Adrenocortical carcinoma with circulating tumor DNA analysis at post-operative recurrence: a case report with review of literature

Abstract

Recently, the usefulness of circulating tumor DNA (ctDNA) analysis in various malignancies has been reported. However, reports on ctDNA analysis in adrenocortical carcinoma (ACC) are few. Therefore, this study aimed to examine the detectability of genetic mutations in ctDNA and the association between ctDNA allelic ratio and disease progression in a patient with post-operative recurrence of ACC. A 77-year-old woman presented with a 5.4 cm left adrenal mass, which was clinically diagnosed as subclinical cortisol-producing ACC on close examination. She underwent left adrenalectomy and was diagnosed with stage II (T2N0M0) ACC. Post-operatively, adjuvant chemotherapy with mitotane was commenced because of histologically high-grade ACC. However, 17 months post-operatively, she had a local recurrence at the left adrenalectomy site. FoundationOne^® CDx Cancer Genome Profile showed CTNNB1 G34A mutation in the resected adrenal tumor. She had heart failure and interstitial pneumonia and was treated with radiotherapy for local recurrence. Subsequently, lung and liver metastasis appeared post-operatively at 21 and 23 months, respectively. Serum dehydroepiandrosterone sulfate and computed tomography findings at 27 months post-operatively showed disease progression. We collected the peripheral blood at 23 and 27 months post-operatively and analyzed 18 genes associated with adrenal disease in plasma cell-free DNA and the resected adrenal tumor using a next-generation sequencer. At both time-points, CTNNB1 mutations consistent with the primary tumor were observed in ctDNA, with the allelic ratio increasing over time from 8% to 27%. In conclusion, monitoring the ctDNA allelic ratio may be useful for evaluating disease progression in advanced ACC.

Introduction

Adrenocortical carcinoma (ACC) is rare, with an annual incidence of 0.5–2 cases per million population [1]. ACC is an aggressive malignancy with a poor prognosis based on the low 5-year survival rate below 30% in advanced stages [2]. Therefore, early diagnostic methods and effective treatments for ACC are expected.

Recently, circulating tumor DNA (ctDNA) has shown usefulness as a somatic mutation marker (liquid biopsy) for various malignancies. CtDNA is fragmented DNA released directly by tumor cells into the bloodstream among the circulating cell-free DNA. In a cross-carcinoma report on ctDNA analysis, somatic mutations were detected using ctDNA in 85% of 21,807 patients with cancer [3]. CtDNA may be useful in almost every stage of cancer management, including diagnosis, minimally invasive molecular profiling, treatment monitoring, residual disease detection, and resistance mutation identification [4]. However, studies documenting the applicability of ctDNA to ACC are few [5, 6]. The sample sizes in such studies were limited by the rarity of ACC and suboptimal ctDNA concentration for analysis.

Previously, we reported a case of ACC diagnosed 9 years after the discovery of adrenal incidentaloma [7]. In this study, we aimed to evaluate the detectability of genetic mutations in ctDNA using next-generation sequencing and the association between ctDNA allelic ratio and disease progression in the same patient with post-operative recurrence of ACC.

Materials and Methods

Case description

A 5.4 cm left-sided subclinical cortisol-producing ACC was clinically diagnosed in a 77-year-old woman with no signs of lymph node or distant metastases as determined using computed tomography (CT) (Fig. 1A), magnetic resonance imaging, and fluorodeoxyglucose positron emission tomography-CT scan. Retrospectively, the left adrenal tumor increased from 3.0 cm to 5.4 cm on CT in 1 year. She underwent laparoscopic left adrenalectomy. The resected adrenal tumor measured 7.0 × 4.0 × 2.5 cm and weighed 80 g (Fig. 2A). The histopathology report described a primary ACC (Weiss score, 4), with a mitotic rate of 25 mitoses/50 HPF, <25% clear cell component, tumor necrosis, and venous invasion (Fig. 2B, C). Immunohistochemically, the Ki67 labeling index was 20% in hot spots and β-catenin staining showed accumulation in the nucleus of the tumor (Fig. 2D, E). The ACC was diagnosed as Stage II (T2N0M0) based on the European Network for the Study of Adrenal Tumors (ENSAT) staging system [2].

Fig. 1  Clinical course

(A) She had a 5.4 cm left-sided adrenal tumor pre-operatively.

(B, C) No recurrence or metastasis of ACC was observed at 6 and 12 months post-operatively. (D) She had a local recurrent lesion at the left adrenalectomy site at 17 months post-operatively. (E, F) The local recurrent lesions were reduced using radiotherapy, but new lung metastases appeared at 19 months post-operatively.

(G–I) She was hospitalized for heart failure at 23 months post-operatively. The local recurrence site increased, and new liver metastases appeared.

(J–L) The local recurrence site and liver metastases were enlarged, the lung metastases were stable, and a new peritoneal dissemination was observed at 27 months post-operatively. Serum DHEA-S levels increased sharply from 23 to 27 months post-operatively.

Fig. 2  Histopathological examination and immunohistochemical analysis of the resected adrenal mass

(A) The resected adrenal tumor measured 7.0 × 4.0 × 2.5 cm and weighed 80 g.

(B, C) The hematoxylin-eosin staining led to a diagnosis of ACC (Weiss score, 4) with high mitotic rate (C); <25% clear cell component, tumor necrosis (indicated by arrow, B), and venous invasion (magnification B [×100], C [×200]).

(D) Ki67-positive cells exhibited 20% hot spots (magnification ×200).

(E) β-catenin immunohistochemical staining showed accumulation in the nucleus of the tumor (magnification ×400).

Post-operatively, adjuvant chemotherapy with mitotane, a dichloro-diphenyl-trichloro-ethane derivative, was started based on the ENSAT practice algorithm for resectable locally advanced ACC because of histologically high-grade ACC (Ki67 labeling index >10%) [2]. Mitotane was continued in small doses (1.5–3.0 mg/day) because of the side effects of gastrointestinal symptoms and fatigue. The patient was doing well with no recurrence or metastasis of ACC (Fig. 1B, C), but a local recurrent lesion appeared at the left adrenalectomy site at 17 months post-operatively (Fig. 1D). FoundationOne^® CDx Cancer Genome Profile, using the formalin-fixed paraffin-embedded (FFPE) samples, was performed as a companion diagnosis to multiple molecular targeted drugs [8]. It showed CTNNB1 G34A mutation in the resected adrenal tumor, with an allelic ratio of 60%. She had heart failure owing to severe mitral regurgitation and interstitial pneumonia associated with rheumatoid arthritis. Therefore, radiation therapy was used for local recurrence instead of reoperation or chemotherapy at 19 months post-operatively. The local recurrent lesions were reduced via radiotherapy (Fig. 1E), but lung metastasis appeared at 21 months post-operatively (Fig. 1F); hence, radiotherapy was added for the lung lesions at 22 months post-operatively. She was hospitalized for heart failure at 23 months post-operatively (Fig. 1H). The local recurrence site increased (Fig. 1G), and new liver metastases appeared (Fig. 1I). Serum dehydroepiandrosterone sulfate (DHEA-S) rose sharply from 24 μg/dL to 255 μg/dL at 23 and 27 months post-operatively, respectively. Additionally, CT findings showed increased local recurrent lesions (Fig. 1J), stable lung metastasis (Fig. 1K), increased liver metastasis, and new peritoneal dissemination (Fig. 1L). Her disease condition subsequently worsened, and she passed away at 31 months post-operatively.

Genetic analysis

In the present study, ctDNA analysis was performed on peripheral blood samples drawn during routine medical care at 23 and 27 months post-operatively. The blood samples were collected and placed in tubes with EDTA-2K as an anticoagulant. Cell-free DNA was extracted from 5 mL of the plasma. Genetic analysis was also performed using tissue samples from the resected adrenal tumor. A portion of the resected adrenal tumor was collected at the time of surgery for research purposes, and DNA was isolated from the frozen tumor tissue. It is unknown whether the tumor sites we sampled and the sites the pathologist submitted to Foundation One CDx are in proximity. DNA was extracted using the DNeasy Blood and Tissue Kit manufactured by QIAGEN (Hilden, Germany). The extracted DNA was subjected to library preparation based on the protocol provided with the KAPA HyperPlus Kit (KAPA Biosystems, Wilmington, MA, USA). Eighteen adrenal disease-associated genes were tested: KCNJ5, ATP1A1, ATP2B3, CACNA1D, CNCNA1H, PRKACA, GNAS, ARMC5, CTNNB1, PRKAR1A, MEN1, NF1, RET, VHL, APC, CYP11B1, HSD3B2, and PDE1A (Supplementary Table 1). For target gene sequencing, the xGen next-generation sequencing hybridization capture system, custom designed, was used for enrichment, and sequencing on the NextSeq 2000 system (Illumina, San Diego, CA, USA) or MiSeq system (Illumina, San Diego, CA, USA) followed. In the analysis of sequence data, the best practices of GATK (https://gatk.broadinstitute.org/) were utilized for calling variants in germline and somatic mutations. ANNOVAR (http://annovar.openbioinformatics.org/) was used for variant annotation. Additionally, the gnomAD database (https://gnomad.broadinstitute.org/), was used to determine the population frequencies of variants. Somatic mutations were determined using zero population frequency of germline variants as an indicator. Furthermore, the pathogenicity of variants was assessed using ClinVar (https://www.ncbi.nlm.nih.gov/clinvar/). The variants were categorized into four based on their clinical significance: Pathogenic, likely pathogenic, uncertain significance, and others. Overall, this comprehensive variant analysis pipeline allowed for identifying and annotating genetic variants with potential clinical significance, providing valuable insights into the underlying genetic causes of diseases.

This study was conducted in accordance with the Declaration of Helsinki, in consideration of the ethics, human rights, and protection of personal information of the participants/patients. The Ethics Review Committee of Kanazawa University reviewed and approved this study protocol (no. 2022-013). Written informed consent was obtained from the proband who participated in the study.

The genetic variant of the DNA extracted from the frozen specimens was CTNNB1, as in Foundation One CDx, with an allelic ratio of 1% (Table 1). DNA was isolated from plasma at 23 and 27 months post-operatively, yielding 156 and 97 ng/mL plasma, respectively. The percentages of cell-free DNA at 23 and 27 months post-operatively were 88% and 96%, respectively. At 23 and 27 months post-operatively, CTNNB1 mutation (c.101G>C: p.G34A) consistent with the primary tumor was identified in ctDNA. No mutations were found in 17 adrenal disease-associated genes other than the CTNNB1 gene. The allelic ratio of CTNNB1 mutation in ctDNA increased over time from 8% to 27%.

Table 1 Monitoring of the allelic ratio of ctDNA

	DNA concentration (ng/mL)	Mutation	DNA quality (%)	Allelic number	NGS depth	NGS allelic ratio (%)
Primary tumor	6,580	CTNNB1:c.101G>C:p.G34A	—	2	201	1
ctDNA
23 months post-operative	156	CTNNB1:c.101G>C:p.G34A	88	159	2,035	8
27 months post-operative	97	CTNNB1:c.101G>C:p.G34A	96	284	1,039	27

Abbreviations: ctDNA, circulating tumor DNA; NGS, next-generation sequencing.

Discussion

We analyzed ctDNA collected from the peripheral blood of patients with post-operative recurrent ACC. CTNNB1 mutations in ctDNA were consistent with those of the resected adrenal tumor. Furthermore, the ctDNA allelic ratio increased with disease progression. This finding is rare because reports on ctDNA analysis of ACC are few.

Cancer genome profiling was performed on 344 cases with 28 cancer types using tissue and blood specimens, and the concordance rate of genetic mutations detected in tissue and blood samples was 75% (932/1,249) [9]. However, there are only two reports on the concordance rate of genetic variants between tissue and blood samples in ctDNA analysis for ACC.

First, Creemers et al. evaluated ctDNA in six patients with ACC [5]. In three of the six cases, no genetic mutations were detected in the primary tumors. In two of the three cases that had genetic mutations in the primary tumors, the mutations were detected in the tissue samples only. Therefore, tumor-specific mutations were only detected in ctDNA of one of the three cases. In this case, ctDNA allelic ratios were analyzed before and after surgery (case 2, Table 2). The lower cell-free DNA yield and the lower mutation frequencies in the post-operative plasma of the patient could be explained by the lower tumor load because the primary tumor and most metastases were surgically removed. Furthermore, the patient had been treated with mitotane in combination with etoposide, doxorubicin, and cisplatin before the post-operative blood sample was taken.

Table 2 Report on the association between ctDNA allelic ratio and disease progression in ACC

Primary tumor features									cfDNA					Patients feature at the time of cfDNA sampling
Case # [ref]	Sex	Age (years)	Steroid secretion	ENSAT stage	Tumor size (cm)	Weiss score	Ki67 (%)	Mutated genes (allelic ratio)	at the time of cfDNA sampling	DNA concentration (ng/mL of plasma)	Mutations	NGS depth	NGS allelic ratio (%)	Metastases	Disease progression
1 [our case]	F	77	GC	2	5.4	4	20	CTNNB1 (1%)	23mo Post-operative	156	CTNNB1	2,035	8	Lung, liver	Progression
									27mo Post-operative	97	CTNNB1	1,039	27	Lung, liver, peritoneum	Progression
2 [5]	M	57	N/A	4	15	6	50	NRAS (70%)	Pre-operative	123.6	NRAS	4,580	64	Lung, peritoneum, omental, right adrenal	N/A
								TP53 (60%)			TP53	2,566	32
								TERT (28%)			TERT	134	2
									6mo Post-operative	1.99	NRAS	2,408	52	Liver, mesentery, pelvis, subcutaneous, retroperitoneal, leptomeningeal intramuscular	Progression
											TP53	1,127	16
											TERT	128	3
3 [6]	M	44	None	4	8	7	40	TP53 (93%)	21mo Post-operative	17.3	TP53	301,308	11.7	Pancreatic	Progression
								CTNNB1 (92%)			CTNNB1	155,564	16.6
								RB1 (93%)			RB1	117,413	14.3
									27mo Post-operative	N/A	TP53	N/A	N/A	Pancreatic, liver	Progression
											CTNNB1	N/A	N/A
											RB1	N/A	27
4 [6]	F	42	GC	4	15	N/A	80	TP53 (71%)	Pre-operative	422	TP53	147,166	13.8	Lung, liver	N/A
						(lung biopsy)			1mo Post-operative	N/A	TP53	N/A	8.9	Lung, liver	N/A
									2mo Post-operative	N/A	TP53	N/A	31.6	Lung, liver	Progression

Abbreviations: ctDNA, circulating tumor DNA; ACC, adrenocortical carcinoma; cfDNA, cell-free DNA; F, female; M, male; GC, glucocorticoids; N/A, not applicable; mo, month.

Second, Garinet et al. investigated ctDNA in 11 patients with ACC [6]. In three of these cases, no genetic mutations were detected in the tissue or blood samples. In 6 of the 11 cases, genetic mutations were detected in the tissue samples only. Genetic mutations of ctDNA consistent with those of the primary tumors were identified in two cases. In these two cases (cases 3 and 4, Table 2) with positive detection, ctDNA and tumor progression were monitored. In case 3 (Table 2), the ctDNA allelic ratio of RB1 rose from 21 to 27 months post-operatively because new liver metastases appeared. In case 4 (Table 2), the increased ctDNA allelic ratio of TP53, corresponded to worsened pulmonary lesions.

We observed a parallel increase in ctDNA allelic ratio along with tumor growth in recurrent or metastatic lesions of ACC. Cases 2–4 (Table 2) both had massive metastasis and rapid progression. Similarly, our patient also developed rapidly worsening recurrent and metastatic lesions from 23 to 27 months post-operatively. The increase in the allelic ratio of ctDNA was paralleled by a rapid and massive increase in metastases. These results support the applicability of ctDNA for assessing temporal tumor heterogeneity in ACC. However, ctDNA gene mutations were detected in some aggressive ACCs, while it was not detected in other ACCs even with high tumor volume [5, 6]. Factors influencing the concentration of ctDNA include tumor size, stage, and cancer type [3, 4]. The mean ctDNA content was high in colorectal cancer and low in pancreatic cancer, renal cancer, and glioblastoma; however, an explanation for the cause is lacking [3]. Additionally, in ACC, ctDNA gene mutations may be less likely to be detected. Further studies are needed to test this hypothesis.

CtDNA analysis can be a useful tool in personalized, targeted treatment options for aggressive malignancies. In Japan, the FoundationOne Liquid CDx cancer genome profile has been approved by pharmaceutical affairs bodies and covered by insurance, enabling clinical application of ctDNA-based cancer genome profile testing [8]. Characteristic genomic profiles were detected in 80% (96/126) of ACCs, and almost half had genetic mutations with drugs that are effective in other carcinomas [10]. Our case reported CTNNB1 mutation in 10–15% of ACCs [11, 12]. CTNNB1 encodes beta-catenin, a key downstream component of the Wnt pathway [13]. Abnormal beta-catenin protein expression is associated with advanced tumor stage, metastasis, and poor prognosis in ACC [12]. Some clinical trials targeting this pathway are underway and are expected to lead to the development of effective therapeutics.

Our study had some limitations. First, we could only evaluate our patient’s ctDNA twice, at 23 and 27 months post-operatively. The analysis of this patient formed part of our study on ctDNA analysis in patients with adrenal disease. Because this study was initiated during the patient’s disease progression, analyses were only performed twice. We would have preferred routine ctDNA analysis on this patient before surgery and before post-operative recurrence. Second, we could not identify enough genes that cover ACC. Given the intra-tumor and intra-time heterogeneity, ACC-associated genes other than CTNNB1 may also be identified [14]. Despite these limitations, the study has some strengths. Our results showed that a genetic mutation consistent with a primary tumor in advanced ACC could be detected in ctDNA in peripheral blood and that the allele ratio increased with disease progression. As the cause of the low detection rate of ctDNA mutations in ACC is unknown, analyzing the clinical findings of each ACC case where a ctDNA mutation is detected is important. This study’s findings raise the question of whether ctDNA analysis could be a more sensitive marker for detecting ACC recurrence or metastasis than serum DHEA-S or CT findings.

In conclusion, we reported ctDNA analysis in a patient with post-operative recurrence of ACC (Graphical Abstract). Monitoring ctDNA could help to evaluate disease progression and predict prognosis in patients with advanced ACC. However, in clinical practice, methods for the early prediction of future ACC progression or diagnosis of ACC before it progresses must be assessed. Further studies in larger cohorts are needed to demonstrate ctDNA’s potential utility.

Graphical Abstract In this study, we analyzed ctDNA collected from the peripheral blood of the patient with post-operative recurrent ACC. At 23 and 27 months post-operatively, CTNNB1 mutation consistent with the resected left adrenal tumor tissue was identified in ctDNA. The allelic ratio of ctDNA increased from 8% to 27% as the disease progressed.

Acknowledgments

We would like to thank Editage (www.editage.com) for English language editing. This work was supported by the research grant program of the Hokkoku Cancer Foundation and JSPS KAKENHI (Grant Number: 22K16390).

Author Contributions

DA, SeK, MK, TY, and ShK managed the patient and obtained her information. DA and TK wrote the manuscript. DA, TK, AM, SaK, MK, and KH performed the genetic analyses. DA, SeK, MK, TY, KH, and ShK contributed to the discussions. All the authors have read and agreed to the published version of the manuscript.

Disclosure

None of the authors have any potential conflicts of interest associated with this research.

Supplementary Table 1 Target gene sequencing

Target	# Regions	Size (bp)	Percent Covered	Chr	Start	Stop	Strand
APC (324)	19	8,931	100	5	112043415	112179823	+
ARMC5 (79798)	6	4,760	94.01	16	31469718	31478210	+
ATP1A1 (476)	25	4,007	96.88	1	116916134	116947353	+
ATP2B3 (492)	21	3,837	100	X	152801706	152845681	+
CACNA1D (776)	50	6,650	99.97	3	53529315	53845433	+
CACNA1H (8912)	34	7,095	100	16	1203738	1270994	+
CTNNB1 (1499)	14	2,346	100	3	41265560	41280833	+
CYP11B1 (1584)	11	1,725	100	8	143955789	143961229	–
GNAS (2778)	22	5,539	96.71	20	57415162	57485884	+
HSD3B2 (3284)	3	1,119	100	1	119958043	119965243	+
KCNJ5 (3762)	2	1,260	100	11	128781169	128786626	+
MEN1 (4221)	9	1,848	100	11	64571806	64577581	–
NF1 (4763)	61	9,486	97.85	17	29422328	29701173	+
PDE1A (5136)	18	3,125	100	2	183011806	183387366	–
PRKACA (5566)	13	1,793	87.06	19	14203924	14228240	–
PRKAR1A (5573)	11	4,169	100	17	66511541	66547265	+
RET (5979)	21	4,127	99.83	10	43572707	43623717	+
VHL (7428)	3	1,044	100	3	10183532	10192051	+

References

• 1   Terzolo  M,  Daffara  F,  Ardito  A,  Zaggia  B,  Basile  V, et al. (2014) Management of adrenal cancer: a 2013 update. J Endocrinol Invest 37: 207–217.
• 2   Fassnacht  M,  Dekkers  OM,  Else  T,  Baudin  E,  Berruti  A, et al. (2018) European Society of Endocrinology Clinical Practice Guidelines on the management of adrenocortical carcinoma in adults, in collaboration with the European Network for the Study of Adrenal Tumors. Eur J Endocrinol 179: G1–G46.
• 3   Zill  OA,  Banks  KC,  Fairclough  SR,  Mortimer  SA,  Vowles  JV, et al. (2018) The landscape of actionable genomic alterations in cell-free circulating tumor DNA from 21,807 advanced cancer patients. Clin Cancer Res 24: 3528–3538.
• 4   Wan  JCM,  Massie  C,  Garcia-Corbacho  J,  Mouliere  F,  Brenton  JD, et al. (2017) Liquid biopsies come of age: towards implementation of circulating tumour DNA. Nat Rev Cancer 17: 223–238.
• 5   Creemers  SG,  Korpershoek  E,  Atmodimedjo  PN,  Dinjens  WNM,  van Koetsveld  PM, et al. (2017) Identification of mutations in cell-free circulating tumor DNA in adrenocortical carcinoma: a case series. J Clin Endocrinol Metab 102: 3611–3615.
• 6   Garinet  S,  Nectoux  J,  Neou  M,  Pasmant  E,  Jouinot  A, et al. (2018) Detection and monitoring of circulating tumor DNA in adrenocortical carcinoma. Endocr Relat Cancer 25: L13–L17.
• 7   Aono  D,  Kometani  M,  Asano  T,  Konishi  S,  Karashima  S, et al. (2022) A case of adrenocortical carcinoma diagnosed 9 years after the discovery of adrenal incidentaloma. Endocr J 69: 1415–1421.
• 8   Yoshii  Y,  Okazaki  S,  Takeda  M (2021) Current status of next-generation sequencing-based cancer genome profiling tests in Japan and prospects for liquid biopsy. Life (Basel) 11: 796.
• 9   Shu  Y,  Wu  X,  Tong  X,  Wang  X,  Chang  Z, et al. (2017) Circulating tumor DNA mutation profiling by targeted next generation sequencing provides guidance for personalized treatments in multiple cancer types. Sci Rep 7: 583.
• 10   Nazha  B,  Zhuang  TZ,  Dada  HI,  Drusbosky  LM,  Brown  JT, et al. (2022) Blood-based next-generation sequencing in adrenocortical carcinoma. Oncologist 27: 462–468.
• 11   Ross  JS,  Wang  K,  Rand  JV,  Gay  L,  Presta  MJ, et al. (2014) Next-generation sequencing of adrenocortical carcinoma reveals new routes to targeted therapies. J Clin Pathol 67: 968–973.
• 12   Ragazzon  B,  Libé  R,  Gaujoux  S,  Assié  G,  Fratticci  A, et al. (2010) Transcriptome analysis reveals that p53 and {beta}-catenin alterations occur in a group of aggressive adrenocortical cancers. Cancer Res 70: 8276–8281.
• 13   Bonnet-Serrano  F,  Bertherat  J (2018) Genetics of tumors of the adrenal cortex. Endocr Relat Cancer 25: R131–R152.
• 14   Scherer  F (2020) Capturing tumor heterogeneity and clonal evolution by circulating tumor DNA profiling. Recent Results Cancer Res 215: 213–230.