment of recurrent malignant gliomas with status and issues of interest. Int J Radiat Oncol Biol Phys
stereotactic intensity modulated radiation therapy. Am J Clin 2001;51:880914.
Oncol 2002;25:60611.
58. Thilmann C, Schulz-Ertner D, Zabel A, Herfarth KK,
Wannenmacher M, Debus J. Intensity-modulated radio-
therapy of sacral chordomaa case report and a comparison
with stereotactic conformal radiotherapy. Acta Oncol 2002;
41:3959.

96 The British Journal of Radiology, February 2004