Keywords
18F-DCFPyL - avascular necrosis - positron-emission tomography/computed tomography -
prostate cancer - prostate-specific membrane antigen
Introduction
Prostate cancer (PCa) is the most common noncutaneous malignancy in men. Over the
past more than three decades, computed tomography (CT) and99m Tc-methylene diphosphonate
bone scan have remained the mainstays for imaging PCa. These imaging modalities, however,
offer only limited sensitivity and specificity for detecting sites of disease.[1] In recent years, molecular imaging with small-molecule radiotracers has gained traction
as a means to more reliably image men with PCa. Perhaps, the most promising target
for PCa molecular imaging is the transmembrane Type II glycoprotein prostate-specific
membrane antigen (PSMA), which is highly expressed on the vast majority of PCa tumors.[2] Indeed, a number of clinical studies have shown promising results for PSMA-targeted,
small-molecule, positron-emission tomography (PET) agents in the initial staging,
and follow-up of men with PCa.[3],[4]
Despite the high sensitivity and specificity of PSMA-targeted agents for imaging PCa,
a number of interpretive pitfalls have been described which have the potential to
confuse radiologists and treating physicians alike.[5] For instance, a number of benign bone lesions have previously been described that
might mistakenly be interpreted as evidence of M1b disease.[5]
In this manuscript, we describe the case of a patient with recurrent PCa who underwent
a restaging PET/CT examination with the PSMA-targeted small molecule 18F-DCFPyL and was found to have uptake in an area of avascular necrosis (AVN) of the
femoral head. This case adds to the growing body of the literature on interpretive
pitfalls with PSMA-targeted imaging.
Case Report
A 74-year-old man with a history of Gleason 3 + 3 = 6 PCa treated with external beam
radiation therapy subsequently developed biochemical failure with a most recent serum
prostate-specific antigen (PSA) level of 40.9 ng/mL. Rebiopsy of the treated gland
demonstrated recurrent PCa (Gleason score 4 + 4 = 8). Given the high PSA and associated
suspicion for metastatic disease, the patient underwent restaging with conventional
imaging as well as with PSMA-targeted 18F-DCFPyL-PET/CT.
The conventional imaging staging evaluation was reported to be unremarkable except
for the appearance of increased uptake of99m Tc-methylene diphosphonate in the region
of the left femoral head, a new finding that had not been present on a bone scan performed
6 months previously [Figure 1]. Although this would be a typical site for degenerative change and would normally
be viewed with low suspicion, the new appearance of this lesion in a patient with
known active PCa raised concern that this may be a metastatic lesion.
Figure 1 (a) Planar, whole-body 99mTc-methylene diphosphonate bone scan image from 6 months before 18F-DCFPyL-positron-emission tomography/computed tomography imaging demonstrates multifocal
degenerative changes, but no suspicious uptake that would suggest osseous metastatic
disease. (b) Planar, whole-body 99mTc-methylene diphosphonate bone scan image obtained contemporaneously with the 18F-DCFPyL-positron-emission tomography/computed tomography demonstrates new uptake
in the left hip (red arrow), raising the possibility of a new metastatic bone lesion
at this site
Particular attention was paid to the left femoral head during interpretation of the
18F-DCFPyL-PET/CT. This area was noted to have mild, but definitive and focal, radiotracer
uptake (lean body mass corrected standardized uptake value 1.3) in an area corresponding
to the new bone scan finding [Figure 2]. However, the CT morphology of the bone lesion fusing to the 18F-DCFPyL uptake was not consistent with metastatic PCa. This location, as well as
the dense peripheral sclerosis with less sclerosis centrally, was most compatible
with AVN of the left femoral head. Although the degree of uptake in this case is subtle,
it is important for any interpreter of PSMA-targeted PET scans to be aware of such
pitfalls, as this patient could have been misclassified as having M1b disease instead
of M0, leading to very different therapeutic options.
Figure 2 (a) Axial computed tomography. (b) Axial 18F-DCFPyL-positron-emission tomography. (c) Axial 18F-DCFPyL-positron-emission tomography/computed tomography images through the pelvis
shows a focus of uptake in the left femoral head fusing to sclerotic bone changes
(red arrows). Excreted radiotracer in the left ureter and bladder is present; the
patient's biopsy-proven locally recurrent disease is not visible on these images but
was intensely radiotracer-avid. (d) Coronal computed tomography. (e) coronal 18F-DCFPyL positron-emission tomography. (f) Coronal 18F-DCFPyL positron-emission tomography/computed tomography recapitulates the findings
on the axial images, with more clear delineation that the sclerotic changes are along
the articular surface of the left femoral head. These findings confirmed that the
newly appearing uptake on the bone scan shown in Figure 1b was most consistent with
left femoral head avascular necrosis
In the parlance of the recently-introduced PSMA reporting and data system (PSMA-RADS)
version 1.0, uptake in AVN would be considered a PSMA-RADS-1B lesion, i.e., radiotracer
uptake lesion that is definitively benign on anatomic imaging.[6],[7]
Discussion
PSMA-targeted radiotracers have generally demonstrated better diagnostic performance
than other molecular imaging agents in detecting sites of PCa, even at low PSA levels.[6] However, as a consequence of the growing clinical experience with PSMA-targeted
radiotracers, a number of potential interpretive pitfalls have been described in the
literature.[8] Findings from multiple groups suggest increased PSMA-targeted radiotracer uptake
can occur in a variety of malignant and benign entities.[5],[9] Moreover, careful comparison to conventional imaging such as CT and bone scintigraphy,
and the occasional use of other advanced modalities such as magnetic resonance imaging,
may be necessary to arrive at a correct diagnosis.
AVN of the femoral head may lead to progressive destruction of the hip joint. Incidence
in the United States is estimated at 20,000–30,000 new AVN cases/year and most frequently
affects patients in the third-to-fifth decades of life. Risk factors include alcohol
consumption, trauma, corticosteroid therapy, coagulation abnormalities, and radiation
therapy (a possible cause in this patient's case); however, its pathobiology has not
been completely elucidated.[8],[10] Given that patients with PCa may be exposed to multiple risk factors for femoral
head AVN, cross-sectional imaging in these patients should be carefully scrutinized
for findings compatible with AVN and abnormal radiotracer uptake with either bone-seeking
or tumor-targeted agents should not be erroneously interpreted.
While it is a limitation of this report that a pathologic diagnosis of the left femoral
head lesion is not available, the findings on anatomic cross-sectional imaging are
essentially pathognomonic for AVN. In the future, we plan to assess the frequency
with which these findings are present on PSMA-targeted PET scans and determine the
range of uptake levels that can be seen.
Conclusion
Uptake of PSMA-targeted radiotracers can be observed in the femoral head AVN and interpreting
imaging specialists and referring clinicians should be aware of this imaging pitfall.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms.
In the form the patient(s) has/have given his/her/their consent for his/her/their
images and other clinical information to be reported in the journal. The patients
understand that their names and initials will not be published and due efforts will
be made to conceal their identity, but anonymity cannot be guaranteed.
