A treatment algorithm for Hepatic Angiomyolipomas: a long-term follow-up study and a systematic review

Hepatic angiomyolipoma (HAML) is a rare mesenchymal liver tumor that belongs to
a group of perivascular epithelioid cell tumors called PEComa. It is composed
of blood vessels, smooth muscle and adipose cells. The proportions of these
tissues can vary greatly (1). Previous observations demonstrate that most
patients are asymptomatic and are discovered incidentally during regular health
check-ups or follow-up examinations for other diseases (2-4). If patients do
present themselves with symptoms they usually complain of abdominal discomfort
(5).The natural history of HAML*s has not yet been clarified (1, 5). There is
an association with tuberous sclerosis in more than 50% of the AML*s in the
kidney, but this has been estimated to be only 5-15% in the liver (1). HAML*s
usually occur in non-cirrhotic livers and are not accompanied by serological
abnormalities (2, 4). Also they occur more frequently in women, however
sex-hormones do not seem to play a role in the pathogenesis and growth of HAML
(2, 6).

On imaging HAML*s have various features closely related to the proportions and
distribution of the different tissue components. Mature adipose tissue and
central thick feeding blood vessels are the most important distinctive
characteristics. HAML*s with domineering fat are easy to diagnose but the low-
and non-fat HAML*s have less typical manifestations and may easily be
misdiagnosed as other benign or malignant tumors of the liver (2). Thus, the
diagnosis of HAML based on imaging can be very difficult due to heterogeneity
(4). Therefore, pathological examination and HMB-45 staining, a
histopathological biomarker of AML, remain an important step towards the
diagnosis (4, 5).

The majority of the tumors are believed to be clinically benign, however an
increasing number of cases and aggressive changes including growth in size,
recurrence after surgical resection, metastasis and invasive growth patterns
into the parenchyma and along the vessels have been reported (1). Meanwhile,
there are numerous therapeutic strategies available but the best treatment
option for hepatic AML has remained controversial (4). Many researchers
advocate surgical resection over a non-surgical approach. However a
non-surgical approach is also being recommended, especially for patients who
are asymptomatic and have small tumors. A non-surgical approach involves close
follow-up with repeated imaging, preferably with MRI (5).

Despite the growing number of reports on HAML, little is known about the long
term follow-up outcomes after a surgical or non-surgical treatment approach.
Recent reports on possible malignant cases have led to our interest in
performing research on patients who have previously been diagnosed with HAML
and have received either a surgical or a non-surgical treatment in one of the
academic medical centres in the Netherlands and Belgium. The aim of this study
is to gain insight into the long-term clinical course of HAML*s. We will
investigate whether there is recurrence, growth, development of new tumors or
metastasis in patients previously diagnosed with HAML. The results could lead
to the development of a treatment algorithm.

The aim of this study is to get insight into the long-term clinical course of
HAML*s in order to develop a treatment algorithm.

Primary Objective: The main study parameters are recurrence, growth or
development of new HAML tumors.

The study will be a cross sectional cohort study. All patients who have
previously been diagnosed with HAML can be included in this study. These
patients will be contacted by post and asked to undergo a MRI scan in one of
the academic medical centers. The inclusion period will end 3 months after
sending out a reminder after 1 month to all potential participants. Depending
on the primary treatment (surgical and non-surgical), we will evaluate whether
the HAML*s recur, grow or whether new HAML tumors develop.

There are no known harmful side-effects with temporary exposure to the strong
magnetic field used during an MRI scan (7). Potential AE*s associated with MRI
are related to the administered contrast agents (usually gadolinium based). The
frequency of all acute adverse events ranges from 0.07% to 2.4% with the vast
majority of mild reactions (e.g. coldness at injection site, nausea, headache,
dizziness). Allergic responses (e.g. urticaria, rash hives) are very unusual
and vary from 0.004% to 0.7%. Severe, life-threatening anaphylactoid reactions
are even more rare and range from 0.001% to 0.01%. The frequency of acute
adverse reactions is higher in patients with previous reactions and second
reactions can be more severe than the first, thus those patients are excluded
from this study. Gadolinium agents are considered to have no nephrotoxicity at
approved dosages. However patients with end stage renal disease have the
potential to develop nephrogenic systemic fibrosis, thus those patients are
excluded from this study as well (8).
If an anaphylactic reaction does occur, though chances are exceedingly rare,
the radiology personnel in each participating center is trained to respond to
life-threatening reactions. The principles of advanced cardiac life support
will be followed (e.g. stabilisation of airway, cardiac function, blood
pressure and administration of epinephrine).

A potential benefit is the development of a treatment algorithm. This may lead
to refined treatments or confirm that the initial chosen treatment is still the
best option for the patients.