Initial validation of the clinical significance of the NETest in Japanese gastroenteropancreatic neuroendocrine tumor patients

Abstract

As novel biomarkers for gastroenteropancreatic neuroendocrine tumors (GEPNET) are in demand, we aimed to validate the clinical value of the NETest in Japanese patients. Between 2021 and 2023, blood and clinical data were collected from patients with GEPNET. Among 35 patients (median age: 59 [49–66] years), 27 cases originated from the pancreas and eight from the gastrointestinal tract. Of 69 samples sent to the laboratory, 56 (81.2%) underwent NETest. The diagnostic sensitivity was 97.1%. Among three patients who underwent R0 resection and four treated with peptide receptor radionuclide therapy, the changes in NETest scores closely correlated with disease progression. The NETest demonstrated high diagnostic efficacy and accurate therapeutic monitoring capabilities in a Japanese population.

Introduction

Neuroendocrine neoplasms (NENs) constitute a heterogeneous group of tumors having origins in the neuroendocrine system and the potential to occur in various organs [1]. The incidence and prevalence is increasing. The tumors exhibit unique features that exhibit the characteristics of both endocrine and neural cells, contributing to their diverse clinical presentations and rendering diagnosis difficult. A critical clinical unmet need is the availability of an accurate blood biomarker.

In the past Neuron-specific enolase (NSE) and chromogranin A (CgA) have been considered as serum biomarkers for assessing neuroendocrine tumor activity [2]; however, despite their utility, these biomarkers have certain limitations including lower sensitivity [3], non-specific false positives, and susceptibility to various medications, particularly proton pump inhibitors [4]. Therefore, there is a critical demand for the development of novel biomarkers that can offer more precise monitoring of disease activity and enhance diagnostic accuracy and treatment decision-making. The recognition that monoanalyte markers do not have the capacity to define the complex biology of a tumor has led to the development of multianalyte markers that identify the specific biology of an NEN. This strategy is referred to as liquid biopsy. In particular, the development of an mRNA-specific group of NEN tumor markers that are effective in blood has demonstrated a significant increase in sensitivity and specificity compared to CgA. Indeed, the NCCN has removed CgA from the standard of care given its lack of accuracy as a biomarker.

In recent years, researchers have actively sought more precise and noninvasive methods for diagnosing and monitoring these diseases. Among these methods, the NETest stands out as a blood-based multi-analyte gene testing approach designed to assess the activity levels of neuroendocrine tumors within patients [5, 6]. This test detects specific gene expression levels in the blood, providing valuable insights into tumor activity and progression. This non-invasive diagnostic method offers a more convenient means of monitoring for both patients and healthcare providers, particularly in cases where frequent monitoring is necessary.

Although this method has been extensively validated in the United States [7, 8] and European regions [9, 10], there remains a dearth of validation information applicable to the Japanese population, specifically for individuals with NENs. Recognizing this gap, Hokkaido University initiated a project in 2021 aimed at “demonstrating the usefulness of NETest in Japanese NET patients” that contributed valuable data to the broader understanding of NEN diagnostics and management in diverse populations. The study was approved by the Ethics Committee of Hokkaido University (020-0465).

Materials & Methods

Patients of Japanese origin with NEN who consented to participate and received treatment at Hokkaido University Hospital in Japan between July 2021 and May 2023 were included in this study. A total of 3 mL whole blood was collected from each participant in tubes containing stabilization buffer. The samples were then promptly stored in a –80°F freezer until they were transported to Wren Laboratory in Connecticut, USA. Thawing occurred at the time of mailing, and the samples were then transported at room temperature for NETest analysis. The average transport time from Hokkaido, Japan to Connecticut, USA was approximately three days. Analysis was conducted immediately upon sample arrival, ensuring RNA Integrity Number appropriateness, and quality of analyzed RNA. Transcripts (mRNA) are isolated, and real-time PCR is conducted following the specified protocol. Subsequently, a crossover analysis of 51 gene markers related to NENs and an algorithm are employed to calculate the corresponding score [11]. As per the defined parameters [6], NETest outcomes were scored on a scale ranging from 0 to 100 points. The normal cutoff was set at 20 points or less, stability was indicated within the range of 21–40 points, and scores exceeding 41 points were indicative of progressive disease. Clinical and pathological analyses were conducted, documenting information on the primary tumor site, pathological grading, and other relevant details. Staging for each patient was performed according to the 7th edition of the Union for International Cancer Control (UICC), categorizing them into stages 1 through 4 [12]. Meanwhile, according to the pathological staging of WHO in 2017, the tumors were classified into G1–G3 stages [13]. Additionally, we measured the NETest score for each case and assessed the changes in scores during various treatment periods. This approach allowed for a comprehensive evaluation of disease activity and treatment responses in patients with NETs. The follow-up intervals varied from 1 month to 6 months depending on the patient’s condition. For example, patients who underwent radical surgery typically had routine follow-up visits at 1 month and 2 months after curative surgery, while those undergoing PRRT treatment were routinely followed up after each treatment cycle. Other patients were scheduled for follow-up every six months. The median follow-up time was 26.3 months. In our study, patients who underwent follow-up assessments showed varying tumor responses: including Partial response (PR), stable disease (SD), and Progressive Disease (PD).

Variables were recorded as medians (P25 and P75) and analyzed using the Mann-Whitney U-test. Categorical variables were compared using Fisher’s exact test or chi-squared test. Comparisons between groups were performed using the log-rank test. Differences were considered significant at p < 0.05. Statistical analyses were performed using JMP 14 software (SAS Institute Inc., Cary, NC, USA).

Results

A total of 35 patients were included, from whom blood was sent to Wren Laboratory 69 times. NETest analysis was successfully conducted in 56 instances (56/69, 81.2%), and the data were returned. The other 13 samples were not processed due to long-distance transportation damage or delay exceeding 2 weeks. The cohort comprised 16 men and 19 women with a median age of 59 (49–66) years, with primary sites of neuroendocrine tumors including the pancreas in 27 patients (77.1%), rectum in 4 patients (11.4%), duodenum in 2 patients (5.7%), small intestine in 1 patient (2.9%), and stomach in 1 patient (2.9%). Among these cases, 20 individuals, or 57.1% of the total cohort were found to have asynchronous or synchronous liver metastases at the time of detection.

During the NETest evaluation, 4 patients (11.4%) were classified as stage I, 2 (5.7%) as stage II, 5 (14.3%) as stage III, 21 (60.0%) as stage IV; the remaining 3 (8.6%) were unstageable. The distribution of NETest scores across the UICC stages was the same (p = 0.640). According to the WHO staging system, six patients (17.1%) had stage G1, 27 (77.1%) had stage G2, and two (5.7%) had stage G3. Similarly, there was no difference in the distribution of NETest scores according to WHO stage (p = 0.828). All cases were classified as NETs based on histopathological results under the diagnosis of pathologists in our hospital, and no cases of neuroendocrine carcinoma. The last follow-up was conducted on December 1, 2023 (Table 1).

Table 1

General clinical patient characteristics

No.	Age	Sex	Origin	NELM	WHO Stage	UICC Stage
1	72	F	Pancreas	Yes	G2	IV
2	40	M	Duodenum	No	G2	NED
3	44	M	Pancreas	No	G2	NED
4	74	F	Pancreas	No	G2	III
5	49	F	Pancreas	Yes	G2	IV
6	44	F	Stomach	Yes	G2	IV
7	66	M	Pancreas	No	G2	III
8	79	M	Duodenum	No	G2	IV
9	56	M	Pancreas	No	G1	I
10	46	F	Rectum	No	G2	NED
11	65	M	Pancreas	Yes	G2	IV
12	70	F	Ileum	Yes	G1	IV
13	26	F	Rectum	Yes	G2	IV
14	59	F	Pancreas	Yes	G2	IV
15	60	F	Pancreas	Yes	G2	IV
16	54	F	Pancreas	No	G1	II
17	43	F	Pancreas	Yes	G2	IV
18	79	F	Pancreas	No	G2	III
19	58	F	Pancreas	Yes	G2	IV
20	40	F	Pancreas	Yes	G2	IV
21	61	F	Pancreas	No	G1	IB
22	71	M	Pancreas	Yes	G2	IV
23	61	M	Pancreas	Yes	G3	IV
24	64	M	Pancreas	Yes	G2	IV
25	51	F	Pancreas	No	G2	I
26	67	F	Pancreas	No	G2	III
27	62	M	Pancreas	Yes	G1	IV
28	48	M	Pancreas	No	G2	I
29	57	M	Pancreas	Yes	G2	IV
30	60	F	Rectum	No	G2	III
31	49	F	Pancreas	Yes	G1	IV
32	59	M	Pancreas	Yes	G3	IV
33	78	M	Rectum	Yes	G2	IV
34	58	M	Pancreas	Yes	G2	IV
35	42	M	Pancreas	No	G2	II

NELM: Neuroendocrine liver metastasis; WHO: World Health Organization; UICC: Union for International Cancer Control; PRRT: peptide receptor radionuclide therapy; NED: Patients who had undergone curative resection do not have tumor existed.

Of the 35 patients, the serum NSE level was 11.8 ng/mL with a reference range of 13.3 to 19.0 ng/mL, and the Ki-67 index was 5.0% with a range of 3.0% to 8.1%. The CgA immunohistochemistry results of tumors (t-CgA) was as follows: 29 cases were positive, four were negative, and two were undetected, with a positivity rate of 87.9% (29/33). Meanwhile, in 33 cases of tumors that underwent immunohistochemical testing for synaptophysin, the outcomes were all positive. Both circulating CgA and synaptophysin levels were not measured in these patients.

Among these patients, 7 had preoperative pancreatic NET and 4 underwent corresponding surgical treatments (3 distal pancreatectomies and splenic resections and 1 pancreatoduodenectomy). Eighteen patients underwent pharmacological intervention, including 1 with Multiple Endocrine Neoplasia type 1. Additionally, seven patients proceeded to receive Peptide Receptor Radionuclide Therapy (PRRT) following pharmacological treatment. The three patients who had undergone curative surgical procedures in the past for tumors originating from the pancreas, duodenum, and rectum were subsequently followed up. During the observation period, two patients died, both with tumors originating from the pancreas, due to causes unrelated to the neuroendocrine tumors (Table 2).

Table 2

Clinical indicators and NETest scores of patients

No.	NSE (ng/mL)*	t-CgA	Syn.	Ki67 (%)#	Treatment methods	1st	2nd	3rd	4th
1	14.1	+	+	5.1	chemotherapy	20.0
2	11.5	+	+	4.4	observation	26.7
3	/&	+	+	2.77	observation	73.3	x
4	10.3	+	+	7.9	chemotherapy	26.7
5	16.4	+	+	3.0	PRRT	53.3	27.0	x	x
6	13.1	+	+	N/A	chemotherapy	93.3	x
7	/	/	/	5.0	surgery	73.3	40.0
8	13.4	–	+	10.2	chemotherapy	73.3	x
9	14.1	+	+	6.6	surgery	26.7	x
10	12.4	–	+	3.0	observation	40.0	13.0
11	/	+	+	10.2	chemotherapy	26.7
12	11.8	+	+	1.0	chemotherapy	33.3
13	16.7	+	+	11.6	chemotherapy	46.7
14	19.5	+	+	3.0	chemotherapy	73.3
15	19.7	+	+	7.8	chemotherapy	60.0
16	13.1	+	+	2.4	surgery	60.0	53.3
17	12.9	+	+	15.0	chemotherapy	46.7
18	38.0	–	+	4.1	chemotherapy	60.0
19	12.9	+	+	6.3	PRRT	86.7	60.0	50.0	40.0	x
20	10.3	/	/	7.8	chemotherapy	46.7
21	/	+	+	1.0	chemotherapy	60.0	x
22	19.6	+	+	5.0	chemotherapy	46.7
23	9.7	+	+	20.1	PRRT	60.0	40.0	73.0	33.0	x
24	11.6	+	+	5.0	chemotherapy	46.7
25	/	+	+	5.0	surgery	60.0	x
26	11.5	+	+	12.2	surgery	60.0	40.0
27	18.0	+	+	2.0	chemotherapy	60.0
28	12.1	+	+	4.2	surgery	60.0	x
29	22.7	+	+	5.0	chemotherapy	40.0
30	13.0	+	+	5.3	PRRT	40.0	47.0	x	x
31	13.9	+	+	2.8	PRRT	33.3	39.9	46.6	53.0	x
32	92.3	+	+	27.7	chemotherapy	27.0
33	29.9	–	+	3.0	PRRT	47.0	x		x
34	18.8	+	+	8.7	PRRT	27.0	13.0	26.5	40.0
35	11.0	+	+	5.5	surgery	40.0

*: The Neuron-Specific Enolase (NSE) values within three months before and after sampling. ^#: The values of primary tumor. ^&: No results. 1^st–4^th: NETest results at different time points. t-CgA: The chromogranin A value of pathological tumors. Syn.: Synaptophysin. X: Samples that have incurred damage. PRRT: peptide receptor radionuclide therapy.

Regarding the distribution of NETest scores, 1 patient (2.9%) had a normal score, 12 patients (34.3%) had a stability range, and 22 patients (62.9%) had scores over 40 indicating progressive disease. Thus, approximately 97.1% of the patients had NET scores above 20, demonstrating a high sensitivity for NEN diagnosis. A comparison with t-CgA positivity showed no statistically significant difference (p = 0.160).

Among the 4 patients who underwent radical surgical resection, 3 had complete pre- and post-operative NETest scores. The outcomes indicated a significant decrease in NETest scores, with two patients showing stable scores postoperatively (z = –2.023; p = 0.043). During follow-up, none of the 3 patients relapsed, although 2 died of other illnesses 6.6 months and 10.1 months after surgery, respectively. Another patient with a postoperative score of 53.3 was followed up for 26.8 months without recurrence (Fig. 1).

Fig. 1

NETest score changes in patients undergoing R0 resection

The NETest scores for cases 7, 16, and 26 exhibited a significant decline one month after surgery (z = –2.023; p = 0.043).

Of the 7 pancreatic NET patients who received PRRT, 4 underwent complete NETest assessments before treatment and after three cycles of PRRT. These 4 cases all originated from the pancreas with liver metastasis during PRRT, and surgical resection was performed on 3 of the primary tumors. Before PRRT, two patients exhibited a progressive NETest score, while the other two demonstrated a stable score. After three rounds of PRRT, both patients with initially progressive scores showed a reduction in the stable range. Among the initially stable patients, 1 maintained a stable post-treatment score, whereas the other exhibited a transition from stable to progressive, with the score escalating from 33.3 to 53 after 3 cycles of PRRT. Subsequent follow-up revealed that the three 3 with stable NETest scores included 1 with a partial response and two with SD. In contrast, patients with elevated NETest scores were classified as having progressive disease during follow-up (Figs. 2, 3).

Fig. 2

NETest score changes in PRRT patients

The NETest scores of cases 34, 19, and 23 declined after three cycles of PRRT, whereas the score of case 31 increased (z = –0.29, p = 0.772).

PRRT: Peptide Receptor- Radionuclide Therapy.

Fig. 3

Imaging changes in patients undergoing PRRT

A: before PRRT treatment; B: after 3 cycles of PRRT

The tumors in cases 34, 19, and 23 exhibited shrinkages after three cycles of PRRT and were evaluated as PR, SD, and SD, respectively. However, the tumor diameter in case 31 increased and was assessed as PD. The imaging findings correspond to the curve changes observed in Fig. 2.

PRRT: Peptide Receptor-Radionuclide Therapy; PR: Partial response; SD: Stable disease; PD: Progressive disease.

Discussion

Like the European and American population, this study validated the high sensitivity of the NETest for diagnosing NETs in a Japanese population. As a blood-based neuroendocrine cancer diagnostic tool, the NETest can play a crucial role in diagnosing, recommending treatment, and monitoring the real-time progression of patients with neuroendocrine tumors. The test involves collecting a sample of blood. Thereafter the specific identification and measurement of gene transcripts from various gastrointestinal and pancreatic neuroendocrine tumors are used to develop a NET score. This process utilizes a crossover analysis of 51 specific markers and a unique algorithm to derive the test output [6]. In our study cohort we identified the sensitivity of the NETest as approximately 97.1% to 100%. This is consistent with the reported sensitivity in the literature of 90 to 99% [6, 14]. The specificity and accuracy of the NETest further emphasize its potential clinical utility.

In this study, when comparing the diagnostic efficacy of t-CgA and NETest, NETest showed higher sensitivity than CgA, but the difference was not statistically significant. However, the literature has previously confirmed that the diagnostic performance of NETest is significantly more effective than that of serum CgA [6, 15]. A literature review from 2021 comparing preoperative NET scores and serum CgA values in 103 patients with NET demonstrated that the diagnostic sensitivity of the NETest reached 100%, indicating its high sensitivity compared to serum CgA [6]. A potential reason for this difference may be that compared to serum CgA, t-CgA exhibited significantly higher levels (78.3% vs. 53.3%, p = 0.004) [16]. Unfortunately, there were not enough serum CgA values available to compare the diagnostic sensitivity between serum CgA and NETest in this study, as well as to monitor disease progression [10]. It should be noted that the NCCN has recently removed CgA from the standard of care.

The existing literature has reported 64–70% of patients undergoing R0 resection to have had normalized NET scores (0–20 scores) 30 days postoperatively, while patients undergoing R1 and R2 resections maintained scores over 20 [17, 18]. Moreover, patients who underwent R0 resection with increased NET scores 30 days postoperatively had a higher recurrence rate than normal patients [17, 18], suggesting that the NET scores perform well in predicting postoperative NET recurrence [9]. But according to the study [18], the postoperative NETest score of 1/3 patients with R0 resection is higher than 20. The potential reason is the identification of minimal residual pancreatic disease by imaging was difficult, which may lead to a higher postoperative NETest score.

Among the three pancreatic originate patients in our cohort who underwent R0 resection, a significant decrease was observed but not bellowed 20 and no recurrence was noted during follow-up. But it should be emphasized that the post-operation NETest scores were all exceeded 20, especially in case 16 with scores over 53. In the Modlin’s study [18], there was also one patient with around 50 post-operation scores had not recurrence during the 18 months follow-up time. Professor Modlin attributed it to the difficulty identification of minimal residual pancreatic tumors and relatively short follow-up time for the indolent tumors. This argument seems to perfectly explain the current situation of NETest level in our three patients and requires us to follow up these patients with a longer period to monitor recurrence. Originally, we expected to have a different NETest scores at 90 days postoperatively. Unfortunately, owing to specimen tube damage during shipment, scores for the three patients could not be successfully obtained. Based on postoperative imaging follow-up, the scores at 90 days post-surgery should be below 20; nevertheless, our data have also proven the role of NETest in monitoring disease progression [7, 19].

NETest accurately monitors PRRT response and is an effective surrogate marker of PRRT radiological response [20], especially for those categorized as PRRT responders; the scores after four cycles of PRRT should decrease to below 40. While according to the large cohort study [20], the NETest score decrease under 40 in more than 90% patients after 3 PRRT sessions regardless of the process including 1st PRRT and 2nd PRRT. So, we began to evaluate the treatment after 3 cycles PRRT, and the NETest score curves of 4 patients demonstrated a precise alignment with clinical tumor treatment progression. The tumor volume of Case 31 (assessed as PD) increased significantly, and the NETest score continued to increase. In case 19 (assessed as SD), the volume of the left liver tumor seemed increased, but the blood supply was obviously reduced and the necrosis was existed, which may be the main factor for the seemed larger tumor; in addition, the tiny space occupying the right posterior lobe of the liver has basically disappeared after three cycles of PRRT. Therefore, the NETest score of case 19 also shows a downward trend. While in case 23 (assessed as SD), there was an obvious fluctuation between the 1st round and the 3rd round. Unfortunately, we could not know the specific reason for this because there were no obvious changes happened in the disease course and the lack of CT evaluation of the tumors at the 2nd round PRRT. The NETest score of case 34 (assessed as PR) increased slightly, but it was not higher than 40. Considering this patient was pancreatic originated, maybe it is because the potentially existed tiny residual pancreatic tumor was not identified yet.

Due to significant tube damage or delays in sample receipt at the laboratory, we were unable to obtain the NETest scores for these patients after the fourth cycle. However, the scores of the three PRRT-responsive patients did not exceed 40. This observation strongly substantiates the potential role of NETest in monitoring the efficacy of PRRT; hence, monitoring the efficacy of surgical interventions or PRRT and analyzing the trends and intervals of NETest scores can provide initial indications of disease stability. A consistent upward trend within the progressive disease range may raise a strong suspicion of disease progression, necessitating potential adjustments to the corresponding treatment strategies as deemed necessary.

In summary, this study validated the high sensitivity of the NETest for diagnosing NETs in a Japanese population. But because of the sample damage or delays, the relatively small number of cases, and short observation period (especially in PRRT), the bias should be introduced in our study. Then, there should be some differences in the NETset scores when European and American subjects were compared with Japanese subjects as control. Also, we cannot deduce the same results should be obtained if the study was performed among other races, such like Caucasian or African populations.

Conclusion

As a blood-based testing method, the NETest has been validated the high sensitivity for accurately diagnosing NETs and the preliminary clinical significance in assessing the therapeutic efficacy of tumor surgery and monitoring treatment outcomes in a Japanese population for the first time. While these holds promise for guiding early diagnoses and treatment decisions, it is necessary to establish local laboratories to run these assays to make the results more accurate and efficient. As research on the NETest continues, we expect this method to play a more significant role in high-risk population screening and disease treatment monitoring.

Acknowledgments

Takahiro Tsuchikawa and Hao Zhang conceived and designed the study and wrote the first draft of manuscript; Hao Zhang, Satoshi Takeuchi, Kimitak Tanaka, Aya Matsui, and Yoshitsugu Nakanishi collected the data; Hao Zhang, Takahiro Tsuchikawa, Kenji Hirata, and Satoshi Hirano contributed to analysis and manuscript preparation; Satoshi Takeuchi, Satoshi Hirano, Toshimichi Asano, Takehiro Noji, Toru Nakamura, Shintaro Takeuchi, Masataka Wada, and Satoshi Hirano commented on previous versions of the manuscript; All authors read and approved the final manuscript.

Meanwhile, we gratefully acknowledge the advice and support of Dr. M. Kidd and Prof I. Modlin in helping with the preparation of this manuscript. In addition, we appreciate the pro bono measurement of the samples.

Disclosure

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

ZHANG HAO: China Scholarship Council; No. 202108515065.

Data Availability Statement

All data generated or analyzed during this study are included in this published article.

References

• 1   Liu  M,  Li  N,  Tang  H,  Chen  L,  Liu  X, et al. (2023) The mutational, prognostic, and therapeutic landscape of neuroendocrine neoplasms. Oncologist 28: e723–e736.
• 2   Kečkéš  Š,  Palaj  J,  Waczulíková  I,  Dyttert  D,  Mojtová  E, et al. (2021) Pretreatment levels of chromogranin A and neuron-specific enolase in patients with gastroenteropancreatic neuroendocrine neoplasia. In Vivo 35: 2863–2868.
• 3   van Adrichem  RC,  Kamp  K,  Vandamme  T,  Peeters  M,  Feelders  RA, et al. (2016) Serum neuron-specific enolase level is an independent predictor of overall survival in patients with gastroenteropancreatic neuroendocrine tumors. Ann Oncol 27: 746–747.
• 4   Corsello  A,  Di Filippo  L,  Massironi  S,  Sileo  F,  Dolcetta Capuzzo  A, et al. (2018) Vasostatin-1: a novel circulating biomarker for ileal and pancreatic neuroendocrine neoplasms. PLoS ONE 13: e0196858.
• 5   Modlin  IM,  Kidd  M,  Bodei  L,  Drozdov  I,  Aslanian  H (2015) The clinical utility of a novel blood-based multi-transcriptome assay for the diagnosis of neuroendocrine tumors of the gastrointestinal tract. Am J Gastroenterol 110: 1223–1232.
• 6   Modlin  IM,  Kidd  M,  Frilling  A,  Falconi  M,  Filosso  PL, et al. (2021) Molecular genomic assessment using a blood-based mRNA signature (NETest) is cost-effective and predicts neuroendocrine tumor recurrence with 94% accuracy. Ann Surg 274: 481–490.
• 7   Bodei  L,  Raj  N,  Do  RK,  Mauguen  A,  Krebs  S, et al. (2023) Interim analysis of a prospective validation of 2 blood-based genomic assessments (PPQ and NETest) to determine the clinical efficacy of (177)Lu-DOTATATE in Neuroendocrine Tumors. J Nucl Med 64: 567–573.
• 8   Al-Toubah  T,  Cives  M,  Valone  T,  Blue  K,  Strosberg  J (2021) Sensitivity and specificity of the NETest: a validation study. Neuroendocrinology 111: 580–585.
• 9   Puliani  G,  Di Vito  V,  Feola  T,  Sesti  F,  Centello  R, et al. (2022) NETest: a systematic review focusing on the prognostic and predictive role. Neuroendocrinology 112: 523–536.
• 10   van Treijen  MJC,  Korse  CM,  Verbeek  WH,  Tesselaar  MET,  Valk  GD (2022) NETest: serial liquid biopsies in gastroenteropancreatic NET surveillance. Endocr Connect 11: e220146.
• 11   Pavel  M,  Jann  H,  Prasad  V,  Drozdov  I,  Modlin  IM, et al. (2017) NET blood transcript analysis defines the crossing of the clinical rubicon: when stable disease becomes progressive. Neuroendocrinology 104: 170–182.
• 12  Brierley JD, Gospodarowicz MK, Wittekind C (eds) (2017) TNM classification of malignant tumours (8th). Wiley-Blackwell, Chichester, UK.
• 13   Rindi  G,  Klimstra  DS,  Abedi-Ardekani  B,  Asa  SL,  Bosman  FT, et al. (2018) A common classification framework for neuroendocrine neoplasms: an International Agency for Research on Cancer (IARC) and World Health Organization (WHO) expert consensus proposal. Mod Pathol 31: 1770–1786.
• 14   Pęczkowska  M,  Cwikla  J,  Kidd  M,  Lewczuk  A,  Kolasinska-Ćwikła  A, et al. (2017) The clinical utility of circulating neuroendocrine gene transcript analysis in well-differentiated paragangliomas and pheochromocytomas. Eur J Endocrinol 176: 143–157.
• 15   Modlin  IM,  Kidd  M,  Falconi  M,  Filosso  PL,  Frilling  A, et al. (2021) A multigenomic liquid biopsy biomarker for neuroendocrine tumor disease outperforms CgA and has surgical and clinical utility. Ann Oncol 32: 1425–1433.
• 16   Zhang  C,  Huang  Y,  Long  J,  Yao  X,  Wang  J, et al. (2019) Serum chromogranin A for the diagnosis of gastroenteropancreatic neuroendocrine neoplasms and its association with tumour expression. Oncol Lett 17: 1497–1504.
• 17   Bodei  L,  Kidd  M,  Modlin  IM,  Severi  S,  Drozdov  I, et al. (2016) Measurement of circulating transcripts and gene cluster analysis predicts and defines therapeutic efficacy of peptide receptor radionuclide therapy (PRRT) in neuroendocrine tumors. Eur J Nucl Med Mol Imaging 43: 839–851.
• 18   Modlin  IM,  Kidd  M,  Oberg  K,  Falconi  M,  Filosso  PL, et al. (2021) Early identification of residual disease after neuroendocrine tumor resection using a liquid biopsy multigenomic mRNA signature (NETest). Ann Surg Oncol 28: 7506–7517.
• 19   Öberg  K,  Califano  A,  Strosberg  JR,  Ma  S,  Pape  U, et al. (2020) A meta-analysis of the accuracy of a neuroendocrine tumor mRNA genomic biomarker (NETest) in blood. Ann Oncol 31: 202–212.
• 20   Bodei  L,  Kidd  MS,  Singh  A,  van der Zwan  WA,  Severi  S, et al. (2020) PRRT neuroendocrine tumor response monitored using circulating transcript analysis: the NETest. Eur J Nucl Med Mol Imaging 47: 895–906.