CHAPTER 1
1.1 INTRODUCTION
Cancer is one of the leading cause of morbidity and mortality worldwide, affecting millions
of people across the globe and still remains a significant global health challenge. According
to GLOBOCAN 2022 data, over 20 million new cancer cases and 9.7 million cancer-related
deaths occurred in 2020 alone1. In India, the year 2022 saw approximately 1.4 million newly
diagnosed cancer cases and around 0.8 million fatalities¹. Thoracic cancers—which include
malignancies of the lung, breast, and esophagus—constitute about 35.2% of all cancer types¹.
Esophageal cancer is the sixth most common cause of cancer-related death worldwide and is
therefore a major global health challenge. The two major subtypes of esophageal cancer are
esophageal squamous cell carcinoma (OSCC) and esophageal adenocarcinoma (OAC), which
are epidemiologically and biologically distinct. OSCC accounts for 90% of all cases of
esophageal cancer globally9 .
Over the past decade, there have been notable advancements in cancer treatment, particularly
with the evolution of radiation therapy techniques⁴. Thoracic cancers are typically managed
through a multidisciplinary approach, with radiation therapy (RT) serving as the most
commonly utilized non-surgical treatment option⁴. Radiation therapy requires high level of
precision. Accurate toxicity prediction models are essential due to the considerable risk of
both acute and long-term radiation-induced complications commonly associated with
radiation therapy (RT)⁵. For optimal irradiation of esophageal cancer, accounting for
geometric uncertainties such as respiratory motion and day-to-day position variability of the
tumor is required. To compensate for these uncertainties, planning target volume (PTV)
margins are applied. Frequently these margins are generous, isotropic, and equal for all
patients. Consequently, organs at risk adjacent to the CTV are exposed to high levels of
irradiation, resulting in increased risk of toxicity of the heart or lungs. 5,6 Detailed knowledge
of the position variability, respiratory motion, and identification of the region of interest
(ROI) could improve target alignment in esophageal cancer irradiation. Various potential
1
sources of error can arise during the planning and delivery of treatment—defined as any
discrepancy between the intended and actual delivery—which may compromise treatment
effectiveness⁶.
1.1.1 SYSTEMATIC ERROR AND RANDOM ERROR
Systematic errors (Σ) are deviations that are consistently present across all treatment fractions
and represent reproducible setup inaccuracies. They are critical in determining planning target
volume (PTV) margins, as they shift the entire dose distribution away from the clinical target
volume (CTV).5 This errors represent uncertainties that occur in the same direction and
magnitude for each treatment session, making them the most significant contributor to target
miss if not accounted for6. Systematic setup errors typically stem from factors such as patient
positioning variations at simulation vs treatment, transfer-related setup inconsistencies,
mechanical uncertainties of treatment equipment, and human error⁷.
In contrast, random errors (σ) are unpredictable and vary from fraction to fraction. These are
caused by variable factors like breathing, organ motion, or setup inconsistencies. They lead to
dose blurring rather than a consistent shift in dose distribution 5. Random errors occur
differently for each treatment fraction and result in a random variation in dose delivery. These
need to be accounted for when determining the overall treatment margins 6 .Random setup
errors are primarily caused by internal organ motion due to respiration, digestion, and
physiological changes during treatment—such as tumor shrinkage or progression. These
errors arise from differences that accumulate during the course of treatment planning process
and the treatment session. Systematic errors are more detrimental than random errors because
they affect every treatment fraction in the same way. This leads to a shift in the cumulative
dose distribution, which can cause under-coverage of the target volume or unnecessary
radiation to organs at risk.
Accurate setup uncertainty estimation in radiation therapy is crucial for several key reasons,
such as ensuring precise tumor coverage & safe tissue dose, determining appropriate margins
(CTV → PTV), managing dose to organs at risk, improving plan robustness and reducing
replanning. Small setup errors (e.g. 3–5 mm) can significantly degrade the delivered dose to
2
the tumor volume (PTV/CTV), especially in highly conformal treatments like IMRT and
stereotactic radiosurgery. Several techniques are used, depending on the resources available,
treatment site, and required precision. Image guided radiation therapy (IGRT) has been
developed to reduce geometric uncertainties by acquiring images before treatment and
making corrections accordingly. Electronic Portal Imaging Devices (EPIDs) are widely used
for measuring and correcting setup errors in radiation therapy. EPIDs capture images using
the treatment beam itself (typically megavoltage photons) and these images are compared to
Digitally Reconstructed Radiographs (DRRs) generated from the treatment planning
computed tomography (CT) scan. Bony landmarks or fiducial markers are aligned between
EPID images and [Link] techniques include Cone-Beam CT, Orthogonal kilovoltage
imaging, fiducial marker tracking, optical tracking system.
Van Herk et al. noted that the processing chain contains approximately ten potential sources
of error7. This approach enabled the development of a formal method to calculate the
necessary margins that ensure the clinical target volume (CTV) receives at least the 95%
isodose in a uniform patient group, with a 90% level of confidence 8. The planning target
volume (PTV) serves as a safety buffer designed to ensure the CTV consistently receives the
full prescribed radiation dose.
1.1.2 van Herk’s method of setup error estimation
Van Herk's method is a widely used analytical approach in radiation therapy to estimate
margins for treatment planning. It provides a systematic way to calculate the necessary
Planning Target Volume (PTV) margins that account for both systematic and random setup
errors, ensuring adequate dose coverage to the Clinical Target Volume (CTV) across a
population of patients. The method defines the PTV margin as:
Margin=2.5⋅Σ+0.7⋅σ
where Σ (systematic error) is the standard deviation of the mean setup error across
patients, and σ (random error) is the standard deviation of daily variations within a patient.
3
The factor 2.5 is derived to ensure that 90% of patients in a population receive at least 95%
of the prescribed dose to the CTV. This term accounts for systematic errors, which
consistently shift the dose distribution for a given patient across all treatment fractions and
have the most significant impact on long-term underdosage of the target.
The 0.7 factor accounts for random errors, which vary from day to day and cause a blurring
of the dose distribution rather than a consistent shift. While they are less impactful on
population-wide coverage, they still reduce the minimum dose received by the CTV due to
intra-patient variability.
These constants were derived analytically using dose-population histograms and Gaussian
error models, with assumptions of a uniform dose gradient at the edge of the target and a
normal distribution of setup errors.
1.2 AIM
1.3. OBJECTIVE OF THE STUDY
The primary objective of setup error estimation in the treatment of esophageal cancer is to
ensure accurate and consistent delivery of the prescribed radiation dose to the clinical
target volume (CTV) while minimizing exposure to surrounding healthy tissues, such as the
heart, lungs, and spinal cord. To ensure this, calculation of PTV margin is a necessary step.
This study focuses on computing the PTV margin by estimating systematic and random
errors for a set of patients with esophageal cancer.
1.4 LITERATURE REVIEW
1. The paper titled “Analysis of setup errors and determination of planning target volume
margins for thorax (lung, esophagus and breast) cancers” conducted at the Regional
4
Radiotherapy Centre in Kashmir assessed patient setup inaccuracies in thoracic cancers—
specifically, lung, esophagus, and breast cases—to determine the necessary margin between
the Clinical Target Volume (CTV) and Planning Target Volume (PTV). Using data from 51
patients across 1,308 portal images, the authors compared daily electronic portal imaging
(EPID) with digitally reconstructed radiographs (DRRs) to quantify displacements in three
orthogonal axes: medio-lateral (ML), cranio-caudal (CC), and antero-posterior (AP). Both
systematic (consistent) and random (day-to-day) errors were calculated to evaluate
precision in patient positioning.
The analysis revealed systematic errors ranging approximately from 1.0 to 1.8 mm in ML and
CC directions, and up to around 3.1 mm in AP. Random errors were lower: typically between
0.4 and 0.8 mm in ML and CC, and up to 1.7 mm in AP direction. Applying the van Herk
margin recipe, margins were estimated per site: for lung cancer, 4.7 mm (ML), 3.3 mm (CC),
and 8.8 mm (AP); for esophagus, 3.6 mm, 2.7 mm, and 5.7 mm respectively; and for breast,
3.0 mm, 4.9 mm, and 8.6 mm in ML, CC, and AP axes, respectively.
The authors conclude that extending the CTV by approximately 8–9 mm in the AP axis
provides sufficient margin to ensure that 90% of patients receive at least 95% of the
prescribed dose to the CTV. These site-specific margins serve as upper limits aimed at
optimizing radiotherapy precision and maintaining clinical target coverage while accounting
for observed uncertainties.
2. The paper titled “Patient setup error and day-to-day esophageal motion error analyzed by
cone-beam computed tomography in radiation therapy” by Hideomi Yamashita et al.
focuses on analyzing two main sources of error in radiation therapy for esophageal cancer:
patient setup errors and day-to-day esophageal motion. The authors used cone-beam
computed tomography (CBCT) to evaluate these uncertainties, aiming to enhance the
accuracy of radiation delivery to the tumor while minimizing damage to surrounding healthy
tissues.
5
The study found that patient setup errors, which occur when patients are positioned
incorrectly for each treatment session, were generally small, with displacements averaging 2-
3 mm in the superior-inferior, left-right, and anterior-posterior directions. These errors are
typically systematic, meaning they occur consistently across treatments, and random,
meaning they vary from session to session. While the errors were relatively small, they could
still impact the precision of radiation delivery, especially in areas close to critical structures.
A more significant factor was the motion of the esophagus itself during treatment. The study
demonstrated that the esophagus undergoes considerable day-to-day motion, particularly in
the superior-inferior direction. This organ motion can vary each day, making it difficult to
target the tumor accurately without compensating for the motion. The study emphasized the
need to consider this motion when defining the planning target volume (PTV) margins to
ensure that the tumor receives the prescribed dose while avoiding healthy tissues.
The authors calculated the necessary PTV margins by combining both setup errors and
esophageal motion. They recommended a margin of approximately 5-7 mm in all directions
to account for both systematic and random errors in positioning and organ motion. This
margin helps ensure that the tumor is adequately covered, while minimizing unnecessary
radiation to surrounding tissues. The study aligns with other research that supports the use of
personalized margins based on real-time imaging, rather than relying on fixed margins that
might not account for patient-specific motion or setup errors.
3. The study "Setup Variations in Radiotherapy of Esophageal Cancer: Evaluation by Daily
Megavoltage Computed Tomographic Localization" explores the impact of setup variations
in the radiation treatment of esophageal cancer, using daily megavoltage computed
tomography (MVCT) for precise patient localization. The authors aim to assess how these
setup variations affect the accuracy of radiation delivery and to determine whether adjusting
the planning target volume (PTV) margins can minimize treatment uncertainties.
The authors note that in radiotherapy for esophageal cancer, patient setup errors can
significantly impact the precision of radiation delivery, leading to either under- or over-
6
irradiation of the tumor. To evaluate these errors, they utilized daily MVCT scans, which
allowed for a more accurate localization of the patient’s anatomy compared to traditional
methods. The study found that setup variations were common, with the greatest discrepancies
occurring in the superior-inferior (SI) direction, followed by the left-right (LR) and anterior-
posterior (AP) axes. These variations were typically within 3–4 mm, but errors in the SI
direction were more prominent and contributed to the overall uncertainty in treatment
delivery. The authors also distinguished between systematic and random errors. Systematic
errors were observed to be more consistent across treatment sessions, whereas random errors
varied more from day to day. They determined that systematic errors were typically small but
significant enough to impact treatment accuracy. Random errors, while smaller, still
contributed to uncertainty, necessitating larger PTV margins to account for these variations.
By identifying these errors through daily MVCT scans, the study demonstrated the
importance of real-time imaging to monitor and correct patient positioning during each
treatment session.
The study’s key conclusion was that larger PTV margins are needed to account for these
setup variations, particularly in the SI direction, where errors were more pronounced. The
authors recommended that margins be adjusted according to the observed setup deviations,
suggesting a margin of at least 5–7 mm to ensure tumor coverage and minimize radiation to
surrounding healthy tissues. This personalized approach to determining margins contrasts
with the traditional use of fixed margins and emphasizes the importance of daily imaging for
improved treatment accuracy.
4. The study titled “Esophageal Motion Characteristics of Setup Errors in Esophageal
Cancer” by Das et al. investigates both intra-tumoral motion and setup uncertainties in
patients receiving definitive or preoperative RT. Their findings highlight that tumor motion is
most pronounced in the superior-inferior (SI) direction, with mean movements reaching
nearly 3.7 mm, followed by moderate displacements in the mediolateral (ML) and
anteroposterior (AP) directions. The magnitude of motion was especially significant in
tumors located in the lower third of the esophagus, including the gastroesophageal junction
(GEJ), underscoring the need for site-specific planning margins.
7
The study also compares setup errors using two image-guidance techniques: portal imaging
and cone-beam computed tomography (CBCT). Portal imaging showed larger setup
deviations—up to 13 mm in the ML and 11 mm in the SI directions—while CBCT resulted in
smaller errors, suggesting improved precision due to the ability to visualize and correct for
rotational and anatomical misalignments. By applying van Herk's margin recipe, the authors
proposed planning target volume (PTV) margins of approximately 12.65 mm (SI), 11 mm
(ML), and 5.2 mm (AP) to ensure adequate target coverage while compensating for motion
and setup variability. These findings are consistent with a broader body of literature that
emphasizes the predominance of SI motion in the esophagus, particularly in the lower third
and GEJ region.
In conclusion, the study by Das et al. contributes valuable data supporting the implementation
of direction-specific and site-specific margins in RT planning for esophageal cancer. Their
work aligns with existing literature advocating for individualized treatment strategies based
on tumor location and patient-specific motion characteristics. Integration of modern imaging
modalities such as CBCT and possibly MR-guided systems offers enhanced precision,
enabling smaller margins and potentially reducing radiation-induced toxicity to adjacent
organs.
5. In the study titled “Residual setup errors and dose variations with less-than-daily image
guided patient setup in external beam radiotherapy for esophageal cancer” Han et al.
investigated how different frequencies of image-guided radiotherapy (IGRT) affected residual
setup errors and dose delivery in patients with esophageal cancer receiving helical
tomotherapy. The authors analyzed data from 25 consecutive patients who underwent daily
megavoltage CT (MVCT) imaging before each treatment fraction, accumulating a total of
625 imaging sessions. They simulated seven distinct IGRT schedules—ranging from no
imaging at all to approximately 60 % imaging frequency—and calculated the corresponding
residual shifts that would remain uncorrected under each protocol by comparing the actual
daily couch shifts with what would have been applied per each less-frequent imaging
strategy.
8
Dosimetric consequences of these residual errors were quantified in a subset of five
representative patients using adaptive planning software. Even with an optimized strategy
involving imaging on first five fractions followed by alternate days (≈ 60 %), they found that
10% of treatment fractions still experienced a >10% reduction in CTV D95, indicating
undercoverage. Moreover, sizable increases were seen in doses delivered to organs at risk—
lung V0.8Gy increased in 14% of fractions, and heart V1.2Gy increased in 13%, compared
with daily imaging scenarios.
From the simulation data, Han et al. derived PTV margins necessary for different IGRT
frequencies: with no imaging, margins of approximately 13 mm (left-right), 14 mm (superior-
inferior), and 5 mm (anteroposterior) were required; even with 60% imaging, the margins
decreased modestly to 10 mm Left-right (LR), 11 mm SI, and 5 mm AP to maintain target
coverage.
Ultimately, the study concluded that inadequate imaging frequency leads to substantial
residual setup uncertainties and clinically meaningful dose variations, even when using highly
conformal delivery. Consequently, they strongly recommended daily image guidance
throughout the entire course of conformal radiotherapy for esophageal cancer to ensure
consistent target coverage and minimize unintended dose exposure to critical structures.
9
10