Abstract
Purpose: Although somatostatin analogues (SSA) and peptide receptor radionuclide therapy (PRRT) are validated therapies in patients with advanced gastroenteropancreatic neuroendocrine tumors (GEP-NET), it remains unclear whether SSA combined with PRRT or as maintenance therapy can provide prolonged survival compared with patients treated with PRRT alone. In this retrospective study, we aimed to investigate whether there is a survival benefit to adding SSA to PRRT as a combination therapy and/or maintenance therapy.
Patients and Methods: The investigation included 168 patients with unresectable GEP-NETs treated at the University Hospital Bonn, Bonn, Germany. The patients were divided into two main groups: PRRT monotherapy (N = 81, group 1) and PRRT plus SSA (N = 87, group 2) as combined therapy with PRRT and/or as maintenance therapy after PRRT.
Results: Data for overall survival (OS) were available from 168 patients, of whom 160 had data for progression-free survival (PFS). The median PFS was 27 months in group 1 versus 48 months in group 2 (P = 0.012). The median OS rates were 47 months in group 1 and 91 months in group 2 (P < 0.001). The death-event rates were lower in group 2 (26%) than in group 1 (63%). SSA as a combination therapy with PRRT and/or as a maintenance therapy showed a clinical benefit rate (objective response or stable disease) of 95%, which was significantly higher than group 1 (79%).
Conclusions: SSA as a combination therapy and/or maintenance therapy may play a significant role in tumor control in patients with GEP-NET who underwent a PRRT. Clin Cancer Res; 24(19); 4672–9. ©2018 AACR.
Results of the recent NETTER-1 trial suggest that patients with advanced midgut neuroendocrine tumors (midgut-NET) treated with peptide receptor radionuclide therapy (PRRT) in combination with somatostatin analogues (SSA) had a better survival than the cohort treated with SSA alone. However, it is unclear whether the combination of PRRT and SSAs provides a survival benefit compared with PRRT alone.
Here, we retrospectively examined the survival and response rate of patients treated with SSA as a combination therapy with PRRT and/or as a maintenance therapy after PRRT. In this analysis, a significantly better progression-free survival, overall survival, and lower death-event rates could be achieved with the combination of SSA and PRRT in patients with advanced gastroenteropancreatic neuroendocrine tumors (GEP-NET), compared with cohort of patients treated with PRRT alone. SSA as a combination therapy and/or maintenance therapy may play a significant role in tumor control in patients with GEP-NET who underwent a PRRT.
Introduction
Therapy management of patients with gastroenteropancreatic neuroendocrine tumors (GEP-NET) can be challenging because of the heterogeneous biology and differing malignancy potentials of the various tumor entities (1–3). The complexity of NETs requires close cooperation within an interdisciplinary team (4). The main components of NET therapy are surgery, somatostatin analogues (SSA; long-acting Octreotide and Lanreotide), peptide receptor radionuclide therapy (PRRT), targeted therapy with everolimus or sunitinib, chemotherapy, and local ablative therapies (4–9).
Well-differentiated GEP-NETs usually overexpress somatostatin receptors (SSTRs; refs. 10, 11). SST is a hypothalamic peptide that downregulates pituitary-growth hormone secretion, exerts antiproliferative effects, and regulates the immune system. SSTRs are widely expressed in the body, including in pancreatic islet cells, the gastrointestinal tract, the endocrine system, and the hematopoietic and immune organs (12–17).
SSTRs are usually homogeneously localized throughout GEP-NETs, thus making them efficient targets of SSA therapies. SST has a very short half-life in vivo and hence cannot be used for therapy. Therefore, stable synthetic SSAs, such as long-acting Octreotide and Lanreotide, have been developed for use in therapy.
SSAs have been approved for treating functioning neuroendocrine tumors showing hormone-related symptoms (18, 19). Results of the prospective, randomized, double-blind, placebo-controlled phase III trials PROMID (on long-acting Octreotide) and CLARINET (on Lanreotide) showed that SSAs were efficient for treating carcinoid syndrome and exerted antitumor effects, with a positive impact on the progression-free survival (PFS) of patients with intestinal and pancreatic NETs (18–22).
Radioactively labeled SSAs have been developed for treating NETs. Results of a recent NETTER-1 trial, the first randomized phase III trial for evaluation of the effects of PRRT, showed the advantage of combination therapy with PRRT using [177Lu]Lu-Octreotate + long-acting Octreotide (30 mg) over monotherapy with long-acting Octreotide (60 mg) for treating patients with unresectable metastatic midgut NETs. Patients who received PRRT did not reach the median PFS (65.2% progression free after 20 months); the PFS of patients in the control group was 8.4 months. These results confirm that radioactively labeled SSAs have better antitumor activity than radioactively unlabeled SSAs. Results of the NETTER-1 trial suggest that PRRT can be used as second-line therapy for treating patients with advanced NETs (23). However, it is unclear whether the combination of PRRT and SSAs provides a survival benefit compared with PRRT alone.
The addition of SSAs to PRRT may enhance its antineoplastic effect, and maintenance therapy with SSA after PRRT may prolong the duration of response. To our knowledge, no study has examined the use of SSAs as maintenance therapy in patients with advanced metastatic GEP-NETs. Therefore, we retrospectively examined the survival and response rate of patients treated with SSA as a combination therapy with PRRT and/or as a maintenance therapy after PRRT.
Patients and Methods
Patients
The inclusion criteria for this study were: having a histologically confirmed well-differentiated GEP neuroendocrine tumor (G1–G2); multiple unresectable metastases; adequate tracer accumulation on the SSTR imaging (Ga-DOTATOC-PET/CT images or alternatively Octreoscan); and Eastern Co-operative Oncology Group (ECOG) performance status ≤2 (Karnofsky performance status ≥60). Furthermore, the follow-up interval from the beginning of the therapy (overall follow-up) should be at least 6 months.
We divided the patients into two groups. The first group included patients who underwent a PRRT as a monotherapy. These patients received no combination or maintenance therapy with SSA. In group 2, patients received SSA as a combination therapy with PRRT and/or as a maintenance therapy after PRRT. For purposes of clarity, pretreatment with SSA was allowed in both groups. An overview of both groups is shown in Supplementary Fig. S1. Patients who could not be clearly classified into these groups were excluded. These were patients who either discontinued the combination therapy and received a PRRT monotherapy afterward or who initially underwent PRRT monotherapy followed by a combination therapy with SSA. This retrospective study was approved by the local Ethics Committee and were in accordance with the Declaration of Helsinki, International Ethical Guidelines for Biomedical Research Involving Human Subjects (CIOMS), and comparable ethical standards. Informed consent was obtained from all patients included in the study.
Treatment
All patients received a PRRT with [177Lu]Lu-Octreotate. Labeling of the peptide with Lutetium-177 (IDB Holland B. V., Baarle-Nassau, the Netherlands) was performed locally. The patients received a nephroprotective infusion of amino acids (2.5% lysine and 2.5% arginine in 1 L 0.9% NaCl) for over 4 hours, starting 30 minutes before therapy. To avoid nausea due to the amino-acids infusions, the patients were premedicated with ondansetron (8 mg i.v.). The radiopharmaceutical was administered via an indwelling catheter over a maximum of 30 minutes. The median cumulative administered activity was 30.0 GBq in group 1 and 32.4 GBq in group 2, with a median dose of 7.4 GBq and 7.1 GBq per cycle in groups 1 and 2, respectively. Both groups received a median of four cycles of PRRT. The administered activity was adapted individually and lowered in patients with reduced blood counts or abnormal renal parameters.
The combined therapy consisted of PRRT (repeated at 2–3-month intervals) and SSA-injections (long-acting Octreotide or Lanreotide every 4 weeks). To avoid interference with PRRT, short-acting SSA was discontinued for one day and long-acting release formulations for 4–6 weeks before PRRT. SSA dosages varied between the different formulations: long-acting Octreotide (median 30 mg, range 20–40 mg) or Lanreotide (median 120 mg, range 60–120 mg). The maintenance therapy consisted of monthly applications of SSA: long-acting Octreotide (median 30 mg, range 10–50 mg) or Lanreotide (median 120 mg, range 30–120 mg). Because of severe carcinoid syndrome, nine patients received a short-acting SSA (median 600 μg/day, range 200–2,000 μg/day) during the combination therapy and three patients during the maintenance treatment (median 600 μg/day).
Disease monitoring
For evaluation of the response to the treatment, we used laboratory data including assessment of Chromogranin A (CgA) and Neuron Specific Enolase (NSE) levels and SSTR-imaging at stated follow-up intervals (every six months beginning 3 months after the last therapy, with eventually annual examinations).
For the biochemical response, an increase or decrease of ≥25% was considered as clinically significant. For this calculation, we compared the baseline tumor marker with post-therapeutic values and follow-up data within the first year. An increase of ≥25% of the tumor-marker levels within the first 6 months after the last PRRT application was regarded as biochemical progress.
The imaging response was assessed using Ga-68-DOTATOC-PET/CT and classified according to adapted WHO criteria for CT and SSTR-status in PET. The findings were graduated as follows: complete response (CR), partial response (PR), minimal response (MR), stable disease (SD), and progressive disease (PD). The overall response rate included CR + PR + MR; the clinical benefit rate (CBR) was defined as CR + PR + MR + SD.
Statistical analyses
Patient data were gathered in a database. Descriptive statistics, frequency analyses, and significance tests were performed using SPSS Statistics 22.0 (IBM Corp.). To explore the significance of the results, we conducted the chi-square (χ2) test. Results with a P-value <0.05 were considered as statistically significant.
Survival rates were estimated using the Kaplan–Meier method. Included were only patients with a minimum overall follow-up of 6 months. Overall survival (OS) and PFS in the patients were compared in the two main groups. Patients with no documented progression in the follow-up were censored for PFS and OS. Furthermore, we compared OS and PFS in subgroups of patients according to primary tumor entity, tumor functionality, tumor burden, and Ki67-index, conducting the log-rank test and the χ2 test. Response rates (imaging and biochemical) in the two groups were also compared using the χ2 test. Finally, group 2 was divided into two subgroups and then compared: SSA maintenance therapy after PRRT (group 2A) and combined therapy with PRRT and SSA followed by maintenance therapy (group 2B).
Results
Patients
After retrospective evaluation of 421 patients with GEP-NET who were treated with PRRT in the period between April 2004 and July 2017, 168 patients (89 males, 79 females) were eligible according to our inclusion criteria. Supplementary Fig. S1 is represented a diagram illustrating the process of patients' selection. Patients who did not meet the inclusion criteria were excluded. These were mainly patients with pulmonary NET, NET with unknown primary, G3-tumors and patients with overall-follow up of <6 months. The patients' characteristics at baseline are summarized in Table 1. To assess the effect of SSA in combination with PRRT, we divided the patients into two groups. Group 1 included 81 patients who were treated with a PRRT monotherapy and received no maintenance treatment with SSA. Group 2 included 87 patients who either received SSA combined with PRRT (77.5% long-acting Octreotide, 20.5% Lanreotide, and 2.5% initially long-acting Octreotide, followed by Lanreotide mainly because of side effects) followed by a maintenance therapy with long-acting Octreotide or patients who received a maintenance therapy with SSA (64.9% long-acting Octreotide, 31.2% Lanreotide, and 3.9% short-acting octreotide) after PRRT. For purposes of clarity, pretreatment with SSA was allowed in all groups, because this factor statistically did not significantly influence the outcome of this study. The groups did not significantly differ in respect to age (mean values of group 1 = 63.1 and group 2 = 63.1, P = 0.68), gender distribution (group 1 male: female = 1.5, group 2 male: female = 1.4, P = 0.32), Ki67-index (mean values of group 1 = 5.8 and group 2 = 5.4%, P = 0.56), ECOG status (both groups median ECOG 1, P = 0.11). Also, the difference in tumor burden, spread of the tumor and the tumor-type distribution between groups 1 and 2 was not significant. Furthermore, about the same number of patients in group 1 (N = 58, 72%) and group 2 (N = 66, 76%) were progressive prior to PRRT. However, regarding the treatment previous to PRRT, there were significantly more patients from group 2 (64%) who received SSA than from group 1 (49%). From the patients, who received previously SSA 87.9% from group 1 and 92.9% from group 2 were progressive prior to PRRT, respectively. The difference was not statistically significant.
Patients' characteristics
. | All pts . | Group 1 . | Group 2 . | . |
---|---|---|---|---|
Characteristic . | n (%) . | n (%) . | n (%) . | P(χ2) . |
Sex | ||||
Male | 89 (53.0) | 40 (49.4) | 49 (56.3) | n.s. |
Female | 79 (47.0) | 41 (50.6) | 38 (43.7) | |
Age | ||||
Mean | 63.1 | 63.1 | 63.1 | n.s. |
Patients ≤65 | 92 (54.8) | 43 (53.1) | 49 (56.3) | |
Patients >65 | 76 (45.2) | 38 (46.9) | 38 (43.7) | |
EGOG-index | ||||
Median | 1 | 1 | 1 | n.s. |
ECOG 0 | 34 (20.2) | 11 (13.6) | 23 (26.4) | |
ECOG 1 | 114 (67.9) | 60 (74.1) | 54 (62.1) | |
ECOG 2 | 20 (11.9) | 10 (12.3) | 10 (11.5) | |
Type of GEP-NET | ||||
Pancreas | 84 (50) | 48 (59.3) | 36 (41.4) | n.s. |
Midgut | 46 (27.4) | 18 (22.2) | 28 (32.2) | |
Others | 38 (22.6) | 15 (18.5) | 23 (26.4) | |
Tumor functionality | ||||
Functioning | 99 (58.9) | 44 (54.3) | 55 (63.2) | n.s. |
Nonfunctioning | 69 (41.1) | 37 (45.7) | 32 (36.8) | |
Ki67 | ||||
Mean | 5.6 | 5.8 | 5.4 | n.s. |
<10% | 113 (76.9) | 54 (76.1) | 59 (77.6) | |
≥10 to ≤20% | 34 (23.1) | 17 (23.9) | 17 (22.4) | |
Tumor burdena | ||||
Low/moderate | 98 (58.7) | 47 (58.0) | 51 (59.3) | n.s. |
High | 69 (41.3) | 34 (42.0) | 35 (40.7) | |
Tumor marker (mean/median) | ||||
Cg A | 2141/429 | 2699/450 | 1621/421 | n.s. |
NSE | 25.7/18.9 | 28.2/21.7 | 23.3/16.2 | n.s. |
Metastases of the tumor | ||||
Liver metastases | 154 (91.7) | 73 (90.1) | 81 (93.1) | n.s. |
Bone metastases | 71 (42.3) | 28 (34.6) | 43 (49.4) | n.s. |
Pulmonal metastases | 9 (5.4) | 3 (3.7) | 6 (6.9) | n.s. |
Lymph node metastases | 110 (65.5) | 52 (64.2) | 58 (66.7) | n.s. |
Other metastases | 40 (23.8) | 14 (17.3) | 26 (29.9) | n.s. |
Progress prior to PRRT | ||||
Progress | 124 (73.8) | 58 (71.6) | 66 (75.9) | n.s. |
Stable disease | 44 (26.2) | 23 (28.4) | 21 (24.1) | |
Treatment previous to PRRT | ||||
Surgery | 99 (58.9) | 50 (61.7) | 49 (56.3) | n.s. |
SSA | 96 (57.1) | 40 (49.4) | 56 (64.4) | P = 0.05 |
Progress prior SSA | 9 (9.4) | 5 (12.5) | 4 (7.1) | n.s. |
Stable disease prior SSA | 87 (90.6) | 35 (87.5) | 52 (92.9) | n.s. |
Chemotherapy | 33 (19.6) | 16 (19.8) | 17 (19.5) | n.s. |
Radiotherapy | 7 (4.2) | 3 (3.7) | 4 (4.6) | n.s. |
Other RNT | 8 (4.8) | 2 (2.5) | 6 (6.9) | n.s. |
LAP | 27 (16.1) | 12 (14.8) | 15 (17.2) | n.s. |
. | All pts . | Group 1 . | Group 2 . | . |
---|---|---|---|---|
Characteristic . | n (%) . | n (%) . | n (%) . | P(χ2) . |
Sex | ||||
Male | 89 (53.0) | 40 (49.4) | 49 (56.3) | n.s. |
Female | 79 (47.0) | 41 (50.6) | 38 (43.7) | |
Age | ||||
Mean | 63.1 | 63.1 | 63.1 | n.s. |
Patients ≤65 | 92 (54.8) | 43 (53.1) | 49 (56.3) | |
Patients >65 | 76 (45.2) | 38 (46.9) | 38 (43.7) | |
EGOG-index | ||||
Median | 1 | 1 | 1 | n.s. |
ECOG 0 | 34 (20.2) | 11 (13.6) | 23 (26.4) | |
ECOG 1 | 114 (67.9) | 60 (74.1) | 54 (62.1) | |
ECOG 2 | 20 (11.9) | 10 (12.3) | 10 (11.5) | |
Type of GEP-NET | ||||
Pancreas | 84 (50) | 48 (59.3) | 36 (41.4) | n.s. |
Midgut | 46 (27.4) | 18 (22.2) | 28 (32.2) | |
Others | 38 (22.6) | 15 (18.5) | 23 (26.4) | |
Tumor functionality | ||||
Functioning | 99 (58.9) | 44 (54.3) | 55 (63.2) | n.s. |
Nonfunctioning | 69 (41.1) | 37 (45.7) | 32 (36.8) | |
Ki67 | ||||
Mean | 5.6 | 5.8 | 5.4 | n.s. |
<10% | 113 (76.9) | 54 (76.1) | 59 (77.6) | |
≥10 to ≤20% | 34 (23.1) | 17 (23.9) | 17 (22.4) | |
Tumor burdena | ||||
Low/moderate | 98 (58.7) | 47 (58.0) | 51 (59.3) | n.s. |
High | 69 (41.3) | 34 (42.0) | 35 (40.7) | |
Tumor marker (mean/median) | ||||
Cg A | 2141/429 | 2699/450 | 1621/421 | n.s. |
NSE | 25.7/18.9 | 28.2/21.7 | 23.3/16.2 | n.s. |
Metastases of the tumor | ||||
Liver metastases | 154 (91.7) | 73 (90.1) | 81 (93.1) | n.s. |
Bone metastases | 71 (42.3) | 28 (34.6) | 43 (49.4) | n.s. |
Pulmonal metastases | 9 (5.4) | 3 (3.7) | 6 (6.9) | n.s. |
Lymph node metastases | 110 (65.5) | 52 (64.2) | 58 (66.7) | n.s. |
Other metastases | 40 (23.8) | 14 (17.3) | 26 (29.9) | n.s. |
Progress prior to PRRT | ||||
Progress | 124 (73.8) | 58 (71.6) | 66 (75.9) | n.s. |
Stable disease | 44 (26.2) | 23 (28.4) | 21 (24.1) | |
Treatment previous to PRRT | ||||
Surgery | 99 (58.9) | 50 (61.7) | 49 (56.3) | n.s. |
SSA | 96 (57.1) | 40 (49.4) | 56 (64.4) | P = 0.05 |
Progress prior SSA | 9 (9.4) | 5 (12.5) | 4 (7.1) | n.s. |
Stable disease prior SSA | 87 (90.6) | 35 (87.5) | 52 (92.9) | n.s. |
Chemotherapy | 33 (19.6) | 16 (19.8) | 17 (19.5) | n.s. |
Radiotherapy | 7 (4.2) | 3 (3.7) | 4 (4.6) | n.s. |
Other RNT | 8 (4.8) | 2 (2.5) | 6 (6.9) | n.s. |
LAP | 27 (16.1) | 12 (14.8) | 15 (17.2) | n.s. |
aIndex: low/moderate tumor burden: few metastases (<10) and A <600 ng/mL and liver involvement <50%; high tumor burden: multiple metastases (>10) or chromogranin A >600 ng/mL or liver involvement >50%; CgA, chromogranin A; NSE, neuron-specific Enolase; pts., patients; n.s., not significant (P > 0.05); RNT, radionuclide therapies, such as [90Y]Y-DOTATOC or [131I]I-MIBG; LAP, local ablative procedures.
Survival
One hundred and sixty-eight patients were assessable for survival analyses. The median follow-up period from start of treatment was 43.5 months (range 7–147 months). The median follow-up period after the last therapy (at least one treatment cycle) was 24 months (range 2–109 months). The median PFS was 27 months in group 1 (PRRT monotherapy) and 48 months in group 2 (SSA combined with PRRT and/or maintenance SSA treatment after PRRT). The median OS values were 47 and 91 in groups 1 and 2, respectively. The difference in survival between the groups was significant: PFS (log rank) P = 0.012 and OS (log rank) P < 0.001 (χ2 = 21.63). An overview of the survival data via Kaplan–Meier curves is shown in Fig. 1.
Survival analyses (Kaplan–Meier curves) of groups 1 and 2. The PFS was calculated for 160 patients: 75 in group 1 (PRRT* as monotherapy), 85 in group 2 (combined therapy PRRT + SSA** with/or maintenance therapy). OS data were available from 168 patients: 81 in group 1 and 87 in group 2. *, PRRT; **, SSAs.
Survival analyses (Kaplan–Meier curves) of groups 1 and 2. The PFS was calculated for 160 patients: 75 in group 1 (PRRT* as monotherapy), 85 in group 2 (combined therapy PRRT + SSA** with/or maintenance therapy). OS data were available from 168 patients: 81 in group 1 and 87 in group 2. *, PRRT; **, SSAs.
Subgroup analyses
We identified three subgroups of patients who showed the most significant survival benefit from SSA as a maintenance or combination therapy. The first subgroup comprised patients with Ki67 ≥10%, who achieved a prolonged PFS compared with the patients without any SSA treatment (median PFS of group 2: 42 months vs. 17 months in group 1, P = 0.004) and had a greater OS (median OS of group 2: 90 months vs. 32 months in group 1, P < 0.001, χ2 = 20.34). The second subgroup was of patients with a high tumor burden, which was defined as the presence of >10 metastases and/or chromogranin A >600 ng/mL or liver involvement >50%. Both PFS (median of group 1: 23 vs. 39 in group 2, P = 0.014) and OS (median of group 1: 34 months vs. 69 months in group 2, P < 0.001) were longer with SSA as a maintenance or combination therapy. In general, nonfunctioning tumors had a better survival than functioning tumors. In both functioning and nonfunctioning tumors, we could see a survival benefit from SSA as a combination therapy with PRRT or as a maintenance therapy. However, the most significant survival benefit from SSA as a maintenance or combination therapy was in patients with functioning tumors (PFS, P = 0.01; OS, P < 0.001). Patient characteristics of the three subgroups are listed in Supplementary Table S1.
In the subgroup analyses, we also tested the effectiveness of the SSA among the various tumor subtypes. We selected the four largest groups of tumor entities in our patient collective: pancreas (n = 81), midgut (n = 18), duodenum (n = 11), and cecum/ileocecal valve (n = 9). There was no significant difference in the survival distribution of PFS or OS among the subgroups. The poorer survival was in those patients with a PRRT monotherapy without maintenance therapy with SSA.
Table 2 and Figure 2 illustrate the survival of the three subgroups of patients that showed the most significant survival benefit from SSA as a maintenance or combination therapy: NET with Ki67 ≥10%, patients with a high tumor burden and patients with functioning tumors.
Subgroup survival analyses
. | Ki67 ≥10% . | Pa . | High tumor burdena . | Pa . | Functioning tumors . | Pa . | |||
---|---|---|---|---|---|---|---|---|---|
Group | 1 | 2 | 1 | 2 | 1 | 2 | |||
Median PFS (months) | 17 | 42 | 0.004 | 23 | 39 | 0.014 | 27 | 42 | 0.010 |
Median overall survival (months) | 32 | 90 | <0.001 | 34 | 69 | <0.001 | 47 | 88 | <0.001 |
. | Ki67 ≥10% . | Pa . | High tumor burdena . | Pa . | Functioning tumors . | Pa . | |||
---|---|---|---|---|---|---|---|---|---|
Group | 1 | 2 | 1 | 2 | 1 | 2 | |||
Median PFS (months) | 17 | 42 | 0.004 | 23 | 39 | 0.014 | 27 | 42 | 0.010 |
Median overall survival (months) | 32 | 90 | <0.001 | 34 | 69 | <0.001 | 47 | 88 | <0.001 |
aHigh tumor burden: presence of >10 metastases or chromogranin A >600 ng/mL or liver involvement >50%
P = P-value based on the Mantel–Cox log-rank test, P < 0.05 is regarded as significant.
Survival analyses (Kaplan–Meier curves) of subgroups regarding Ki67, tumor burden, and tumor functionality. PFS of the both main groups of patients (group 1 and 2*) divided into three subgroups: patients with Ki67 <10% vs. ≥10%, patients with low/moderate vs. high tumor burden**, and patients with nonfunctioning vs. functioning tumors. *, Group 1: PRRT monotherapy. Group 2: combined therapy PRRT + SSAs and/or maintenance therapy with SSA. **, High tumor burden: the presence of >10 metastases or chromogranin A >600 ng/mL or liver involvement >50%.
Survival analyses (Kaplan–Meier curves) of subgroups regarding Ki67, tumor burden, and tumor functionality. PFS of the both main groups of patients (group 1 and 2*) divided into three subgroups: patients with Ki67 <10% vs. ≥10%, patients with low/moderate vs. high tumor burden**, and patients with nonfunctioning vs. functioning tumors. *, Group 1: PRRT monotherapy. Group 2: combined therapy PRRT + SSAs and/or maintenance therapy with SSA. **, High tumor burden: the presence of >10 metastases or chromogranin A >600 ng/mL or liver involvement >50%.
To compare the various treatment protocols of group 2, we divided the patients into two subgroups: SSA maintenance therapy after PRRT (group 2A) and combined therapy with PRRT and SSA followed by a maintenance therapy (group 2B). There was no significant difference in the median PFS: 48 months in group 2A versus 44 months in group 2B. However, the better median OS was in group 2A: 110 months versus 68 months in group 2B (P = 0.02). Therefore, we tested whether group 2A and 2B had similar patient and tumor characteristics. There were no significant differences in gender distribution, age, Ki67-index, ECOG, tumor origin, tumor spread or progress prior to PRRT, and in the majority of previous therapies. However, the main significant difference between those groups was that more patients from group 2B had previous SSA and eventually progressed than from group 2A: 31 patients vs. 25 patients (P = 0.001). Furthermore, no patient in group 2A had another radionuclide therapy prior to PRRT, in contrast to group 2B: six patients (P = 0.003). For better clarity, we omitted a further representation of other subgroups.
Imaging response rate
One hundred and fifty-nine patients were assessable for imaging response. Figure 3 shows the response rates of the three groups. In the first group, 40.0% of patients responded: 0.0% CR, 30.7% PR, and 9.3% MR. Another 38.7% of patients showed a stable disease; the CBR was 78.7%. The response rates in group 2 (OR: 63.1%; CBR: 95.2%) were higher than in group 1. The survival data and response rates are summarized in Table 3.
Tumor response in the imaging according to modified WHO criteria in CT and SSTR-status in PET. Response rate in groups 1 and 2. CR, no tumor detection in the imaging; PR, ≥50% decrease of the tumor burden; MR, ≥25% to <50% decrease of the tumor burden; SD, <25% change of the tumor burden; PD, ≥25% increase of the tumor burden.
Tumor response in the imaging according to modified WHO criteria in CT and SSTR-status in PET. Response rate in groups 1 and 2. CR, no tumor detection in the imaging; PR, ≥50% decrease of the tumor burden; MR, ≥25% to <50% decrease of the tumor burden; SD, <25% change of the tumor burden; PD, ≥25% increase of the tumor burden.
Overview of survival analyses and response rate
. | Group 1 . | Group 2 . | P-value . |
---|---|---|---|
Median PFS (months) | 27 | 48 | 0.012a |
Median overall survival (months) | 47 | 91 | <0.001a |
Objective response rate (%) | 40.0 | 63.1 | Difference in response 0.008b |
Stable disease (%) | 38.7 | 32.1 | |
Clinical benefit rate (CBR) (%) | 78.7 | 95.2 | CBR vs. PD 0.002b |
Progressive disease (PD) (%) | 21.3 | 4.8 | |
Death events (%) | 63.0% | 35.6% | <0.001b |
. | Group 1 . | Group 2 . | P-value . |
---|---|---|---|
Median PFS (months) | 27 | 48 | 0.012a |
Median overall survival (months) | 47 | 91 | <0.001a |
Objective response rate (%) | 40.0 | 63.1 | Difference in response 0.008b |
Stable disease (%) | 38.7 | 32.1 | |
Clinical benefit rate (CBR) (%) | 78.7 | 95.2 | CBR vs. PD 0.002b |
Progressive disease (PD) (%) | 21.3 | 4.8 | |
Death events (%) | 63.0% | 35.6% | <0.001b |
aP-value based on the Mantel–Cox log-rank test.
bP-value based on the chi-square test.
P < 0.05 is regarded as significant.
Biochemical response
The mean CgA levels before the therapy and after the first follow-up (1–3 months after the end of treatment) were 2,699 versus 3,928 ng/mL in group 1. In group 2, the value fell from 1,621 to 641 ng/mL. We observed a CgA decrease of at least 25% in 45.8% of patients in group 1 and 68.9% in group 2. A CgA increase was registered in 29.2% of patients from group 1 and in 9.8% from group 2. The remaining patients showed largely stable CgA values in the course of the therapy and during follow-up examinations (25.0% group 1, 21.3% group 2).
The mean NSE measured before and in the follow-up after the treatment was 28.2 ng/mL versus 40.0 ng/mL in group 1 and 23.3 ng/mL vs. 17.7 ng/mL in group 2. A biochemical response according to NSE was observed in 34.2% of group 1 and 35.4% of group 2. An NSE increase was detected in 23.7% of patients from group 1 and 20.8% from group 2. The largest proportion of patients was stable: 40.1% in group 1 and 43.8% in group 2.
Discussion
At the time the study was initiated, the role of PRRT in the treatment of NET was not yet established (24–28). Meanwhile, from prospective studies we learned more about PRRT and, especially, from the randomized phase III NETTER-1 trial in which there was a significant clinical benefit from the therapy in patients with midgut NET (23, 29–31). Compared with SSA alone, the combination of PRRT and SSA could achieve a prolonged PFS (median not reached, approximately 40 months, P < 0.001) and overall response of 18%. These results prove that PRRT in combination with SSA is superior to SSA alone (23). However, it is still unclear whether there is an enhancement of the antiproliferative effect by adding SSA to PRRT.
The PROMID and CLARINET studies showed an antiproliferative effect of SSA. However, a significant benefit of SSA in the overall survival has not yet been proven. The main reason for this might be that crossover of patient groups was allowed (18–22).
The present retrospective study examines the effect of SSA in combination therapy with PRRT or as a maintenance therapy after PRRT in patients with metastatic GEP-NET. These patients were compared with a cohort of patients, who received PRRT monotherapy. Supplementary Figure S2 demonstrates the trend of the patients treated with PRRT over the years in both groups. The comparison of these groups can be challenging because of the long follow-up needed to distinguish between the treatment regimens. In our study, we had a follow-up of up to 109 months after the end of treatment (mean 32 months) and an overall follow-up of up to 147 months from starting the therapy (mean 49 months).
As shown in the Kaplan–Meier curves (Fig. 2), most of the patients had a better PFS and OS in group 2 than in group 1: PFS 48 months versus 27 months (P = 0.012) and OS 91 months versus 47 months (P < 0.001). The imaging response rate was 40.0% (CBR 78.7%) in group 1 versus 63.1% (CBR 95.2%) in group 2. These results support the hypothesis that the combination of SSA during or after PRRT can be advantageous for patients with NET.
CgA is a sensitive tumor marker for well-differentiated NETs and correlates with hepatic tumor burden. An abrupt increase of CgA is associated with tumor progress and short survival (32). A higher biochemical response according to CgA measurements was seen in group 2 (68.9%) than in group 1 (45.8%). NSE is an established tumor marker for poorly differentiated neuroendocrine carcinomas and is associated with poor survival (33, 34). The response according to NSE measurements was very similar in both groups: decrease in 34.2% (group 1) and 35.4% (group 2); increase in 23.7% (group 1) and 20.8% (group 2); stable in 40.1% (group 1) and 43.8% (group 2). However, in the follow-up, we observed an increase of the median CgA and NSE in group 1: from 2,699 to 3,928 ng/mL and from 28.2 ng/mL to 40.0 ng/mL, respectively. In contrast, the mean tumor-marker levels in group 2 decreased after the therapy. As shown in Table 1, the difference in the baseline CgA and NSE levels between both groups was not significant. The tumor-marker elevation in group 1 can be a sign that the aggressiveness of the tumor is increasing after PRRT monotherapy. These results substantiate the assumption of our previous study in patients with multiple PRRT cycles, which showed PFS to decline steadily after each salvage PRRT, which might be also a sign of dedifferentiation and higher aggressiveness of NET (35).
When we compared the survival in group 2A (maintenance therapy after PRRT monotherapy) with group 2B (combined therapy with PRRT and SSA followed by a maintenance therapy), there was no significant difference in the PFS, but the overall survival in group 2A was better. However, there were two differences in the patient cohorts in both groups, which might be relevant for the difference in survival: First, in their medical history, more patients in group 2B (N = 31) than in group 2A (N = 25) had SSA previous to PRRT and the majority of patients eventually progressed (N = 28 in group 2B and N = 24 in group 2A). Nevertheless, the main reason for the combination therapy was the better symptom control, such as reduction of flushing and diarrhea, but also increased capability to interact socially, which is relevant for quality of life (36–38). The second difference in the prior therapies was another radionuclide treatment, which was performed only in group 2B (N = 6). However, it is debatable whether the differences in pretreatment can explain the differing overall survival of group 2A and 2B while at the same time accounting for the lack of difference in PFS.
Subgroup analyses showed that especially patients with a high tumor burden, Ki67 between 10% and 20% and functioning tumors had the most significant survival benefit from SSA as a maintenance or combination therapy. A possible explanation could be the higher need for continuous antiproliferative therapy in these groups of patients. Because of the high radiation exposure, PRRT cannot be applied continuously. In contrast, long-acting Octreotide and Lanreotide are long-acting “non-radioactive” SST analogues, which can be administered every 4 weeks and are applicable for the maintenance treatment.
To the strengths of this study counts, the large number of patients included. However, there are several limitations. A possible limitation is the fact that eligibility did not exactly match the NETTER-1 trial (e.g., WHO response criteria used instead of RECIST, patients were not required to have progressed on SSA prior to PRRT). Another limitation is its retrospective design lacked standardized data collection, which may have induced bias. As stated previously, patients in each group were treated with SSAs before administering PRRT. This was because we did not observe a significant difference in the survival of patients pretreated with or without an SSA. It should be mentioned that more patients from group 2 (64.4%) than group 1 (49.4%) were treated with SSA prior to PRRT. This might be a potential bias, because patients who progress rapidly on SSA are more likely have their SSA stopped when they begin subsequent therapy such as PRRT, whereas patients with minor progression are more likely to continue SSA treatment. These biases might contribute to differences in PFS and OS between groups 1 and 2. However, there was no significant difference in both groups regarding the number of patients who were progressive under SSA prior PRRT: 87.9% from group 1 and 92.9% from group 2. Furthermore, we were concerned that inclusion of more groups in the study might increase the incidence of bias. After proper selection according to inclusion/exclusion criteria, the patients in both groups showed no significant differences in gender distribution, age, ECOG, tumor origin, tumor distribution, tumor burden, or progressive disease prior to PRRT. Otherwise, such differences in the cohorts might have affected the outcome of these patients irrespective of the treatment administered. However, prospective studies should be performed in the future to confirm the efficacy of SSAs for combination treatment with PRRT or as a maintenance treatment. For better comparability between the trials, the inclusion criteria of and the response evaluation should match the recent NETTER-1 trial.
In conclusion, this study indicates that SSA might play a significant role in tumor control in patients with GEP-NET who underwent a PRRT, especially as a maintenance therapy. Based on these results, future prospective randomized trials are warranted.
Disclosure of Potential Conflicts of Interest
A. Yordanova reports receiving commercial research support from Novartis and IPSEN for attending symposia. M.M. Wicharz reports receiving commercial research support from Novartis for attending symposia. H. Ahmadzadehfar reports receiving speakers bureau honoraria from Beyer, IPSEN, Novartis, and SIRTex, and reports receiving commercial research grants from Novartis. No potential conflicts of interest were disclosed by the other authors.
Ethical Approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. This article does not contain any studies with animals performed by any of the authors.
Informed Consent
Informed consent was obtained from all individual participants included in the study.
Authors' Contributions
Conception and design: A. Yordanova, M. Essler, H. Ahmadzadehfar
Development of methodology: A. Yordanova, M. Essler, H. Ahmadzadehfar
Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): A. Yordanova, M.M. Wicharz, K. Mayer, M.A. Gonzalez-Carmona, C.P. Strassburg
Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): A. Yordanova, M.M. Wicharz, P. Brossart, M.A. Gonzalez-Carmona, R. Fimmers, M. Essler, H. Ahmadzadehfar
Writing, review, and/or revision of the manuscript: A. Yordanova, P. Brossart, M.A. Gonzalez-Carmona, C.P. Strassburg, M. Essler, H. Ahmadzadehfar
Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): A. Yordanova, M.M. Wicharz, M. Essler, H. Ahmadzadehfar
Study supervision: A. Yordanova, M. Essler, H. Ahmadzadehfar
Acknowledgments
We are grateful to the nursing staff of the treatment ward in our department. We give special thanks to our study nurse, Mrs. Ulrike Kuhn-Seifer (Department of Nuclear Medicine Bonn).
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.