c-KIT oncogene expression in PRKAR1A-mutant adrenal cortex

Abstract

Protein kinase A (PKA) regulatory subunit type 1A (PRKAR1A) defects lead to primary pigmented nodular adrenocortical disease (PPNAD). The KIT protooncogene (c-KIT) is not expressed in normal adrenal cortex (AC). In this study, we investigated the expression of c-KIT and its ligand, stem cell factor (SCF), in PPNAD and other cortisol-producing tumors of the adrenal cortex. mRNA and protein expression, by qRT-PCR, immunohistochemistry (IHC), and immunoblotting (IB), respectively, were studied. We then tested c-KIT and SCF responses to PRKAR1A introduction and PKA stimulation in adrenocortical cell lines CAR47 and H295R which were also treated with the KIT inhibitor, imatinib mesylate (IM). Then, mice xenografted with H295R cells were treated with IM. There was increased c-KIT mRNA expression in PPNAD; IHC showed KIT and SCF immunoreactivity within certain nodular areas in PPNAD. IB data was consistent with IHC and mRNA data. PRKAR1A-deficient CAR47 cells expressed c-KIT; this was enhanced by forskolin and lowered by PRKAR1A reintroduction. Knockdown of PKA’s catalytic subunit (PRKACA) by siRNA reduced c-KIT levels. Treatment of the CAR47 cells with IM resulted in reduced cell viability, growth arrest, and apoptosis. Treatment with IM of mice xenografted with H295 cells inhibited further tumor growth. We conclude that c-KIT is expressed in PPNAD, an expression that appears to be dependent on PRKAR1A and/or PKA activity. In a human adrenocortical cell line and its xenografts in mice, c-KIT inhibition decreased growth, suggesting that c-KIT inhibitors may be a reasonable alternative therapy to be tested in PPNAD, when other treatments are not optimal.

INTRODUCTION

Genetic defects in protein kinase A (PKA) regulatory subunit type 1A (PRKAR1A) lead to increased cyclic adenosine mono-phosphate (cAMP) signaling and primary pigmented nodular adrenocortical disease (PPNAD) associated with corticotropin-independent Cushing syndrome (Kirschner et al., 2000, Horvath et al., 2010). PPNAD is a rare form of bilateral adrenocortical hyperplasia that is genetically distinct form other forms of cortisol-producing pituitary tumors, such as primary macronodular adrenal hyperplasia (PMAH), cortisol-producing adrenal adenomas (CPA), and isolated micronodular adrenocortical disease (Hannah-Shmouni and Stratakis, 2020). PPNAD is characterized by nodules that are less than 1 cm-large, often microscopic, and that produce cortisol in a corticotropin-independent manner, unlike the surrounding adrenal cortex which is frequently atrophic; these nodules may be detected by the cellular accumulation of pigment (i.e. lipofuscin) and using specific immunostaining markers such as synaptophysin (Stratakis et al., 1999).

The KIT protooncogene (c-KIT, CD117), encoded by the KIT gene, is not expressed normally in adrenocortical cells (Zhang et al., 2003) except in adrenal cortex-residing mast cells (Duparc et al., 2015). We recently showed that in PPNAD and macronodular hyperplasia, there is increased serotonin signaling (Bram et al., 2016, Le Mestre et al., 2019), and our previous transcriptomic studies had suggested increased c-KIT expression in PPNAD and macronodular hypeprlasias compared to normal adrenoglandular tissue total RNA (Almeida et al., 2011, Horvath et al., 2006b). However, we also showed that increased serotonin signaling was not due to higher number of mast cells in PPNAD and related adrenocortical lesions (Bram et al., 2016), although mast cells were not searched for in macronodular hypeprlasia (Le Mestre et al., 2019).

In a previous study from our laboratory, the expression of stem cell factor (SCF) that is the ligand for the KIT tyrosine receptor and is encoded by the KITLG gene, was found to be upregulated in the presence of increased cAMP signaling in testicular germ cell tumors (Azevedo et al., 2013). Additional in vitro studies were performed in a testicular germ cell cancer cell line (Azevedo et al., 2013), as KIT and KITL are known to be involved in both germ cell differentiation and Leydig cell steroidogenesis (Gu et al., 2009, Rothschild et al., 2003). Despite the similarities between adrenocortical and testicular cells both in terms of embryologic origin and regulation of steroidogenesis, there has been essentially no work on potential interplay between c-KIT and cAMP signaling, in light of the fact that repeated attempts to identify KIT-expressing cells in the adrenal cortex (other than mast cells) have failed (Duparc et al., 2015, Zhang et al., 2003).

In the present study, we investigated KIT and SCF expression in PPNAD caused by PRKAR1A mutations that lead to increased cAMP signaling (Horvath et al., 2010, Kirschner et al., 2000), and compared their expression with that in normal adrenal cortex and other adrenocortical lesions that cause corticotropin-independent Cushing syndrome, including macronodular hypeprlasia. The data support the presence of KIT and SCF-expressing clusters of cells in PPNAD and potential regulation of c-KIT and SCF expression by PRKAR1A and/or cAMP signaling. Furthermore, we showed that c-KIT inhibition may lead to decreased growth and proliferation of adrenocortical tumor cells.

These data are important as they suggest that in at least some forms of adrenocortical neoplasms, those that depend on increased cAMP signaling, c-KIT (and possibly other tyrosine kinase receptor) inhibitors may be useful therapeutic alternatives, when other treatment options are not available, or have an unfavorable safety profile or long-term outcomes.

MATERIALS & METHODS

Clinical and DNA, RNA studies

All patients were recruited under protocols 95CH0059 and 00CH160 approved by National Institutes of Health (NIH) intramural institutional review boards for the study of PPNAD and other adrenocortical tumors, including macronodular hyperplasia and cortisol-producing adenomas. The patient’s parents’ informed consent was obtained and the patient’s assent as appropriate. Tumor samples were obtained from patients at surgery and DNA was extracted from peripheral blood and fresh-frozen tissues, as previously described (Azevedo et al., 2013, Hannah-Shmouni and Stratakis, 2020, Horvath et al., 2010). All patients and/or samples have been screened for germline mutations of PRKAR1A, PDE11A, PRKACA, and related genes, as part of older studies (Angelousi et al., 2017, Beuschlein et al., 2014, Horvath et al., 2006a).

We studied 11 samples with PPNAD from patients with known germline PRKAR1A-inactivating mutations (Suppl. Table S1) and 6 from patients with macronodular hyperplasia and another 6 from patients with cortisol-producing adenomas without any known somatic or germline genetic defects. DNA was extracted using DNA extraction kit (Qiagen, Inc. Venlo, NL) and total RNA was isolated using Trizol (Invitrogen, Waltham, MA) according to manufacturer’s instructions. Proteins were extracted from the tissue samples using the TPER Reagent (Pierce Biotechnology, Waltham, MA), as per the manufacturer’s recommendations. A sample from a gastrointestinal stromal tumor (GIST) with a known KIT mutation was used a control (courtesy of Dr. Electron Kebebew, National Cancer Institute, NIH, Bethesda, MD, USA).

Cell lines

The H295R cell line obtained from American Type Culture Collection (ATCC), Manassas, VA (https://www.atcc.org/; NCI-H295R [H295R] [ATCC® CRL-2128]) was cultured in Dulbecco’s Modified Eagle Medium (DMEM) F12 medium (Invitrogen, Waltham, MA) with 2% Nu serum (BD Biosciences) supplemented with 1:100 dilution of ITS Universal Cell Culture Supplement Premix (BD Biosciences, Jan Jose CA) and 5mLs of antibiotics (Invitrogen, Waltham, MA). CAR 47, a cell line previously established in our laboratory (Nesterova et al., 2008), from a patient with PPNAD was maintained and cultured in DMEM medium supplemented with 20% serum and antibiotics as described previously (Nesterova et al., 2008). The cell line is also available through ATCC (CAR47 [ATCC® CRL-3235]).

Cortisol measurements

Cortisol levels from the supernatant of H295R cells that were treated by imatinib or vehicle were measured at 24 and 48h, using enzyme-linked immunosorbent assay (ELISA) kit (ADI-900–071, Enzo Life Sciences, Inc. Farmingdale, NY), as per the manufacturer’s instructions.

KIT sequencing

KIT exons 8, 9, 10, 11, 13 and 17 were screened for mutations by direct sequencing of PCR products (using the Sanger method; the sequence of the primers is provided in Suppl. Table S2) amplified from tumor samples excised from PPNAD nodules. The amplified PCR products were run on a gel to check for purity and the bands were gel excised and DNA was extracted from gel slices using Gel Extraction Kit (Qiagen, Inc. Venlo, NL). Sequencing was performed and the sequences were analyzed using Vector NTI Advance™ 11 software https://www.thermofisher.com/us/en/home/life-science/cloning/vector-nti-software.html (Invitrogen by Life Technologies, Carlsbad, CA, USA).

Quantitative real time polymerase chain reaction (qPCR) analyses

Reverse transcription was performed using iScript reverse transcription PCR kit (Bio-Rad, Inc., Hercules, CA) with a minimum of 250–1ug of RNA in a final volume of 20ul according to the instructions of the manufacturer. Real time PCR was performed using Power SYBR green PCR Master Mix (Applied Biosystems, Foster City, CA) in a 25 uL reaction containing 1uL of the reverse transcription product, 1X Power SYBR green PCR Master Mix (P/N 4368577; Applied Biosystems, Foster City, CA), 0.5 μmol/L forward primer and 0.5 μmol/L reverse primer on a 96 well PCR plate at 95⁰C for 10 min followed by 40 cycles of 95⁰C for 15 s and 60⁰C for 1 min (Bio-Rad, Inc., Hercules, CA).

Each sample was analyzed in triplicate. qPCR was used to quantify the expression of KIT and SCF. The expression of GAPDH was used as an endogenous reference control. The amount of target is normalized to an endogenous reference and relative to a calibrator (cDNA prepared from three commercially available normal adrenal total RNA pools obtained from Clontech, Mountain View, CA; BioChain, Hayward, CA; and Ambion, Austin, TX), is given by 2-ΔΔCt comparative Ct method as described previously (Patterson et al., 2011).

Immunoblotting

For immunoblotting/Western analysis, cells were collected, washed in phosphate-buffered saline (PBS), and lysed in M-PER protein extraction reagent (Pierce Biotechnology, Rockford, IL) unless otherwise stated. Lysates from primary tumors were made by homogenizing the tissues in T-PER protein extraction reagent (Pierce Biotechnology, Rockford, IL). After centrifugation at 14,000 rpm for 15 min equal amounts of total protein from the cell and tumor lysates were resolved on 10% denaturing polyacrylamide gels (Invitrogen) and transferred to nitrocellulose membranes (Pall Biotech, Port Washington, NY). After blocking with tris-buffered saline (TBS) and Polysorbate 20 (TBST)-containing 5% non-fat milk (Blotto, Bio-Rad, Inc., Hercules, CA) for 1h, the membranes were incubated with indicated primary antibodies at 4⁰C overnight and then washed three times with TBST. All membranes were probed with horseradish peroxidase conjugated secondary antibodies (Jackson Immune research laboratories, West Grove, PA) for 1h at room temperature and washed three times with TBST. The blots were then developed with Western Lightning reagents (Perkin-Elmer, Inc., Waltham, MA). The following primary antibodies were used. KIT (SC-17806), SCF (SC-13126), from Santa Cruz Biotechnology, Inc. (Dallas, TX) phospho-KIT (Y568+Y570 ab78247) GAPDH (ab9485) and p-AKT (#9271S), P44/42 MAPK (#9101S), AKT (#9272), p44/42 (#9102), caspase-3 (#9662) from Cell Signaling Technology, (Danvers, MA).

Immunohistochemistry (IHC)

The expression of KIT and SCF in tumor sections was examined by IHC that was performed on paraffin embedded tissue sections using rabbit polyclonal (KIT) and mouse monoclonal antibodies (SCF) obtained from Abcam (Cambridge, MA) and Santa Cruz Biotechnology (Dallas, TX) respectively (see above), and were compared with negative controls and surrounding stroma where ever available. All the IHC staining was performed by Histoserve, Inc. (Germantown, MD).

Transfections

For transfections in CAR47 and H295R cells, cells were plated 24h before transfection and were transfected with PRKAR1A plasmids that we have reported elsewhere (Nesterova et al., 2008), using lipofectamine 2000 (Invitrogen, Waltham, MA). Cells were harvested 48h post transfection for protein and RNA preparation, as described above.

Drug treatment and assessment of cell viability and apoptosis assays

The effect of imatinib mesylate on the viability H295R and CAR47 cells was assessed using the WST-1 cell proliferation assay kit from Roche, Basel, CH). Briefly, cells were seeded in 96-well plates at ~2,500 cells per well in their respective growth medium. After 24h, the cells were treated with varying concentrations of imatinib mesylate dissolved in DMSO for 24, 48 and 72h. Controls were treated with DMSO vehicle alone at a concentration to that of drug-treated cells. After 24, 48, and 72h of drug treatment, 20μL of WST-1 reagent was added to each well and the cells were incubated for up to 1~2 h at 37⁰C. After incubation, absorbance was measured at 450 nm using the plate reader. All treatments were evaluated in triplicate in at least three independent experiments.

Xenografts

Female nude mice (nu/nu) were fed on a standard diet and were maintained in regulated light cycle (12h of light, 12h of darkness). About one million cells in 100uL suspension were injected subcutaneously into right flanks of ~six weeks old nude mice. Imatinib mesylate (from LKT Laboratories, St. Paul, MN) was dissolved in 0.2% CMC at concentration of 10 mgs/mL. When the tumors reached ~100 mm3 (~4 weeks after injection), the mice were randomly placed into control and experimental groups (7 mice per group). Group1 mice were gavage orally with carboxymethycellulose (which was used as vehicle) alone and Group 2 with imatinib mesylate, once a day in volumes of 0.1 mL/10 g (mouse body weight, 100mgs/kg) for 7 days. Tumor sizes were measured every other day starting from the day of treatment. Mice were sacrificed on the 8th day and the tumors collected were used for further histological characterization. Histologically, the xenografts were similar to adrenal cancer (data not shown). All experiments were performed according National Institutes of Health Animal Care and Use Committee (ACUC) guide for the care and use of Laboratory animals (NIH Publications No. 8023, revised 1978), under approved protocol ASP# 18–033.

Statistical analysis

All data are presented as the mean +/− SD; P-values were obtained by unpaired t-test or ANOVA carried out using the software GraphPad Prism 6.0 (GraphPad®). We did not use non-parametric testing for any of the comparisons.

RESULTS

Identifying c-KIT over-expression in PPNAD

We analyzed the expression of c-KIT in samples from patients with macronodular hyperplasia (N=6) and cortisol-producing adenoma (N=6) with no known mutations, and from patients with PPNAD (N=11) with known PRKAR1A mutations (Suppl. Table S1).

First, KIT mRNA as detected by qPCR was upregulated in PPNAD samples as a group, although specific samples (3 samples of 11) showed even more expression (Fig. 1A). We then used immunoblotting/Western blotting (Fig. 1B) and IHC (Fig. 1C, D. E); a sample from a KIT-mutant GIST was run also a control, as we have used elsewhere (Azevedo et al., 2013). Western data were consistent with mRNA data: the samples that had the highest c-KIT mRNA also had the highest c-KIT protein as shown in Fig. 1B.

Figure 1.

c-KIT mRNA & protein expression in patient samples by RT-PCR, immunoblotting/Western blotting, and immunohistochemistry

A) Real-time PCR data for c-KIT expression in samples from patients with primary macronodular adrenal hyperplasia (PMAH), cortisol producing adenomas (CPA), and primary pigmented nodular adrenocortical disease (PPNAD) (listed in Suppl. Table S1); the data are normalized against pooled normal adrenal RNA, as described previously (Patterson et al., 2011); B) Western blot data for c-KIT protein expression in PPNAD (panel on the left) against samples with primary macronodular adrenal hyperplasia (panel on the right); a single GIST with a KIT “hot-spot” mutation is included as control in the PPNAD blot. Certain samples (for example, CAR20.14) express high levels of c-KIT; C) and D) Immunohistochemistry using an antibody specific for c-KIT identifies cell clusters that express highly c-KIT, usually small cortical nodules; both the medulla (M) and the rest of the cortex (C) are negative for any staining; E) A sample from a patient with primary macronodular adrenal hyperplasia that shows no immunoreactivity for c-KIT. In all 3 samples the original magnification is 10X and the focused images indicated by the cropped boxes are 40X.

IHC was obtained in all samples (data not shown); representative samples with PPNAD among the 3 highest-expressing are shown in figure 1C and 1D, showing (in magnification) nodules that stained positive for c-KIT. Clearly, not all nodules in each specimen were positive and the surrounding adrenal cortex was negative. Likewise, most cortical staining from other samples was negative; a representative sample is shown in figure 1E. SCF expression was more widespread; clearly, again, PPNAD samples had the highest SCF expression (Suppl. Fig. S1). Following the unexpected detection of c-KIT expression, we sequenced the most commonly KIT-mutated exons (exons 8, 9, 10, 11, 13, and 17) in all samples used in this study and we found no mutations other than known polymorphic variants (data not shown).

PRKAR1A and forskolin regulate c-KIT expression in AC cells; KIT inhibition affects cell viability

We introduced PRKAR1A by transfection in AC cell lines CAR47 that has a known PRKAR1A-inactivating mutation (Nesterova et al., 2008) and H295R. In both cases, the introduction of PRKAR1A led to a decrease in detectable c-KIT by Western (Fig. 2A); on the other hand, forskolin increased c-KIT expression (Fig. 2B). Exposure of CAR47 and H295R cells to increasing doses of the KIT tyrosine receptor inhibitor imatinib mesylate led to decreased cell viability starting at concentration 40 μM (Fig. 2C). We could see that the use of the inhibitor in H295R cells led these cells to express decreasing levels of c-KIT upon increasing doses, which led to largely unaffected levels of the KIT ligand SCF, but affected all downstream indicators of c-KIT activity, such as pAKT and pERK (Fig. 3A). Increasing doses of the inhibitor led these cells to undergo apoptosis (Fig. 3B). The decrease in cell viability was accompanied by decreasing cortisol levels in H295R cells (Fig. 4A). The effect of PRKAR1A and forskolin on H295R cells (Fig. 2A and 2B) was likely mediated by the PKA catalytic subunit, PRKACA, as siRNA to PRKACA led to decreased c-KIT expression in H295R cells (Fig. 4B).

Figure 2.

c-KIT levels and cell viability in CAR47 and H295R cell lines transfected with PRKAR1A gene

A) CAR47 cells show expression of the c-KIT protein that is down-regulated by introduction of a PRKAR1A-expressing plasmid; likewise, H295R cells show expression of the c-KIT protein that is down-regulated by introduction of the PRKAR1A-expressing plasmid; B) Forskolin, leads to up-regulation of c-KIT protein in H295R cells; similar data were seen in CAR47 cells (data not shown); C) Both CAR47 and H295R cells respond with decreased viability to increasing concentrations of imatinib mesylate (IM).

Figure 3.

Western blot evaluation of some markers after use of different doses of imatinib mesylate in H295R cell line

A) H295R cells showed decreased phosphorylation of AKT and ERK in response to increasing concentrations of IM; B) Apoptosis was increased as measured by caspase 3 expression in response to increasing concentrations of IM in H295R cells; CAR47 cells showed similar data (data not shown).

Figure 4.

Cortisol and c-KIT levels in H295R cells and Xenografts of H295R cells in mice

A) Cortisol levels decreased in response to increasing concentrations of IM in H295R cells; B) c-KIT expression was decreased in response to siRNA against the main PKA catalytic subunit PRKACA; C) Xenografts of H295R cells produced tumors in mice that looked histologically like adrenocortical cancer (data not shown); D) In each animal treated with IM, the tumor size stabilized after 3 days of treatment (at a dose of IM 100mg/kg); the experiment ended after 7 days of treatment. The graph shows the per-cent response averaged for the two groups (each comprised of 7 animals), one treated with IM and the other with vehicle.

KIT inhibition led to tumor stasis in xenografts of H295R cells

We tested the possibility of using KIT inhibition on an adrenocortical tumor in vivo. We developed a murine adrenal xenograft model using human adrenocortical cancer H295R cells; tumor-bearing animals were treated orally with imatinib mesylate or vehicle alone for 7 days (at a dose of the inhibitor of 100mg/kg). Seven mice were included per treatment group; the effect of the inhibitor on tumor size was then assessed: there was visible tumor size stabilization starting on day 3 of treatment, without further changes by the time the experiment ended on day 7 (Figs. 4C and 4D). There were no histological differences between tumor samples from the two groups of animals (data not shown).

DISCUSSION

The tyrosine kinase receptor c-KIT is not known to be expressed in normal adrenal cortex or in adrenocortical tumors (Matsuda et al., 1993, Zhang et al., 2003). In one study, c-KIT (also known as CD117) immunoreactivity was detected in 1 of 9 (11%) adrenocortical carcinomas and 0 of 13 adrenocortical adenomas (Zhang et al., 2003). On the other hand, we and others have shown that certain cortisol-producing tumors, such as PPNAD, and other adrenocortical hyperplasias leading to corticotropin-independent Cushing syndrome may express markers that are not normally present in adrenal cortex or in other adrenocortical lesions (Stratakis et al., 1999). KIT is highly expressed in mast cells that are the source of serotonin-signaling in the brain and elsewhere (Theoharides et al., 1982), and mast cells are widely present in the subcapsular region of the adrenal and within zona glomerulosa in both humans and mice (Duparc et al., 2015, Kim et al., 1997). However, when we recently showed widespread serotonin signaling in PPNAD and other forms of adrenocortical hyperplasias, we demonstrated that this was not due to increased mast cell infiltration of these lesions (Bram et al., 2016). Although other cells of non-adrenocortical lineage can often infiltrate cortisol-producing lesions, including adenomas (Willenberg et al., 1998), there is no known lesion-specific infiltration from mast or other non-cortical KIT-expressing cells in PPNAD.

There are some data indicating that KIT may be physiologically expressed in some adrenocortical cell progenitors in mice (Bayne et al., 2008). Interestingly, in a model of mast cell-deficient mice, global adrenal expression of kit mRNA was not different from that in control mice (Boyer et al., 2017), pointing to the presence of non-mast cell derived c-KIT in the normal mouse adrenal gland. Indeed, the present study shows that occasionally in PPNAD there are few, c-KIT-expressing cell clusters; this was apparent also my mRNA and protein studies (Fig. 1). We also could show that two adrenal cortex-derived cell lines, one from a patient with PPNAD (CAR47) and another from adrenocortical cancer (H295R) both expressed c-KIT which could be down-regulated by increased PRKAR1A and up-regulated by increased PKA activity, or down-regulated by siRNA against the main catalytic subunit of PKA (PRKACA) that is regulated by PRKAR1A (Figs. 2 and 4B). When we tested the well-known c-KIT-inhibitor imatinib mesylate, both cell lines responded with decreased viability and increased apoptosis; in the case of H295R cells, tumors derived from xenografts failed to grow further after the administration of the c-KIT inhibitor (Figs. 2, 3, and 4). We should also note that we did not see any mast cells in the xenograft-derived tumors (data not shown), as it would have been possible that the inhibition of tumor growth by the inhibitor-treated cells may have been due to inhibition of mast cell infiltration of the xenografts.

It is worth emphasizing that in our study, too, there was considerable paucity of staining for c-KIT in most adrenal cortex. The patchy presence of c-KIT-positive clusters of cells suggest that certain cells in PPNAD may be subject to somatic genetic or epigenetic changes that led to aberrant KIT expression. Our sequencing did not show any known KIT mutations, but we did not sequence the full KIT gene, nor did we look at changes in methylation or other regulatory variations of the KIT genetic locus. Finally, it is worth noting that the increased expression of SCF (Suppl. Fig. S1) did not lead to an increased presence of mast cells in PPNAD; it is possible that PRKAR1A deficiency albeit associated with increased SCF, may have prevented mast cells from infiltrating the diseased adrenal cortex.

The present study is a small one that may be considered a pilot investigation. However, the data are encouraging for the potential use of c-KIT-inhibitors in adrenocortical neoplasms. Although the first uses of tyrosine-kinase inhibitors in adrenal cancer did not lead to any appreciable benefit (Berruti et al., 2012), recent data are more encouraging (Kroiss et al., 2020, O’Sullivan et al., 2014). However, none of the recently used tyrosine-kinase inhibitors in adrenal cancer inhibited c-KIT (Berruti et al., 2012, Kroiss et al., 2020, O’Sullivan et al., 2014), and it would be great if the data of the present study could be confirmed in a larger study, one that will employ additional cancer samples and lines.

Beyond using tyrosine-kinase inhibitors in cancer, the current data suggest that low-dose imatinib mesylate may also be tried for decreasing cortisol production in patients with PNNAD that are not eligible for surgery. Indeed, a significant number of patients with PPNAD have either mild cortisol production or are burdened by other tumors caused by PRKAR1A mutations; in both cases, clinicians are left with very few reliable options for treatment of the hypercortisolemia. The present study suggests that a proof-of-principle, randomized, placebo-blinded trial of imatinib mesylate at doses lower than those used for routine chemotherapy in patients with PPNAD may be worthwhile.

We conclude that in PPNAD caused by PRKAR1A mutations there appears to be selected c-KIT/CD117-expressing cell populations (the c-KIT ligand stem cell factor or SCF is also expressed in a wider set of cells). Preliminary data show that c-KIT-inhibition could confer a therapeutic advantage decreasing cortisol secretion (at least in H295R cells) and cell viability (in both PRKAR1A-mutant CAR47 and H295R cells), as well as inhibiting the growth of adrenocortical cancer xenografts in mice.