Keywords
hepatocellular carcinoma - interventional procedures - MR-imaging - abdomen
Abbreviations
ADC:
Apparent Diffusion Coefficient
AFP:
Alpha-fetoprotein
BCLC:
Barcelona Clinic Liver Cancer
CR:
Complete response
CT:
Computed tomography
cTACE:
Conventional transarterial chemoembolization
DEB-TACE:
Drug-eluting bead transarterial chemoembolization
DWI:
Diffusion-weighted imaging
EASL:
European Association for the Study of the Liver
FLR:
Focal liver reaction
HCC:
Hepatocellular carcinoma
INR:
International normalized ratio
LI-RADS-TR:
Liver Imaging Reporting and Data System – Treatment Response
mRECIST:
Modified Response Evaluation Criteria in Solid Tumors
MRI:
Magnetic resonance imaging
MWA:
Microwave ablation
PD:
Progressive disease
PR:
Partial response
RFA:
Radiofrequency ablation
RS:
Radiation segmentectomy
RL:
Radiation lobectomy
SBRT:
Stereotactic body radiotherapy
SD:
Stable disease
SIRT:
Selective internal radiotherapy
TACE:
Transarterial chemoembolization
TRA:
Treatment response assessment
US:
Ultrasound
VIBE:
Volumetric interpolated breath-hold examination
Introduction
Over the past two decades, interventional radiology has become integral to the multimodal
treatment of hepatocellular carcinoma (HCC). Particularly for inoperable tumors, important
treatment options include local therapies, such as thermal ablation using radiofrequency
ablation (RFA) or microwave ablation (MWA), transarterial chemoembolization (TACE),
and selective internal radiotherapy (SIRT) [1]. In addition, there is increasing use of high-precision percutaneous radiotherapy
(stereotactic body radiotherapy, SBRT) [2].
Imaging plays a central role in all stages of interventional therapy: from initial
treatment planning to post-procedural success monitoring, and the exclusion of residual
tumor, as well as in structured aftercare to detect any progression at an early stage.
Among imaging techniques, magnetic resonance imaging (MRI) is considered the method
of choice because it offers higher sensitivity with equivalent specificity compared
to computed tomography, especially for lesions less than 2 cm in size. Due to its
superior soft tissue contrast, it provides more precise lesion characterization and
is thus particularly suitable for post-interventional follow-up [3]. In this context, it takes detailed knowledge of treatment-specific image patterns
to assess the morphological response and detect therapy-induced changes [4]
[5]
[6]
[7].
Standardized classification systems, such as mRECIST, EASL, and the treatment response
module of the LI-RADS classification, provide structured criteria for assessing radiological
treatment response. Each of the classification systems has different strengths, and
it is also important to understand their respective limitations [5]
[8]
[9].
The aim of this review article is to present the key aspects of post-interventional
magnetic resonance imaging following local treatment procedures for hepatocellular
carcinoma.
Materials & Methods
For this review, a structured literature search was conducted in PubMed, covering
the period from January 2000 to March 2025. The search terms used were “hepatocellular
carcinoma,” “magnetic resonance imaging,” “thermal ablation,” “transarterial chemoembolization,”
“transarterial radioembolization,” “stereotactic body radiotherapy,” and “treatment
response.” We included English and German-language original articles and reviews about
post-interventional imaging for HCC. Case reports were excluded. A total of 1,689
publications were identified. After removing duplicates and excluding thematically
inappropriate papers, 281 articles remained. After screening the abstracts, 193 papers
were shortlisted. Of these, we reviewed 98 full text versions. In addition, current
national and international guidelines were taken into account, as well as in-house
clinical experience.
Interventional treatment methods
Interventional treatment methods
Transarterial chemoembolization (TACE, DEB-TACE)
TACE is the recommended standard therapy for patients with intermediate-stage HCC
(BCLC B), if curative options such as resection, transplantation, or ablation are
not possible. The prerequisite is preserved liver function (Child-Pugh A to B7) and
the absence of extrahepatic metastases or macroscopic vascular invasion [3]. The procedure is based on the predominantly arterial supply of the HCC, while the
adjacent liver tissue is supplied by portal veins. TACE is performed on a selective
basis and can be used as conventional TACE (cTACE), where a chemotherapeutic agent
is administered intraarterially with Lipiodol as the carrier medium. Alternatively,
with DEB-TACE (drug-eluting bead transarterial chemoembolization) drug-loaded embolization
particles are used. The aim is not only to prevent ischemia-induced tumor necrosis
but also to achieve a local cytostatic effect with low systemic toxicity. The embolic
effect is present immediately after therapy and can be detected in imaging as a lack
of arterial hyper-enhancement.
Thermal ablations (RFA, MWA)
Thermal ablation using radiofrequency ablation (RFA) or microwave ablation (MWA) represents
a curative treatment option for patients with early hepatocellular carcinoma (BCLC
0 and A), particularly for tumors with a diameter of up to three centimeters and a
limited number of lesions. The prerequisite is preserved liver function and the absence
of macroscopic vascular invasion or extrahepatic metastasis [3]. Radiofrequency ablation (RFA) and microwave ablation (MWA) differ in their technical
approach, but both aim to induce irreversible coagulation necrosis of the tumor through
locally induced hyperthermia. While RFA is based on an alternating current in the
range of 400–500 kHz, MWA uses electromagnetic waves with frequencies in the range
of 915 MHz to 2.45 GHz. Compared to RFA, MWA offers higher energy penetration, shorter
application times, and less susceptibility to the “heat sink effect,” which refers
to unwanted heat dissipation through large vessels near the target lesion [10]
[11]. The thermally induced necrosis zone can be detected in the MRI scan immediately
after therapy, and it presents with a regression in size over the longer term.
Selective internal radiotherapy (SIRT)
Selective internal radiotherapy (SIRT), also called transarterial radioembolization
(TARE), is used in patients with advanced or intermediate-stage HCC, especially when
other locoregional procedures are unsuitable or have been exhausted [3]
[12]. β-Emitting microspheres, usually yttrium-90, are administered selectively via the
tumor arterial supply. Due to the radiation’s low penetration depth of 2.4 mm on average,
selective treatment can be carried out and the surrounding tissue can be protected
[13]. Pre-therapeutically, angiographic simulation with technetium-99m MAA is performed
to exclude pulmonary or gastrointestinal shunts so that no unwanted extrahepatic accumulation
occurs during therapy. Indications exist for portal venous tumor invasion without
decompensated liver cirrhosis, as well as for inoperable tumors with preserved liver
function (Child-Pugh A to B7). Originally used as a palliative measure in advanced
disease, TARE is now increasingly being used for bridging therapy before liver transplantation,
for downstaging therapy, as curative therapy in the context of radiation segmentectomy
(RS), and for radiation lobectomy (RL) in preparation for resection [14]. The embolic effect is in the background compared to radiation damage, so that post-therapeutic
changes in imaging only become apparent in the longer term [15].
Stereotactic radiotherapy (SBRT)
According to the current S3 guideline, stereotactic radiotherapy (SBRT) is a non-invasive
treatment option for patients when thermal ablation or resection is not technically
feasible or contraindicated or in the event of recurrence after thermal ablation [3]. It can be used in different stages of HCC [3]
[16]
[17]. In addition, SBRT is also increasingly used as a bridging or downstaging strategy
before liver transplantation. SBRT enables targeted tumor necrosis through precise
dose application with the best possible protection of the surrounding parenchyma [18]. The pathophysiological effects of SBRT are based on both direct DNA double-strand
breaks in tumor cells and secondary biological processes, such as endothelial damage
and a modulation of the tumor-associated microenvironment, which can promote immune-mediated
antitumoral effects [19]. Because these processes are delayed, changes that can be detected using image morphology
only occur over time [20].
Classification systems for treatment evaluation
Classification systems for treatment evaluation
The evaluation of treatment response after locoregional procedures in hepatocellular
carcinoma requires specific classification systems, as conventional size measurement
according to Response Evaluation Criteria in Solid Tumors (RECIST) only inadequately
reflects treatment success or failure. The evaluation of contrast agent behavior,
especially arterial hypervascularization and washout phenomenon, is crucial for assessing
tumor viability. For standardized reporting, various systems have been developed,
such as modified RECIST (mRECIST), the European Association for the Study of the Liver
(EASL) criteria, and the Liver Imaging Reporting and Data System Treatment Response
(LI-RADS-TR) algorithm, each of which has different focuses and limitations. With
version 2024, the LI-RADS Treatment Response Assessment (TRA) has been fundamentally
overhauled. For the first time, a distinction is made between a non-radiation TRA
core for ablation and embolization therapies and a radiation TRA core for radiation-based
procedures (SIRT, SBRT) [21]
[22]
[23]. A structured application of these systems is crucial for valid treatment monitoring
and continued clinical decision-making [5]
[8]
[24].
RECIST 1.1
The Response Evaluation Criteria in Solid Tumors in version 1.1 define treatment response
exclusively by size changes of measurable lesions (largest longitudinal diameter).
A complete response (CR) corresponds to the disappearance of all lesions, a partial
response (PR) requires at least a 30% diameter reduction, while progression (PD) is
interpreted as an increase ≥20% or the appearance of new lesions. A stable finding
is present if the change in size does not correspond to CR, PR, or PD [25].
However, after locoregional therapy of HCC, avascular (but morphologically persistent)
lesions often remain, so that RECIST 1.1 in this context does not allow a statement
to be made about tumor viability. For example, after microwave ablation, the ablation
zone may be even larger than the original target lesion due to the safety margin maintained
and the technical properties of the applicator, without, of course, indicating any
progression.
mRECIST
The modified RECIST criteria were developed specifically for HCC and only take into
account the viable parts of the tumor, which are defined as regional contrast uptake
in the arterial phase. Only the largest dimension of the contrast-enhancing component
is measured. CR is present when there is a complete absence of arterial hyper-enhancement;
PR when there is at least a 30% reduction in arterial hyper-enhancement; PD when there
is at least a 20% increase in arterial hyper-enhancement or the appearance of new
lesions, and SD is present when no significant change is measurable [5]
[26].
The greatest advantage of using the classifications based on mRECIST criteria is that
they capture and evaluate the morphological image correlate of the devascularizing
therapies, namely the decline in the arterialized components.One limitation of this
method is its lack of applicability in poorly vascularized tumors with ambiguous arterial
hyper-enhancement, for example, under anti-angiogenic systemic treatment. In addition,
inflammatory changes or bleeding can affect the contrast behavior and thus complicate
the assessment of viability.
EASL criteria
The European Association for the Study of the Liver’s criteria expand the mRECIST
system by a two-dimensional measurement instead of the one-dimensional measurement
of the contrast-enhancing tumor component in the axial plane and calculation of an
area product. CR is present when contrast uptake is no longer detectable; PR occurs
when there is at least a 50% reduction in the area of the contrast-enhancing portion;
PD occurs when there is at least a 25% increase or the appearance of a new lesion,
and SD is present in the absence of significant change [5]
[24]
[27].
This classification based on EASL criteria enables a more differentiated assessment
of irregularly-shaped lesions. However, it is methodologically more complex, potentially
less reproducible, and difficult to standardize. For this reason, its application
is limited largely to the context of studies.
LI-RADS treatment response
The Liver Imaging Reporting and Data System Treatment Response Module (LI-RADS-TR),
developed by the American College of Radiology, was widely updated in the 2024 version
[21]
[22]
[23]. For the first time, a differentiation is made between radiation-associated (Radiation-TRA)
and non-radiation-associated (Non-radiation-TRA) therapy-induced changes. In addition
to the established categories “non-viable,” “viable,” and “non-evaluable,” the new
category “non-progressing” was introduced in the Radiation-TRA, which describes persistent
but not increasing contrast uptake after SBRT or SIRT. The definition of tumor viability
has been simplified. The only decisive factor now is the presence of a “mass-like
enhancement,” regardless of contrast agent phase. In addition, ancillary features
(i.e. diffusion restriction and mild to moderate T2 hyperintensity) were integrated,
which allow for optional upgrading of lesions. The current version therefore takes
into account both the specific characteristics of radiation-induced changes and the
requirements for standardized and reproducible reporting.
One key advantage of this classification system is that a standardized categorization
is provided for each individual lesion. LI-RADS-TR, however, does not allow for quantification
in percentages over time. This makes it less suitable in many cases for use in studies.
It should also be noted that the LIRADS classification was explicitly developed only
for high-risk patients with pre-existing liver cirrhosis or other risk factors for
HCC. It is not intended for use in patients who do not have chronic liver disease.
A comparison summarizing the rating systems is shown in [Fig. 1].
Fig. 1 Overview of classification systems. CR: complete remission; PR: partial remission;
SD: stable disease; PD: progressive disease; TRA: treatment response assessment; SIRT:
selective internal radiation therapy; SBRT: stereotactic body radiotherapy.
Image morphology according to therapy type
Image morphology according to therapy type
The interpretation of post-interventional imaging requires detailed understanding
of treatment-specific morphology in order to differentiate between treatment effect,
residual tumor viability, or recurrence and complications. The following section presents
the methods already discussed with regard to their morphological image characteristics
over time.
Transarterial chemoembolization (TACE, DEB-TACE)
Post-therapeutic changes after TACE are generally caused by two factors: the ischemic
effect of the embolic agent and the cytotoxic effect of the chemotherapeutic agent
[28]. Differences, nevertheless, arise with regard to visualization of the TACE area
in follow-up imaging due to the choice of therapeutic agent, as the Lipiodol used
in conventional TACE remains in the liver tissue for a longer period of time and leads
to corresponding changes that cannot be distinguished in DEB-TACE using loaded particles.
MRI is superior to CT, particularly with regard to post-interventional imaging after
conventional TACE [29].
Follow-up after TACE
The ischemic and cytotoxic effect of TACE causes cell necrosis, which does not, however,
lead directly to a reduction in the size of the target tumor. As a result, no reduction
in size can be expected in the initial imaging after TACE, which is usually performed
one to three months after therapy. In individual cases, however, a slight increase
in size may occur due to the development of edema and bleeding [8]. Nevertheless, a decrease in the size of the treated tumor will be observed over
time and leads to a decrease in diameter in the long term.
In contrast to the slow shrinkage of the target tumor, devascularization of a previously
hypervascularized target tumor is observed immediately after TACE and should continue
during further aftercare imaging [30]. While focal, intratumoral contrast enhancement detects residual tumor or local
recurrence, a thin, hyperarterialized rim around the treated tumor is considered physiological
([Fig. 2]) [31]. The underlying cause is inflammation, which occurs immediately after therapy and
can persist for over a year. Peripherally located, wedge-shaped areas near the treated
tumor are caused by the embolizing effect of TACE and correspond to the treated liver
volume. These changes are particularly evident in the arterial phase as reduced contrast,
while in later phases the contrast adjusts to that of the surrounding tissue. These
described changes in perfusion resolve in the longer term after TACE [8].
Fig. 2 HCC in segment VIII in a 78-year-old patient. a Planning imaging using cone-beam CT prior to TACE. b The first follow-up MRI, performed Ten weeks after TACE, shows a thin peripheral
enhancement (arrow) of the HCC in the late arterial phase of the fat-saturated T1-VIBE
sequence, which is considered therapy-associated.
The internal signal of a liver tumor treated with TACE appears heterogeneous in T1-weighted
images and predominantly hypointense in T2-weighted images. However, T2-hyperintense
areas can be caused by hemorrhages, liquid necrotic changes, or reactive edema, making
it difficult to assess the treatment outcome based on signal behavior [32]. T1-hyperintense portions may occur as a result of hemorrhages or protein-rich deposits
in the treated tumor. To distinguish these from residual tumor or recurrence in the
arterial phase, subtraction images can be used ([Fig. 3]).
Fig. 3 HCC in segment V in a 69-year-old female patient. a Pretherapeutic angiography prior to TACE of an HCC in segment V (arrow). b The first follow-up MRI, performed four weeks after therapy, shows the treated HCC
as hyperintense (arrow) in the arterial phase of the fat-saturated T1-VIBE sequence.
c However, this hyperintensity is already discernible on unenhanced images. d Subtraction images show no contrast enhancement. The hyperintensity is due to intratumoral
hemorrhage or proteinaceous deposits within the tumor remnant and must not be misinterpreted
as viable residual tumor.
As mentioned above, MRI is superior to CT in assessing the therapeutic effect after
TACE, as it is not influenced by hardening artifacts caused by deposited Lipiodol.
In addition, the opposed-phase T1 sequence can be used to assess the distribution
of Lipiodol within the tumor [8].
Thermal ablation (RFA, MWA)
The ablation area can be delineated in the MRI scan immediately after therapy. It
is usually larger than the treated tumor, due to the safety margin of 5–10 mm and
the minimum size of the therapy zone determined by the electrode or antenna [33]. In some cases, the previously treated lesion can still be delineated within the
ablation zone for some time immediately after ablation ([Fig. 4]). The internal signal for the ablation zone may be heterogeneous. T1-hyperintense
areas often result from bleeding or protein-rich necrosis, whereas T2 hyperintensity
results from edema, liquefaction, or tissue remodeling [34].
Fig. 4 MRI-guided microwave ablation. a Planning imaging using fat-saturated T1-VIBE illustrates the HCC to be treated in
segment VIII of a 72-year-old male patient, showing arterial hyper-enhancement (arrow).
b The lesion exhibits washout in the late phase. c During the intervention, the applicator is advanced percutaneously under image guidance
to the lesion (sagittal plane). d The final contrast-enhanced control imaging shows an ovoid ablation zone, hypo-enhancing
relative to the untreated liver parenchyma (arrowhead). The non-viable lesion itself
remains centrally visible within the ablation zone (arrow). e Due to the puncture, arteriovenous and arterioportal shunts may occur (different
patient), which appear in the arterial phase as wedge-shaped peripheral hyper-enhancement
adjacent to the ablation zone. f Shunts do not exhibit washout in the late phase.
Follow-up after MWA/RFA
The ablation zone typically has a narrow hyperperfused rim, which is due to post-interventional
hyperemia and reactive inflammation, and it exhibits contrast enhancement without
washout. In diffusion-weighted sequences, this shows reduced ADC values [35]. This pattern is considered physiological and persists in individual cases for several
months [4]
[8]. In contrast, the occurrence of nodular, irregular contrast enhancement within or
at the edge of the ablation area can be considered suspicious for residual tumor.
Subtraction images from the arterial and native phase can also help to differentiate
true arterial enhancement from hemorrhages in cases of ambiguous hyperintensity [32]. In the authors’ experience, DWI is only partially suitable for assessing the central
ablation zone. Although recurrences with reduced ADC values and necrosis with increased
ADC values have been described [36], the signal can vary greatly.
Immediate post-interventional, perihepatic fluid collections can occur, particularly
in quite peripheral lesions as a reaction of the liver capsule to thermal injury,
and they are spontaneously regressive. However, it is important to differentiate this
fluid buildup from relevant bleeding after retracting the electrode or applicator.
Individual air figures in the ablation zone are typically visible immediately after
the intervention. These can persist for up to two weeks after therapy, are considered
physiological, and should not be interpreted as an infection. Wedge-shaped, hypervascularized
areas visible immediately after ablation correspond to arteriovenous or arterioportal
shunts resulting from the puncture ([Fig. 4]
e, f). These perfusion changes can persist for weeks to months. In contrast to these vascular
changes, subsegmental bile duct dilations may occur peripherally to the ablation zone
within a few weeks after ablation. These bile duct changes are irreversible, but they
are not a sign of recurrent cholestasis. Rather they are due to thermal damage to
the bile ducts ([Fig. 5]). Over months and years, the ablation zone shrinks and eventually persists as a
scarred, low-signal residue ([Fig. 6]).
Fig. 5 Follow-up imaging after microwave ablation. a At 6-week follow-up after microwave ablation, the ablation zone already shows a reduction
in size. In the contrast-enhanced T1-VIBE arterial phase, hyperintense structures
are visible within the ablation zone (arrow). b The hyperintensity is also discernible in the unenhanced phase and can be attributed
to hemorrhage and proteinaceous deposits. c Subtraction imaging confirms that these are not vascularized components. d The ablation area remains hypointense compared to the untreated liver parenchyma
in the portal venous phase. e At a later follow-up twelve weeks after therapy, dilated bile ducts (arrowhead) can
be delineated peripheral to the ablation zone (T2-weighted sequence). These are due
to thermal injury to the bile ducts and are not indicative of cholestasis caused by
tumor recurrence. Such damage is irreversible. f Coronal T2-weighted sequence. The dilated bile ducts appear hyperintense (arrowhead).
g In the contrast-enhanced T1-weighted sequence, the bile ducts appear hypointense.
Fig. 6 Long-term course after percutaneous microwave ablation of an HCC in segment V. a Fat-saturated T1-VIBE, arterial phase: focal HCC with arterial phase hyper-enhancement
(arrow). b Delayed phase: washout of the lesion. c First follow-up at four weeks, arterial phase: ablation zone extending to the liver
capsule; central T1 hyperintensities (hemorrhage/protein-rich necrosis); no residual
arterial phase hyper-enhancement. d First follow-up at four weeks, delayed phase: hypocontrasted ablation zone with a
thin, reactive hyperemic rim. e Ten months after ablation, arterial phase: marked volume reduction of the ablation
zone; central hyperintensities regressing. f Ten months, delayed phase: clear delineation of the hypocontrasted area; early capsular
retraction. g Twenty-eight months, arterial phase: no evidence of viable tumor. h Twenty-eight months, delayed phase: further shrinkage of the ablation scar and progressive
capsular retraction.
SIRT (selective internal radiotherapy)
After SIRT, the morphological response to treatment sets in with a delay after three
to six months [14]
[37]. Arterial hyper-enhancement with or without tumor washout can therefore persist
without indicating treatment failure ([Fig. 7]) [38]. In the early weeks after therapy, inhomogeneous signal changes are often seen in
MRI in T1- and T2-weighted sequences. T1 hyperintensities occur due to blood degradation
products or protein precipitates; T2 hyperintensities occur due to necrosis or edema.
Initially, the tumor volume often remains stable or even increases slightly over the
first four months due to inflammatory reactions or intratumoral bleeding (pseudoprogression)
[8]. A narrow enhancement rim (typically < 5 mm) at the edge of the area occurs in up
to 50% of cases within half a year and should not be misinterpreted as tumor recurrence
due to its circular configuration [32]
[39]. Nodular arterial contrast enhancement may persist even after three months in successfully
treated tumors, but may regress over time [8]. Granulation tissue can also potentially appear in this form and can often only
be differentiated through further follow-up. Occasionally, radiation-induced changes
occur in the surrounding parenchyma, recognizable as geographic wedge-shaped areas
with transient enhancement changes that may persist for up to six months or longer
and do not represent an infiltrative tumor. Fibrosis, with possible capsular retraction
and volume reduction, appears after months. This effect is particularly pronounced
in radiation lobectomy, which, in addition to atrophy of the treated liver lobe, leads
to a compensatory hypertrophy of the untreated liver [40].
Fig. 7 Follow-up after SIRT of an HCC in liver segment III in a 53-year-old patient (status
post right hemihepatectomy). a Pre-treatment, fat-saturated T1-VIBE, arterial phase: hypervascular HCC adjacent
to the resection margin (arrow). b Pre-treatment, delayed phase: washout of the lesion. c Twelve weeks after SIRT, arterial phase: thin peripheral rim enhancement around the
treated lesion (arrowheads), consistent with post-therapeutic hyperemia. d Twelve weeks, subtraction: persistent central enhancement; after SIRT, central arterial
hyper-enhancement may persist for weeks to months and should not be misinterpreted
as early treatment failure. e Nine months after SIRT, arterial phase: no arterial hyper-enhancement detectable.
f Nine months, delayed phase: subtle residual scar.
Stereotactic radiotherapy (SBRT)
The morphological changes after SBRT also occur with a delay ([Fig. 8], [Fig. 9]). In imaging up to about three months after therapy, the tumor size usually remains
unchanged or may increase slightly [38]. In up to 75% of cases, arterial hyper-enhancement persists for several months,
even in cases of histologically confirmed necrosis, and only regresses significantly
after about six to twelve months [20]. Occasionally, a thin marginal enhancement occurs, which is less frequent and less
pronounced than in SIRT. This enhancement is also caused by inflammation and usually
regresses within six months. In parallel, edematous changes develop in the peritumoral
liver tissue that can persist for up to a year or even longer; these changes are often
referred to as focal liver reaction (FLR), and they are detectable as a geographically
configured T2-hyperintense area with arterial hyper-enhancement and possible hypo-enhancement
in the portal venous phase [20]
[38]. Completely necrotic tumor parts show increased ADC values in diffusion-weighted
sequences (Navin et al., 2022). After about six months, fibrotic changes increasingly
develop, which persist in the long term and can be accompanied by significant capsular
retraction.
Fig. 8 Follow-up after SBRT of a subcapsular HCC in liver segment VII. a A subcapsular HCC in liver segment VII (arrow) appears hyperarterialized on the fat-saturated
T1-VIBE sequence prior to therapy. b The lesion shows mild hyperintensity on the fat-saturated T2-weighted sequence. c At 6-week follow-up after SBRT, the lesion shows slight progression in size. d Peritumoral edematous changes are visible within the previous radiation field (“focal
liver reaction”). e At further follow-up performed twelve weeks after SBRT, the arterial enhancement
of the lesion has markedly regressed. f Peritumoral signal alterations have also decreased. The T2-hyperintense signal changes
in the right lower lung lobe are due to radiation pneumonitis (arrowhead).
Fig. 9 SBRT and TACE of an extensive HCC in segment I in a 54-year-old woman. a Pre-treatment, fat-suppressed T1-VIBE, arterial phase: extensive hypervascular, histologically
confirmed HCC in segment I (arrow). b Pre-treatment, delayed phase: washout of the lesion. c Six months after SBRT, arterial phase: no arterial hyper-enhancement detectable;
no relevant change in lesion size. d Six months after SBRT, delayed phase: tumor size remains unchanged. e Ten months after SBRT, arterial phase: new nodular arterialized foci along the posterior
circumference of the lesion (arrowheads) indicating recurrence; TACE recommended by
the multidisciplinary tumor board. f Ten months, delayed phase: still no relevant size change. g Seven months after successful TACE (17 months after SBRT overall), arterial phase:
no arterially enhancing lesions suspicious for recurrence. h Seven months after successful TACE, delayed phase: marked reduction in lesion size.
[Fig. 10] illustrates the therapy-associated changes over time after interventional treatment.
Fig. 10 Sequence of therapy-associated changes over time after interventional treatment.
Aftercare recommendations and follow-up protocols
Aftercare recommendations and follow-up protocols
Recommendations for imaging follow-up after locoregional therapy of HCC are contained
in both the German S3 guideline and the European EASL Clinical Practice Guideline.
Both guidelines prefer the use of multiphasic dynamic MRI, with contrast-enhanced
CT examinations as a valid alternative [3]
[41].
According to the S3 guideline, the initial imaging follow-up after thermoablation,
SIRT, and TACE should take place between four and twelve weeks; for SBRT, the first
follow-up is recommended after twelve weeks, at the earliest. The aim of these initial
checks is to assess the technical success and to detect residual tumor tissue.
In the continued course of the first year, the S3 guideline recommends regular imaging
checks at intervals of approximately three months. This close-meshed approach is justified
by the significantly increased risk of recurrence within the first year, which is
about 6.5 times higher than in later periods [42].
In the second year, the follow-up interval can be extended to three to six months.
The S3 guideline recommends an aftercare period of at least two years, although the
continuation may be longer depending on the individual risk of recurrence and patient
profile. At the authors’ institution, it is common to extend aftercare to five years.
After completion of the aftercare phase, patients should be transferred to regular
ultrasound screening with semi-annual check-ups according to both guidelines. This
recommendation applies particularly to patients with existing liver cirrhosis or an
increased risk of recurrence due to chronic liver disease.
Summary
Imaging plays a central role in the post-interventional assessment of treatment response
in hepatocellular carcinoma. The type and temporal dynamics of the image morphological
changes vary depending on the type of therapy used. After thermoablation, an ablation
zone immediately appears, which shrinks over time. After successful TACE, there is
an immediate reduction in arterial hyper-enhancement, but the impacts of cytotoxic
effects only become apparent after weeks. However, after SIRT and SBRT, morphological
changes occur with a delay. Initial size stability should not be confused with a lack
of response to treatment. Key elements of the radiology assessment are arterial contrast
enhancement and washout. It can also be helpful to use subtraction images. Understanding
typical pitfalls, such as post-interventional inflammatory hyperemia, is critical
and can help to avoid misinterpretations.
Imaging should be based on standardized aftercare plans such as mRECIST or LI-RADS-TR,
which interpret the viable parts of the tumor. Looking strictly at morphological size
changes, such as in the context of RECIST 1.1, does not adequately reflect treatment
success. The version of LI-RADS-TR updated in 2024 includes the category “non-progressing”
and for the first time takes radiation-associated changes into account; it also simplifies
the definition of tumor viability. In addition, ancillary features such as diffusion
restriction or T2 hyperintensity enable more differentiated classification of ambiguous
findings and help to provide a standardized assessment.
The limitations of this review result from the heterogeneity of the post-interventional
findings, as not all of the changes described occur in every case, their development
over time can vary greatly, and combination therapies can also have an effect. The
impact of systemic treatments was not considered in this study, which limits its transferability
to multimodal therapy concepts.
