SSTR2 scoring and distribution of SSTR2 expression among samples
Representative images of SSTR2 expression score are presented in Figure 2 A) Amount of Low SSTR2 (SSTR2 score 1 and 2) samples in the entire TMA, cohort A and cohort B were similar (Figure 2 B). In 95 samples from the 44 PRRT treated patients (cohort A), 16 samples (16.8 %) had no or low STTR2 expression (score 0 or 1) (Figure 1 B). The majority of samples (n= 79) had medium or high SSTR2 expression (score 2 or 3). In 102 samples from the 34 patients in cohort B (paired samples from primary tumour, lymph node and hepatic metastases), 23 samples (23%) had a low STTR2 expression (score 0 or 1).
To investigate if the SSTR2 expression differed depending on tumour site the samples from cohort B (n=102), with paired samples from three separate locations, were analysed. SSTR2 expression levels did not vary significantly between primary, lymph node or hepatic metastases (average score Primary tumour 2.18 vs average score Lymph node metastases 2.03 vs average score Liver metastases 2.24, n=34 for all groups, p=ns).
Further, to determine if SSTR2 expression was consistent among all lesions in a patient, all samples from cohort B were sorted according to SSTR2 expression in the primary tumour. Three groups of primary tumours (score 1-3, no primary tumour had score 0) were established and SSTR2 expression in corresponding metastases was assessed (Figure 2 C). SSTR2 expression was significantly different in metastases when sorted according to SSTR2 expression in the primary tumors (Group 1-3) ( Lymph node metastases group 1 vs Lymph node metastases group 2, p=**, Lymph node metastases group 1 vs Lymph node metastases group 3, p=***, and Liver metastases group 1 vs Liver metastases group 3 p=** .
SSTR2 expression and Ki-67
To determine the association between SSTR2 and Ki-67%, Ki-67% was assessed in 70 samples from cohort A. Samples were grouped according to SSTR2 expression and Ki-67% in corresponding samples was analysed. There was a non-significant tendency towards higher Ki-67% in tumours with SSTR2 score 2-3 (p=ns) (Figure 1 C)
SSTR2 expression does not correlate with synaptophysin or chromogranin A expression
We also assessed if SSTR2 expression was associated with the IHC expression of established SI-NET markers synaptohysin and chromogranin A (CgA). A subset of samples (n=49) from PRRT treated patients (cohort A) were divided into groups based on SSTR2 expression, and synaptophysin and CgA expression for each sample was assessed. IHC expression of synaptophysin and CgA was consistent among samples, regardless of SSTR2 expression (Figure 1 D)
SSTR2 expression and activity concentration
In PRRT treated patients (cohort A) an estimation of the uptake of radionuclide was done by measuring the activity concentration in 3 tumours per patient (in two patients only 2 tumours were measured). Measurements were possible in 33 patients: 27 with high and 6 with low SSTR2 expressing tumours, respectively. Results showed a large variability between treated patients but consistency among the tumours within patients. There was a tendency towards higher uptake in the Low SSTR2 group but this was not statistically significant (p=0.06) (Figure 4 D). There was no correlation between average tumour activity concentration and average Ki-67% in PRRT treated patients (Figure 4 C).
SSTR2 expression and long-term outcome
The patients in cohort A and cohort B were divided into two groups based on SSTR2 expression. Patients who had at least one lesion with a score of 0 or 1 were assigned to the “Low SSTR2” group. The remaining patients, with a score of 2 or 3 in every sample (n=32), were assigned to the “High SSTR2” group. Patient characteristics are presented in table 1 and table 2.
4 patients had lesions with a Ki-67% over 10%, these patients all were assigned to the “High SSTR2” group. When the sample with the highest Ki-67% for each patient was used, patients with low SSTR2 expressing tumours had significantly lower Ki-67% than patients with high SSTR2 expression (p=0.049).
Both cohorts A and B were analysed regarding long-term outcome and SSTR2 expression. PRRT treated patients in cohort A included 10 patients in the Low SSTR2 group and 32 patients in the High SSTR2 group. Among non-PRRT treated patients (cohort B) 11 patients had low SSTR2 expressing tumours and 17 had high SSTR2 expressing tumours. Kaplan-Meier curves depicting overall survival (OS) based on SSTR2 expression in these two patient cohorts show that there was no statistically significant difference between the groups (p = 0.12 for cohort A; p = 0.11 for cohort B) (Fig 3). Although there was no statistical difference, we observed a trend towards longer OS in the Low SSTR2 group.
SSTR2 expression and treatment patterns
In cohort A all patients received PRRT. After PRRT, patients continued SSA, and were offered additional treatments in case of progression. Eight of the 32 patients (25%) with high SSTR2 expression received additional treatment after initial PRRT. None of the patients (0/10) with low SSTR2 expression received additional treatment during follow-up, Table 3.
Among the non-PRRT treated patients all but two patients, with no evident signs of disease, were treated with SSA after surgery. The main additional treatment was hepatic artery embolization (n=16), followed by interferon-a (n=5) (Table 4).