Abstract
Sufficient expression of somatostatin receptor (SSTR) in well-differentiated neuroendocrine tumors (NETs) is crucial for treatment with somatostatin analogs (SSAs) and peptide receptor radionuclide therapy (PRRT) using radiolabeled SSAs. Impaired prognosis has been described for SSTR-negative NET patients; however, studies comparing matched SSTR-positive and -negative subjects who have not received PRRT are missing. This retrospective analysis of two prospectively maintained NET databases aimed to compare matched metastatic grade 1 or 2 SSTR-positive and –negative NET patients. SSTR-negativity was defined as having insufficient tumor uptake on diagnostic SSTR imaging. Patients that underwent PRRT were excluded. Seventy-seven SSTR-negative and 248 SSTR-positive grade 1–2 NET patients were included. Median overall survival rates were significantly lower for SSTR-negative compared to SSTR-positive NET patients (53 months vs 131 months; P < 0.001). To adjust for possible confounding by age, gender, grade and site of origin, 69 SSTR-negative NET patients were propensity score matched to 69 SSTR-positive NET patients. Group characteristics were similar, with the exception of SSTR-negative patients receiving more often chemotherapy and targeted treatment. The inferior survival outcome of SSTR-negative compared to SSTR-positive NET patients persisted with a median overall survival of 38 months vs 131 months (P = 0.012). This relationship upheld when correcting for the main influencing factors of having a higher grade tumor or receiving surgery in a multivariate Cox regression analysis. In conclusion, we showed that propensity score-matched SSTR-negative NET patients continue to have a worse prognosis compared to SSTR-positive NET patients despite receiving more aggressive treatment. Differences in tumor biology likely underlie this survival deficit.
Introduction
Neuroendocrine neoplasms (NENs) consist of a diverse group of tumors arising from neuroendocrine cells and are mainly localized in the intestine, pancreas and lung (Hofland et al. 2020). Classification of NENs is based on their origin, extension and histological differentiation (Perren et al. 2017). Based on their histology, mitotic count, the expression of the nuclear protein marker Ki-67 and additional markers of differentiation, NENs are classified as well-differentiated grade 1, 2 or 3 neuroendocrine tumors (NETs) or poorly differentiated grade 3 neuroendocrine carcinomas (Nagtegaal et al. 2020). Around 25% of NENs produce hormones or peptides and are the so-called functioning tumors (Zandee et al. 2017). However, the majority of NENs are slow growing non-secreting tumors, leading to a late diagnosis at a metastatic stage (Dasari et al. 2017).
Characteristic features of well-differentiated NETs are the expression of hormone receptors on its tumor surface, with the somatostatin receptor subtype 2 (SST2) being expressed in up to 90% of gastroenteropancreatic NETs (Reubi et al. 2010, Gatto & Hofland 2011). According to this feature, SSTR-based imaging is the current standard for diagnosing and staging NETs (Sundin et al. 2017). Furthermore, treatment with unlabelled somatostatin analogs (SSAs) or radiolabelled SSAs as in peptide receptor radionuclide therapy (PRRT) has shown prolongation of progression-free survival (PFS) in metastatic NET patients (Rinke et al. 2009, Caplin et al. 2014, Strosberg et al. 2017). However, since sufficient uptake of radiolabelled SSAs on SSTR imaging is a prerequisite for those therapies (Hicks et al. 2017), SSA treatment is questionable and PRRT not feasible in SSTR-negative NET patients. To overcome this limitation, systemic treatment options such as chemotherapy and targeted therapy are used in these patients, but chemotherapy is not an option in metastatic midgut NETs (Raj & Reidy-Lagunes 2014). This could partly explain the poorer survival rates in SSTR-negative NET patients (Mehta et al. 2015). Other data have suggested that a more aggressive tumor biology was responsible for the worse outcome, but these results were obtained in uncontrolled studies (Okuwaki et al. 2013, Qian et al. 2016). Studies comparing outcomes and covariates between matched SSTR-negative and -positive NET patients are missing.
Accordingly, the aim of this retrospective analysis was to compare outcomes of metastatic grade 1–2 SSTR-negative and -positive NET patients from the databases of two large NEN cohorts. We hypothesized that the poorer outcome of SSTR-negative NET patients would persist after patient matching using the propensity score method and covariant adjustment.
Methods
Patient characteristics
The data of this retrospective analysis was obtained from the years 2000–2019 from two prospective observational studies of NEN patients from the ENETS Centre of Excellence for Neuroendocrine Tumours Erasmus MC Rotterdam, the Netherlands and the Swiss national register of neuroendocrine tumors SwissNET. Local Institutional Review Boards (IRBs; Rotterdam: IRB of the Erasmus MC; Switzerland: Federal Office of Public Health) approved both databases. Informed consent was obtained of all Swiss patients, the IRB of the Erasmus MC approved a waiver of patient informed consent for the retrospective analysis.
Patients were selected if they had advanced stage well-differentiated grade 1 or 2 NETs. Patients that received treatment with PRRT were excluded from this study because of its beneficial effect on survival in favor of SSTR-positive tumors. Tumor characteristics, grade and stage were documented at the time of diagnosis according to the 2010 WHO grading and ENETS guidelines (Pavel et al. 2016). Treatments were recorded over the entire follow-up period. Highest serum chromogranin A levels (reference <94 µg/L) of the follow-up period were used.
SSTR negativity was defined as having an uptake lower than the liver on SSTR-scintigraphy (Kwekkeboom et al. 2005) and equal to or lower than the liver on SSTR-PET (Werner et al. 2018)at the time of diagnosis.
Primary and secondary outcomes
The primary endpoint was to compare survival outcomes between patients with a SSTR-negative NET and those with a SSTR-positive NET. Secondary endpoints were to define prognostic factors for worse outcome.
Statistics
Descriptive statistics were used to characterize demographic and clinical data, expressed as median with interquartile range (IQR) or frequencies with percentage (%). The association of categorical variables was assessed by Pearson chi-square test, continuous variables by Mann–Whitney U test.
Outcome analysis was divided into three parts. First, data of all eligible patients were evaluated according to their SSTR status. Survival rates (= survival from date of diagnosis to last date of follow-up or death) were calculated with the Kaplan–Meier method, the log-rank test was used to compare survival differences between groups. A Cox proportional hazard model was used to calculate mortality hazard ratios (HR) and 95% CIs. Only variables that significantly affected survival in univariate analysis were included in the Cox model. The proportionality of hazards was evaluated using the Cox regression analysis with time-dependent covariates. The assumption of proportionality of hazards was tested and was not broken in any of the Cox regression models.
Second, all analyses were repeated in a propensity score (PS)-matched cohort. PS analysis optimizes the post-weighting balance of covariates between groups (Austin 2009, 2013). Using the PS methodology, all patients were assigned a weight between 0 and 1 according to the covariates age, gender, tumor grade and site of origin. Using caliper matching without replacement, we used PSs to match SSTR-negative to SSTR-positive NET patients in a 1:1 ratio. A caliper distance of an absolute difference in PS of 0.1 was employed, leading to 69 matched patients in each group, with 8 SSTR-negative NET patients failing to match within the defined scope. Successful matching was indicated by the absence of statistical significance between the variables.
Third, subgroup analysis involving all patients treated with SSA were performed for overall survival and PFS as described previously. PFS was defined as survival without progression from the start of SSA treatment to the last date of follow-up or progression. PS-matching, as described previously, was again used to better compare the SSTR-negative and -positive patients. Also, the effect of SSA treatment among the SSTR-negative patients was investigated.
Statistical analyses were performed using SPSS version 25.0 (IBM Corp.) and R 3.3.3 open-source. P values <0.05 were considered statistically significant.
Results
Complete cohort
Patient characteristics
After excluding all patients receiving PRRT, a total of 77 SSTR-negative NET patients and 248 SSTR-positive NET patients met the inclusion criteria and were included into the analysis. SSTR-status was defined according to SSTR-scintigraphy in 255 (78%) patients, while in 28 (9%) and 42 (13%) Ga-DOTATOC or Ga-DOTATATE PET/CT was used respectively. The two patient groups differed with regard to the distribution of the site of origin, grade, metastatic locations and received treatment (see Table 1 for details). Median (IQR) follow-up time for SSTR-negative and -positive NET patients was 34 (15–67) and 51 (16–83) months, respectively (P = 0.09).
Characteristics of the complete cohort.
| Complete cohort | SSTR status negative (n = 77) | SSTR status positive (n = 248) | P value |
|---|---|---|---|
| Sex (female), n (%) | 41 (53) | 119 (48) | 0.4 |
| Age at diagnosis, years (IQR) | 63 (55–68) | 63 (54–71) | 0.3 |
| Primary tumor site, n (%) | <0.001 | ||
| Gastro-duodenal | 0 | 4 (2) | n.a. |
| Midgut | 24 (31) | 172 (69) | <0.001 |
| Hindgut | 1 (1.3) | 4 (2) | 0.18 |
| Pancreas | 13 (17) | 35 (14) | 0.001 |
| Lung | 24 (31.2) | 11 (4) | 0.028 |
| Unknown | 15 (19.5) | 22 (9) | 0.32 |
| Grade, n (%) | |||
| 1/2 | 29 (38)/48 (62) | 136 (55)/112 (45) | 0.009 |
| Gastro-duodenal | 0 (0) | 2 (50)/2 (50) | n.a. |
| Midgut | 13 (54)/11 (46) | 104 (60)/68 (40) | 0.007 |
| Hindgut | 0 (0)/1 (100) | 2 (50)/2 (50) | 0.66 |
| Pancreas | 4 (31)/9 (69) | 9 (26)/26 (74) | 0.001 |
| Lung | 9 (38)/15 (62) | 7 (64)/4 (36) | 0.61 |
| Unknown | 3 (20)/12 (80) | 12 (54)/10 (46) | 0.32 |
| Hormone secreting, n (%) | 23 (30) | 78 (32) | 0.6 |
| Chromogranin A, ng/l (IQR) | 255 (91–952) | 261 (101–907) | 0.9 |
| Site of metastasis, n (%) | |||
| Lymphnodes | 60 (78) | 217 (89) | 0.2 |
| Liver | 61 (79) | 193 (78) | 0.4 |
| Lung | 11 (14) | 13 (5) | 0.001 |
| Bone | 21 (27) | 28 (11) | <0.001 |
| Treatment, n (%) | |||
| Somatostatin analogs | 35 (46) | 152 (61) | 0.014 |
| Chemotherapy | 23 (30) | 12 (5) | <0.001 |
| Targeted therapy | 22 (29) | 4 (2) | <0.001 |
| Liver directed therapy | 6 (8) | 23 (9) | 0.7 |
| Surgery | 45 (58) | 177 (71) | 0.033 |
| Curative | 19 (42) | 113 (64) | 0.007 |
| Palliative | 26 (58) | 64 (36) | 0.011 |
| Died, n (%) | 43 (56) | 72 (29) | <0.001 |
Bold indicates statistical significance, P < 0.05.
Outcomes
At last follow-up, 56% (n = 43/77) of the SSTR-negative NET patients had died, compared to 29% (n = 72/248) of the SSTR-positive NET patients (P < 0.001). Median (95% CI) overall survival time was significantly lower for SSTR-negative NET patients with 53 months (26–80) compared to 131 months (89–173) for SSTR-positive NET patients (P < 0.001, Fig. 1). Multivariate Cox regression analysis confirmed SSTR-negativity as an independent risk factor for mortality with a hazard ratio (95% CI) of 1.85 (1.21–2.83), P = 0.005. Additional determinants for worse outcome were age above 65 years (HR 2.11 (1.42–3.12), P = 0.001), having a grade 2 tumor (HR 1.75 (1.17–2.61), P = 0.007) and the presence of bone metastases (HR 1.80 (1.14–2.82), P = 0.011) (Table 2). Having received any surgical intervention was associated with a significantly reduced overall mortality risk (HR 0.39 (0.26–0.57), P < 0.001). However, neither the intent of surgery (curative vs palliative: P = 0.42) nor the primary tumor site (P = 0.69) had an effect on survival outcome. Figure 2 shows the survival outcomes of SSTR-negative and -positive patients according to their tumor grade (Fig. 2A) and surgical status (Fig. 2B).
Overall survival complete cohort. Kaplan–Meier analysis showing overall survival of the complete cohort, divided into SSTR-positive (n = 248) and SSTR-negative (n = 77) patients. P-value indicates difference in survival (log rank test). SSTR, somatostatin receptor.
Citation: Endocrine-Related Cancer 27, 11; 10.1530/ERC-20-0340
Overall survival complete cohort. Kaplan–Meier analysis showing overall survival of the complete cohort, divided into SSTR-positive (n = 248) and SSTR-negative (n = 77) patients. P-value indicates difference in survival (log rank test). SSTR, somatostatin receptor.
Citation: Endocrine-Related Cancer 27, 11; 10.1530/ERC-20-0340
Overall survival complete cohort. Kaplan–Meier analysis showing overall survival of the complete cohort, divided into SSTR-positive (n = 248) and SSTR-negative (n = 77) patients. P-value indicates difference in survival (log rank test). SSTR, somatostatin receptor.
Citation: Endocrine-Related Cancer 27, 11; 10.1530/ERC-20-0340
Overall survival complete cohort according to covariates. Kaplan–Meier analysis showing overall survival of the complete cohort, according to (A) SSTR-status and tumor grade 1 (n = 165) or 2 (n = 160) and (B) SSTR status and having received a surgical intervention (n = 222) or not (n = 103). P value indicates difference in survival between the groups (log rank test). SSTR, somatostatin receptor; G1, grade 1; G2, grade 2.
Citation: Endocrine-Related Cancer 27, 11; 10.1530/ERC-20-0340
Overall survival complete cohort according to covariates. Kaplan–Meier analysis showing overall survival of the complete cohort, according to (A) SSTR-status and tumor grade 1 (n = 165) or 2 (n = 160) and (B) SSTR status and having received a surgical intervention (n = 222) or not (n = 103). P value indicates difference in survival between the groups (log rank test). SSTR, somatostatin receptor; G1, grade 1; G2, grade 2.
Citation: Endocrine-Related Cancer 27, 11; 10.1530/ERC-20-0340
Overall survival complete cohort according to covariates. Kaplan–Meier analysis showing overall survival of the complete cohort, according to (A) SSTR-status and tumor grade 1 (n = 165) or 2 (n = 160) and (B) SSTR status and having received a surgical intervention (n = 222) or not (n = 103). P value indicates difference in survival between the groups (log rank test). SSTR, somatostatin receptor; G1, grade 1; G2, grade 2.
Citation: Endocrine-Related Cancer 27, 11; 10.1530/ERC-20-0340
Cox regression analysis complete cohort.
| Complete cohort | Univariate Cox regression analysis | Multivariate Cox regression analysis | Hazard ratio (95% CI) |
|---|---|---|---|
| SSTR negativity | <0.001 | 0.005 | 1.85 (1.21–2.83) |
| Sex (female) | 0.80 | ||
| Age > 65 years | <0.001 | 0.001 | 2.11 (1.42–3.12) |
| Primary tumor site | 0.001 | 0.97 | |
| Grade 2 vs 1 | <0.001 | 0.007 | 1.75 (1.17–2.61) |
| Site of metastasis | |||
| Lymph nodes | 0.07 | ||
| Liver | 0.011 | 0.10 | |
| Lung | 0.60 | ||
| Bone | 0.001 | 0.011 | 1.80 (1.14–2.82) |
| Treatment | |||
| Somatostatin analogs | 0.20 | ||
| Chemotherapy | 0.001 | 0.28 | |
| Targeted therapy | 0.008 | 0.06 | |
| Liver-directed therapy | 0.005 | 0.95 | |
| Surgery | <0.001 | <0.001 | 0.39 (0.26–0.57) |
| Curative vs palliative | 0.42 | ||
| Primary tumor site | 0.69 |
Bold indicates statistical significance, P < 0.05.
Propensity score-matched cohort
Patient characteristics
Propensity score (PS) matching led to two well-balanced groups of 69 patients each with a median (IQR) follow-up time for SSTR-negative NET patients of 34 (14–64) months and SSTR-positive NET patients of 44 (9–76) months, P = 0.53. With the exceptions of SSTR-negative NET patients having a higher prevalence of lung metastases and more often receiving chemotherapy and targeted treatment than patients with SSTR-positive NETs, there were no significant differences between the two groups (Supplementary Table 1, see section on supplementary materials given at the end of this article).
Outcomes
During the observation period, more SSTR-negative NET patients (59%, n = 41/69) than matched SSTR-positive (32%, n = 22/69, P = 0.001) NET patients died. Despite the PS matching, SSTR-negative NET patients still had a worse median (95% CI) overall survival of 38 (19–56) months compared to SSTR-positive NET patients with 131 (54–208) months, P = 0.012 (Fig. 3). SSTR-negativity persisted to be associated with worse outcome in multivariate Cox regression analysis with a HR (95% CI) of 2.58 (1.34–4.99), P = 0.005. Another independent determinant of mortality was the presence of a grade 2 tumor (HR 2.32 (1.18–4.58), P = 0.015), while having a surgical intervention was associated with a lower mortality risk (HR of 0.36 (0.18–0.72), P = 0.004) (Supplementary Table 2). Again, complete surgical resection and primary tumor site did not influence survival outcome (P = 0.30 and P = 0.42, respectively). Supplementary Figure 1 shows the survival curves of the PS-matched SSTR-negative and -positive patients according to their tumor grade (Supplementary Fig. 1A) and surgical status (Supplementary Fig. 1B).
Overall survival propensity score matched cohort. Kaplan–Meier analysis showing overall survival of the PS-matched cohort, divided into SSTR-positive (n = 69) and SSTR-negative (n = 69) patients. P value indicates difference in survival (log rank test). PS, propensity score; SSTR, somatostatin receptor.
Citation: Endocrine-Related Cancer 27, 11; 10.1530/ERC-20-0340
Overall survival propensity score matched cohort. Kaplan–Meier analysis showing overall survival of the PS-matched cohort, divided into SSTR-positive (n = 69) and SSTR-negative (n = 69) patients. P value indicates difference in survival (log rank test). PS, propensity score; SSTR, somatostatin receptor.
Citation: Endocrine-Related Cancer 27, 11; 10.1530/ERC-20-0340
Overall survival propensity score matched cohort. Kaplan–Meier analysis showing overall survival of the PS-matched cohort, divided into SSTR-positive (n = 69) and SSTR-negative (n = 69) patients. P value indicates difference in survival (log rank test). PS, propensity score; SSTR, somatostatin receptor.
Citation: Endocrine-Related Cancer 27, 11; 10.1530/ERC-20-0340
Treatment with somatostatin analogs
In a subgroup analysis, we evaluated the outcomes of all patients receiving SSA treatment. This concerned a total of 35 SSTR-negative NET patients and 152 SSTR-positive NET patients. The two patient groups differed with regard to the distribution of the site of origin, number of patients having bone metastasis and additional received treatments (Supplementary Table 3). Although more SSTR-negative NET patients in this group received additional chemotherapy and targeted treatment, they continued to have a worse overall survival compared to the SSTR-positive NET patients (P = 0.008, data not shown).
SSTR-negative NET patients also had a shorter median (95% CI) PFS after the start of SSA treatment of 15 (7–22) months compared to 54 (44–64) months in SSTR-positive NET patients (P < 0.001, Fig. 4A).
Outcome of patients treated with somatostatin analogs. (A) Kaplan–Meier analysis of progression-free survival of all patients treated with SSA according to SSTR-status (SSTR positive: n = 152, SSTR-negative: n = 35), (B) Kaplan–Meier analysis of overall survival of SSTR-negative patients treated with SSA (n = 35) compared to no SSA treatment (n = 42). P value indicates difference between groups according to log rank test. SSTR, somatostatin receptor; SSA, somatostatin analogs.
Citation: Endocrine-Related Cancer 27, 11; 10.1530/ERC-20-0340
Outcome of patients treated with somatostatin analogs. (A) Kaplan–Meier analysis of progression-free survival of all patients treated with SSA according to SSTR-status (SSTR positive: n = 152, SSTR-negative: n = 35), (B) Kaplan–Meier analysis of overall survival of SSTR-negative patients treated with SSA (n = 35) compared to no SSA treatment (n = 42). P value indicates difference between groups according to log rank test. SSTR, somatostatin receptor; SSA, somatostatin analogs.
Citation: Endocrine-Related Cancer 27, 11; 10.1530/ERC-20-0340
Outcome of patients treated with somatostatin analogs. (A) Kaplan–Meier analysis of progression-free survival of all patients treated with SSA according to SSTR-status (SSTR positive: n = 152, SSTR-negative: n = 35), (B) Kaplan–Meier analysis of overall survival of SSTR-negative patients treated with SSA (n = 35) compared to no SSA treatment (n = 42). P value indicates difference between groups according to log rank test. SSTR, somatostatin receptor; SSA, somatostatin analogs.
Citation: Endocrine-Related Cancer 27, 11; 10.1530/ERC-20-0340
To further validate the previous results, SSA-treated patients were PS-matched according to their SSTR-status, resulting in 35 patients in each group. Patient characteristics were well balanced, with the exception of SSTR-negative patients receiving more often targeted treatment (Supplementary Table 4). In this PS-matched cohort, SSTR-negative patients tended toward a shorter median overall survival time (39 months vs 72 months, P = 0.14) and continued to have a significantly reduced PFS (15 months vs 47 months, P = 0.006). SSTR-negativity and having a grade 2 tumor remained independent determinants of a shorter PFS in multivariate analysis in all SSA-treated patients as well as in the PS-matched cohort (all SSA-treated patients: P = 0.008 and 0.002; PS-matched SSA-treated patients: P = 0.004 and 0.002, respectively).
Finally, we investigated whether SSA treatment leads to a survival benefit in SSTR-negative NET patients. The two groups were comparable with the exception of SSA-treated SSTR-negative patients having a higher rate of liver metastases (91% vs 69%, P = 0.024). Median overall survival was similar between SSTR-negative patients receiving SSA and those without SSA treatment (P = 0.53, Fig. 4B). Having received a surgical intervention was the sole independent predictor for a decreased overall mortality risk in this multivariate Cox regression analysis (P = 0.016).
Discussion
Our study has three major findings: first, we were able to show that SSTR-tumor negativity is an independent determinant of mortality in patients with advanced grade 1–2 NETs. This observation upheld in the PS-matched cohort as well as in multivariate analysis. Second, both a lower tumor grade of the primary tumor and surgical treatment were associated with a better outcome in SSTR-negative NET patients. Third, SSA treatment did not improve progression-free or overall survival in SSTR-negative NET patients.
Our results that SSTR tumor negativity is a negative prognostic biomarker are consistent with previous, more limited studies in gastroenteropancreatic and lung NEN (Corleto et al. 2009, Okuwaki et al. 2013, Mehta et al. 2015, Song et al. 2016, Qian et al. 2016, Vesterinen et al. 2019). The main advantage of our analysis is the in vivo definition of SSTR-negativity by insufficient uptake in SSTR-imaging instead of negative immunohistochemistry. SST2 status evaluation on immunohistochemistry is inferior compared to SSTR-imaging, as it evaluates a single lesion and potentially misses interlesional heterogeneous SST2 expression (Müssig et al. 2010, Brunner et al. 2017). Accordingly, there is no additional benefit of immunohistochemistry compared to SSTR-imaging (van Adrichem et al. 2016). A further limiting factor of the aforementioned studies was their comparison of patients at all stages and tumor grades, which resulted in high data heterogeneity. Also, with the exception of two studies (Song et al. 2016, Vesterinen et al. 2019), data were derived from small cohorts with fewer than 100 patients. Most importantly, all studies failed to match the SSTR-negative NET patients to a control group, which is critical given the differences in outcome according to tumor grade and stage (Okuwaki et al. 2013, Nunez-Valdovinos et al. 2018). We aimed at overcoming those limitations by including only well-differentiated (grade 1–2) advanced stage NETs in our analysis. To further increase comparability, we excluded patients who underwent PRRT, since this treatment option significantly affects prognosis in SSTR-positive NENs (Strosberg et al. 2017). Also, we chose to not only adjust results for possible interaction of co-factors, but to create a PS-matched cohort of SSTR-negative and -positive NET patients, thereby optimizing the post-weighting balance of covariates between groups (Austin 2009, 2013).
While our analysis confirmed the worse outcome for SSTR-negative NET patients, the main question remains what causes this survival deficit. A more aggressive tumor biology has been proposed to be the underlying reason for the worse outcome of SSTR-negative tumor patients. Several studies have demonstrated a decrease in the maximum standardized uptake values (SUV) on SSTR-PET imaging with increasing Ki-67 index (Reubi 2007, Haug et al. 2010, Chan et al. 2019). However, since SSTR-negativity remained an independent predictor for worse outcome after correcting for grade via PS matching and multivariate analysis, other mechanisms presumably play a role here. SSTR-negativity could represent an overall state of epigenetic gene silencing, which might be associated with aberrant cellular growth. Modifications of the epigenome via epigenetic drugs which decrease methylation and augment histone acetylation of the SST2 gene promoter region have been shown to stimulate SST2 expression levels and inhibit cell proliferation in several in vitro and in vivo studies (Taelman et al. 2016, Veenstra et al. 2018, Wanek et al. 2018, Guenter et al. 2019, 2020, Jin et al. 2019). That SST2 expression levels correlate with prognosis is known (Asnacios et al. 2008, Kim et al. 2011); however, the role of epigenetic treatment for NET patients remains speculative.
Another essential finding in our analysis was the positive impact of having a surgical intervention on overall survival of patients with advanced stage NETs, which is in line with other studies (Chawla et al. 2018, Chakedis et al. 2019, Tierney et al. 2019). There is a potential bias in patient selection as patients with better performance status or more easily resectable tumors are more frequently operated on. Although no information on performance status was available in our cohort, the radical nature of the operation and primary tumor site had no impact on survival outcome.
Evidence on the use of SSAs as an antiproliferative treatment in NEN has increased over the past years (Rinke et al. 2009, Caplin et al. 2014). Accordingly, many physicians have also employed SSAs as a treatment for SSTR-negative tumors. An earlier study in 54 small-intestinal NENs reported limited efficacy of this treatment in SSTR-negative tumors with a median PFS of 15.6 months (Qian et al. 2016), which was similar to the PFS observed in our cohort. However, there was no survival benefit from SSA treatment in the SSTR-negative patient group. Ideally, the effect of SSA treatment in SSTR-negative tumor patients should be compared to patients with a watchful waiting strategy. Unfortunately, such a comparison was not possible in our study due to insufficient patient numbers.
One of the limitations of our study is its retrospective design. However, since data were derived from two carefully managed prospective NEN registries, the resulting relatively large cohort from several different expert centers makes the data more comparable. Having included only well-differentiated grade 1 and 2 tumors in our analysis, no statement can be made about the outcome of SSTR-negative grade 3 NETs. Despite PS-matching, there were more lung NETs and a higher rate of lung metastases in SSTR-negative tumor patients. However, since this had no impact on the multivariate analyses, a significant bias can be excluded. Also, among the SSTR-negative patients, pancreatic NET and NET of unknown origin had the highest mortality rate (data not shown).
Another limitation to our study is the different imaging modalities used. Since SSTR scintigraphy has a lower sensitivity compared to SSTR PET/CT, it is theoretically possible that patients scored negative in scintigraphy would have been positive in PET/CT (Hope et al. 2019). However, for most patients, SSTR-negativity was confirmed by immunohistochemistry or PET/CT, leaving only 16 patients with the definition solely based on SSTR scintigraphy. Primary outcomes did not change after exclusion of those 16 patients (data not shown).
Because patients were included over a long study period, different treatment strategies influenced treatment decisions, which could be a reason why nowadays eligible SSTR-positive patients in our cohort did not receive PRRT. However, the bias in treatment between the two cohorts were minimized by excluding PRRT-treated patients and a balanced use of SSA treatment among the PS-matched groups.
In summary, we were able to confirm the inferior prognosis of SSTR-negative compared to SSTR-positive NET patients. The results of our study emphasize the clinical unmet need for more effective treatment options in patients with advanced stage SSTR-negative NETs.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/ERC-20-0340.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
J R supported by a grant from the Swiss National Science Foundation (SNF P2BSP3_181720). T B speaker fees, research support and advisory board AAA. R F research support from Ipsen, Strongbridge and Corcept. Speaker fee HRA Pharma, Novartis, Ipsen. L J H research support from Ipsen, Novartis and Strongbridge. W W D H research support from Ipsen, speaker fee Ipsen, Novartis, AAA, Pfizer, advisory board AAA. J H speaker fee Ipsen, advisory board Novartis.
Author contribution statement
J R involved in data collection, data analysis and interpretation, literature search, wrote the manuscript. W T Z, T B, R F, G F, L J H, E C and W W D H contributed to data collection and edited the manuscript. J H involved in data collection, data analysis and interpretation, edited the manuscript and supervised all steps of the conduct of the analysis.
Acknowledgements
The authors thank all patients for their participation in the respective NEN registers and the research nurses Simone Lanz, Christine Lückl Boyer, Christiane Schwarzenbach for collecting the data for the SwissNET registry.
References
Asnacios A, Courbon F, Rochaix P, Bauvin E, Cances-Lauwers V, Susini C, Schulz S, Boneu A, Guimbaud R & Buscail L 2008 Indium-111-pentetreotide scintigraphy and somatostatin receptor subtype 2 expression: new prognostic factors for malignant well-differentiated endocrine tumors. Journal of Clinical Oncology 26 70. (https://doi.org/10.1200/JCO.2007.12.7431)
Austin PC 2009 Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Statistics in Medicine 28 107. (https://doi.org/10.1002/sim.3697)
Austin PC 2013 The performance of different propensity score methods for estimating marginal hazard ratios. Statistics in Medicine 32 49. (https://doi.org/10.1002/sim.5705)
Brunner P, Jörg AC, Glatz K, Bubendorf L, Radojewski P, Umlauft M, Marincek N, Spanjol PM, Krause T, Dumont RA, et al. 2017 The prognostic and predictive value of sstr(2)-immunohistochemistry and sstr(2)-targeted imaging in neuroendocrine tumors. European Journal of Nuclear Medicine and Molecular Imaging 44 . (https://doi.org/10.1007/s00259-016-3486-2)
Caplin ME, Pavel M, Cwikla JB, Phan AT, Raderer M, Sedlackova E, Cadiot G, Wolin EM, Capdevila J, Wall L, et al. 2014 Lanreotide in metastatic enteropancreatic neuroendocrine tumors. New England Journal of Medicine 371 33. (https://doi.org/10.1056/NEJMoa1316158)
Chakedis J, Beal EW, Lopez-Aguiar AG, Poultsides G, Makris E, Rocha FG, Kanji Z, Weber S, Fisher A, Fields R, et al. 2019 Surgery provides long-term survival in patients with metastatic neuroendocrine tumors undergoing resection for non-hormonal symptoms. Journal of Gastrointestinal Surgery 23 . (https://doi.org/10.1007/s11605-018-3986-4)
Chan H, Moseley C, Zhang L, Bergsland EK, Pampaloni MH, Van Loon K & Hope TA 2019 Correlation of DOTATOC uptake and pathologic grade in neuroendocrine tumors. Pancreas 48 . (https://doi.org/10.1097/MPA.0000000000001356)
Chawla A, Williams RT, Sich N, Clancy T, Wang J, Ashley S, Pezzi C & Swanson R 2018 Pancreaticoduodenectomy and metastasectomy for metastatic pancreatic neuroendocrine tumors. Journal of Surgical Oncology 118 . (https://doi.org/10.1002/jso.25219)
Corleto VD, Falconi M, Panzuto F, Milione M, De Luca O, Perri P, Cannizzaro R, Bordi C, Pederzoli P, Scarpa A, et al. 2009 Somatostatin receptor subtypes 2 and 5 are associated with better survival in well-differentiated endocrine carcinomas. Neuroendocrinology 89 30. (https://doi.org/10.1159/000167796)
Dasari A, Shen C, Halperin D, Zhao B, Zhou S, Xu Y, Shih T & Yao JC 2017 Trends in the incidence, prevalence, and survival outcomes in patients With neuroendocrine tumors in the United States. JAMA Oncology 3 . (https://doi.org/10.1001/jamaoncol.2017.0589)
Gatto F & Hofland LJ 2011 The role of somatostatin and dopamine D2 receptors in endocrine tumors. Endocrine-Related Cancer 18 R233–R2 51. (https://doi.org/10.1530/ERC-10-0334)
Guenter RE, Aweda T, Carmona Matos DM, Whitt J, Chang AW, Cheng EY, Liu XM, Chen H, Lapi SE & Jaskula-Sztul R 2019 Pulmonary carcinoid surface receptor modulation using histone deacetylase inhibitors. Cancers 11 767. (https://doi.org/10.3390/cancers11060767)
Guenter R, Aweda T, Carmona Matos DM, Jang S, Whitt J, Cheng YQ, Liu XM, Chen H, Lapi SE & Jaskula-Sztul R 2020 Overexpression of somatostatin receptor type 2 in neuroendocrine tumors for improved Ga68-DOTATATE imaging and treatment. Surgery 167 . (https://doi.org/10.1016/j.surg.2019.05.092)
Haug AR, Assmann G, Rist C, Tiling R, Schmidt GP, Bartenstein P & Hacker M 2010. Quantification of immunohistochemical expression of somatostatin receptors in neuroendocrine tumors using 68Ga-DOTATATE PET/CT. Der Radiologe 50 54. (https://doi.org/10.1007/s00117-009-1972-2)
Hicks RJ, Kwekkeboom DJ, Krenning E, Bodei L, Grozinsky-Glasberg S, Arnold R, Borbath I, Cwikla J, Toumpanakis C, Kaltsas G, et al. 2017 Enets consensus guidelines for the standards of care in neuroendocrine neoplasia: peptide receptor radionuclide therapy with radiolabeled somatostatin analogues. Neuroendocrinology 105 . (https://doi.org/10.1159/000475526)
Hofland J, Kaltsas G & de Herder WW 2020 Advances in the diagnosis and management of well-differentiated neuroendocrine neoplasms. Endocrine Reviews 41 . (https://doi.org/10.1210/endrev/bnz004)
Hope TA, Calais J, Zhang L, Dieckmann W & Millo C 2019. (111)In-pentetreotide scintigraphy versus (68)Ga-DOTATATE PET: impact on Krenning scores and effect of tumor burden. Journal of Nuclear Medicine 60 . (https://doi.org/10.2967/jnumed.118.223016)
Jin XF, Auernhammer CJ, Ilhan H, Lindner S, Nolting S, Maurer J, Spottl G & Orth M 2019 Combination of 5-fluorouracil with epigenetic modifiers induces radiosensitization, somatostatin receptor 2 expression, and radioligand binding in neuroendocrine tumor cells in vitro. Journal of Nuclear Medicine 60 . (https://doi.org/10.2967/jnumed.118.224048)
Kim HS, Lee HS & Kim WH 2011 Clinical significance of protein expression of cyclooxygenase-2 and somatostatin receptors in gastroenteropancreatic neuroendocrine tumors. Cancer Research and Treatment 43 8. (https://doi.org/10.4143/crt.2011.43.3.181)
Kwekkeboom DJ, Teunissen JJ, Bakker WH, Kooij PP, de Herder WW, Feelders RA, Van Eijck CH, Esser JP, Kam BL & Krenning EP 2005 Radiolabeled somatostatin analog [177Lu-DOTA0,Tyr3]octreotate in patients with endocrine gastroenteropancreatic tumors. Journal of Clinical Oncology 23 62. (https://doi.org/10.1200/JCO.2005.08.066)
Mehta S, De Reuver PR, Gill P, Andrici J, D’Urso L, Mittal A, Pavlakis N, Clarke S, Samra JS & Gill AJ 2015 Somatostatin receptor SSTR-2a expression is a stronger predictor for survival than Ki-67 in pancreatic neuroendocrine tumors. Medicine 94 e1281. (https://doi.org/10.1097/MD.0000000000001281)
Müssig K, Oksüz MO, Dudziak K, ueberberg B, Wehrmann M, Horger M, Schulz S, Häring HU, Pfannenberg C, Bares R, et al. 2010 Association of somatostatin receptor 2 immunohistochemical expression with [111In]-DTPA octreotide scintigraphy and [68Ga]-DOTATOC PET/CT in neuroendocrine tumors. Hormone and Metabolic Research 42 . (https://doi.org/10.1055/s-0030-1253354)
Nagtegaal ID, Odze RD, Klimstra D, Paradis V, Rugge M, Schirmacher P, Washington KM, Carneiro F, Cree IA & WHO Classification of Tumours Editorial Board 2020 The 2019 WHO classification of tumours of the digestive system. Histopathology 76 . (https://doi.org/10.1111/his.13975)
Nunez-Valdovinos B, Carmona-Bayonas A, Jimenez-Fonseca P, Capdevila J, Castano-Pascual Á, Benavent M, Pi Barrio JJ, Teule A, Alonso V, Custodio A, et al. 2018 Neuroendocrine tumor heterogeneity adds uncertainty to the World Health Organization 2010 classification: real-world data from the Spanish Tumor Registry (R-GETNE). Oncologist 23 . (https://doi.org/10.1634/theoncologist.2017-0364)
Okuwaki K, Kida M, Mikami T, Yamauchi H, Imaizumi H, Miyazawa S, Iwai T, Takezawa M, Saegusa M, Watanabe M, et al. 2013 Clinicopathologic characteristics of pancreatic neuroendocrine tumors and relation of somatostatin receptor type 2A to outcomes. Cancer 119 102. (https://doi.org/10.1002/cncr.28341)
Pavel M, O’Toole D, Costa F, Capdevila J, Gross D, Kianmanesh R, Krenning E, Knigge U, Salazar R, Pape UF, et al. 2016 Enets consensus guidelines update for the management of distant metastatic disease of intestinal, pancreatic, bronchial neuroendocrine neoplasms (NEN) and NEN of unknown primary site. Neuroendocrinology 103 85. (https://doi.org/10.1159/000443167)
Perren A, Couvelard A, scoazec JY, Costa F, Borbath I, Delle Fave G, Gorbounova V, Gross D, Grossma A, Jense RT, et al. 2017 Enets consensus guidelines for the standards of care in neuroendocrine tumors: pathology: diagnosis and prognostic stratification. Neuroendocrinology 105 . (https://doi.org/10.1159/000457956)
Qian ZR, Li T, Ter-Minassian M, Yang J, Chan JA, Brais LK, Masugi Y, Thiaglingam A, Brooks N, Nishihara R, et al. 2016 Association between somatostatin receptor expression and clinical outcomes in neuroendocrine tumors. Pancreas 45 . (https://doi.org/10.1097/MPA.0000000000000700)
Raj N & Reidy-lagunes D 2014 Current clinical trials of targeted agents for well-differentiated neuroendocrine tumors. Pancreas 43 9. (https://doi.org/10.1097/MPA.0000000000000232)
Reubi JC 2007 Peptide receptor expression in GEP-NET. Virchows Archiv 451 (Supplement 1) S47–S 50. (https://doi.org/10.1007/s00428-007-0443-2)
Reubi JC, Waser B, Cescato R, Gloor B, Stettler C & Christ E 2010 Internalized somatostatin receptor subtype 2 in neuroendocrine tumors of octreotide-treated patients. Journal of Clinical Endocrinology and Metabolism 95 50. (https://doi.org/10.1210/jc.2009-2487)
Rinke A, Muller HH, Schade-Brittinger C, Klose KJ, Barth P, Wied M, Mayer C, Aminossadati B, Pape UF, Blaker M, et al. 2009 Placebo-controlled, double-blind, prospective, randomized study on the effect of octreotide LAR in the control of tumor growth in patients with metastatic neuroendocrine midgut tumors: a report from the PROMID Study Group. Journal of Clinical Oncology 27 63. (https://doi.org/10.1200/JCO.2009.22.8510)
Song KB, Kim SC, Kim JH, seo DW, Hong SM, Park KM, Hwang DW, Lee JH & Lee YJ 2016 Prognostic value of somatostatin receptor subtypes in pancreatic neuroendocrine tumors. Pancreas 45 92. (https://doi.org/10.1097/MPA.0000000000000493)
Strosberg J, El-Haddad G, Wolin E, Hendifar A, Yao J, Chasen B, Mittra E, Kunz PL, Kulke MH, Jacene H, et al. 2017 Phase 3 trial of (177)Lu-Dotatate for midgut neuroendocrine tumors. New England Journal of Medicine 376 . (https://doi.org/10.1056/NEJMoa1607427)
Sundin A, Arnold R, Baudin E, Cwikla JB, Eriksson B, fanti S, Fazio N, Giammarile F, Hicks RJ, Kjaer A, et al. 2017 Enets consensus guidelines for the standards of care in neuroendocrine tumors: radiological, nuclear medicine and hybrid imaging. Neuroendocrinology 105 . (https://doi.org/10.1159/000471879)
Taelman VF, Radojewski P, Marincek N, Ben-Shlomo A, Grotzky A, Olariu CI, Perren A, Stettler C, Krause T, Meier LP, et al. 2016 Upregulation of key molecules for targeted imaging and therapy. Journal of Nuclear Medicine 57 . (https://doi.org/10.2967/jnumed.115.165092)
Tierney JF, Chivukula SV, Wang X, Pappas SG, Schadde E, Hertl M, poirier J & Keutgen XM 2019 Resection of primary tumor may prolong survival in metastatic gastroenteropancreatic neuroendocrine tumors. Surgery 165 . (https://doi.org/10.1016/j.surg.2018.09.006)
Van Adrichem RC, Kamp K, Van Deurzen CH, Biermann K, Feelders RA, Franssen GJ, Kwekkeboom DJ, Hofland LJ & De Herder WW 2016 Is there an additional value of using somatostatin receptor subtype 2a immunohistochemistry compared to somatostatin receptor scintigraphy uptake in predicting gastroenteropancreatic neuroendocrine tumor response? Neuroendocrinology 103 6. (https://doi.org/10.1159/000441604)
Veenstra MJ, Van Koetsveld PM, Dogan F, Farrell WE, Feelders RA, Lamberts SWJ, De herder WW, Vitale G & Hofland LJ 2018 Epidrug-induced upregulation of functional somatostatin type 2 receptors in human pancreatic neuroendocrine tumor cells. Oncotarget 9 . (https://doi.org/10.18632/oncotarget.9462)
Vesterinen T, Leijon H, Mustonen H, Remes S, Knuuttila A, Salmenkivi K, Vainio P, Arola J & Haglund C 2019 Somatostatin receptor expression is associated with metastasis and patient outcome in pulmonary carcinoid tumors. Journal of Clinical Endocrinology and Metabolism 104 . (https://doi.org/10.1210/jc.2018-01931)
Wanek J, Gaisberger M, Beyreis M, Mayr C, Helm K, Primavesi F, Jager T, Di Fazio P, Jakab M, wagner A, et al. 2018 Pharmacological inhibition of class IIA HDACs by LMK-235 in pancreatic neuroendocrine tumor cells. International Journal of Molecular Sciences 19 3128. (https://doi.org/10.3390/ijms19103128)
Werner RA, Solnes LB, Javadi MS, Weich A, Gorin MA, Pienta KJ, Higuchi T, Buck AK, Pomper MG, Rowe SP, et al. 2018 SSTR-RADS version 1.0 as a reporting system for SSTR PET imaging and selection of potential PRRT candidates: a proposed standardization framework. Journal of Nuclear Medicine 59 . (https://doi.org/10.2967/jnumed.117.206631)
Zandee WT, Kamp K, Van Adrichem RC, Feelders RA & De Herder WW 2017 Effect of hormone secretory syndromes on neuroendocrine tumor prognosis. Endocrine-Related Cancer 24 R261–R274. (https://doi.org/10.1530/ERC-16-0538)
