Page 4 - Microsoft Word - 6942_BPR_Epub

CUAJ – CUA Best Practice Report Guo et al
BPL: AML management

in them being usually diagnosed as either RCCs or indeterminate/fat invisible AMLs. 17,22 However, it
must be noted that there remain rare case reports where these lesions were initially diagnosed as classic
23
AMLs and only differentiated on pathology. Unfortunately, given the lack of evidence, it is difficult to
determine the incidence of such rare misclassifications. EAMLs must remain a consideration in the
evaluation of fat invisible lesions.
As there is no reliable way of imaging to differentiate fat invisible AMLs from atypical
appearing RCCs or EAMLs, they may be managed as indeterminate solid renal masses and may require
biopsy for diagnosis. A recent metanalysis on renal mass biopsies of 57 studies and 5228 patients from
Marconi et al. found an overall accuracy of 92% and only three significant (Clavian grade 2 or greater)
24
complications. They also found that core biopsies had a high sensitivity and specificity compared to
fine needle aspiration, 99.1% and 99.7%, compared to 93.2% and 89.8%, respectively. There is also
evidence that FNA may be particularly challenging in diagnosing AMLs. Zhou et al. reviewed the FNA
biopsies of 33 surgically diagnosed AMLs and found that only 49% of them were diagnosed correctly,
25
with the remainder being non-diagnostic or described as RCCs.
For lesions that are diagnosed as EAMLs or remain undifferentiated after biopsy, surgical
resection is recommended, regardless of lesions size. After resection of pathologically confirmed
EAMLs, there is no evidence for adjuvant therapy and observation is recommended. There is no data or
evidence to suggest a follow-up schedule. However, applying the RCC follow-up guidelines may be
26
reasonable. In cases of metastatic EAMLs, again there is little evidence to guide local or systemic
therapy but there are case reports of response to doxorubicin and everolimus. 27,28

Natural history
The natural history of AMLs has been controversial and has played a significant role in treatment
decision making. Our review of contemporary reviews of AMLs on active surveillance identified nine
articles with 1137 patients. It is important to note that these are all retrospective reviews of patients
selected to be on active surveillance and likely represent a favourable cohort. We found that over the
average follow up period of 37 months, 92% of AMLS observed were asymptomatic. The vast majority
of AMLs remained stable in size, and only 9% of these lesions grew with an average growth rate of
0.4mm/year. The hemorrhage rate was also quite low, at 3%. No lesions that were diagnosed to be
malignant during follow up.
Due to the influence of Oesterling’s original paper, of the four articles that did differentiate
outcomes by the size of the lesion, three used the 4cm cut-off. When stratified by size, we found that
lesions >4cm appeared to be at a higher risk to be symptomatic (34% vs 6%), grow (25% vs 2%) and
hemorrhage (16% vs 1%). Examining individual articles that compared AMLs by size reveal significant
heterogeneity of results. Maclean et al found that lesions >4cm grew significantly faster than those
<2cm (OR of 13.3 and p=0.02) while Bhatt et al found no difference (0.17mm/year vs 0.2mm/year,
p=0.86). 5,29 Ouzaid et al. found that tumour size was significant as an independent predictor of
discontinuation of AS for any reason (HR of 11.2, p=0.001) while Yamakato found that size was not a
significant independent predictor of hemorrhage (p=0.07). 30,31 Based on our systematic review presented