Asphericity of Somatostatin Receptor Expression in Neuroendocrine Tumors: An Innovative Predictor of Outcome in Everolimus Treatment?

1. Introduction

Neuroendocrine tumors (NET) of the bronchopulmonary (lung-) or gastroenteropancreatic system (GEP-NET) are classified as a rare and heterogeneous tumor entity with substantially rising incidence over the last two decades [1,2]. Due to the mostly slow-growing nature of these tumors and the lack of early symptoms, NET are frequently metastasized, and systemic treatment is required. In addition to somatostatin analogues, chemotherapeutic regimes, and radiolabeled peptide receptor treatments (e.g., PRRT), tumor-targeted therapies such as tyrosine kinase inhibitor (TKI, i.e., sunitinib) and especially mammalian target of rapamycin (mTOR) inhibitors (everolimus) are well established [3,4,5,6]. In particular, three large phase III placebo-controlled randomized trials showed the efficacy of everolimus in patients with well-differentiated lung- and gastrointestinal (GI)-NET (RADIANT-2 and RADIANT-4) [3,5] as well as in patients with advanced pancreatic tumors (P)-NET (RADIANT-1 and RADIANT-3) [4,7]. In addition, RADIANT trials demonstrated an improved progression-free survival (PFS) in patients with advanced NET when combining everolimus with somatostatin analogs (SSA), regardless of previous SSA exposure [3,6,7].

Based on these results, the European Neuroendocrine Tumor Society (ENETs) guidelines recommend everolimus as a second-line therapy for advanced P-NEN and as a third-line therapy for GI-NEN [8].

However, due to biological heterogeneity of NET, identification of patients who will respond to treatment remains challenging. As a result, there is an unmet clinical demand to identify patients who are not likely to benefit and thus, avoid the expense and inconvenience of unnecessary therapy. Therefore, prognostic or predictive markers to identify patients that are likely to benefit from everolimus are needed.

Somatostatin receptor (SSR) -specific imaging with scintigraphy (e.g., [¹¹¹In]In-octreotide scintigraphy) or positron emission tomography (PET) is nowadays routinely assessed at initial diagnosis and during follow-up. Despite its use as disease-specific functional imaging, no image-derived parameters are established as prognostic or predictive factors in NET. So far, the asphericity (ASP), a measure of shape irregularity of the functional lesion volume in SSR imaging, has shown prognostic value in NET patients undergoing PRRT [9,10].

This study examined if ASP of NET lesions in SSR imaging before initiation of everolimus treatment could predict PFS.

2. Methods

3. Results

4. Discussion

This retrospective bicentric study investigated the prognostic value of lesional asphericity (ASP) in pretherapeutic [¹¹¹In-DTPA⁰]octreotide (OctreoScan) SPECT/CT in metastatic GEP-NET prior to everolimus treatment. High ASP predicted PFS independently from common clinical and histological parameters. The median Ki-67 index 5% represented well- and intermediate differentiated GEP-NET, and median PFS of all patients was 11.5 months. The median PFS was similar to that published in the RADIANT-3 trial with 11 months in P-NET [4] as well as RADIANT-4 with again 11 months PFS in advanced, progressive, and nonfunctional GI- and lung-NET [5]. In our study, P-, GI-, and CUP-NET were included irrespective of prior ”first-line” treatment strategies, and therapy was combined with somatostatin analogs in most of the cases. Hence, our PFS is comparable with a number of subgroup analyses of the RADIANT-2 trial, which supports the efficacy of everolimus in combination with long-acting repeatable octreotide (octreotide LAR) with a PFS of 13.6 months [3].

Several previous studies investigated different biomarkers and immunohistochemical parameters aimed at differentiation of responders and non-responders to treatment with everolimus but mostly failed [17,18]. Yao et al. reported shorter PFS and overall survival in patients with P-NET with elevated baseline CgA or neuron-specific enolase (NSE) [19]. In a study by Benslama et al. [20], hypercholesterolemia was associated with longer PFS, whereas the presence of bone metastases and overexpression of phosphorylated p70S6 kinase (p-p70S6K) by tumor cells were associated with shorter PFS under everolimus. However, the authors concluded that the absence of bone metastases and high levels of p-p70S6K are unspecific surrogates independently of therapy regime. Moreover, similarly to the histological tumor grade, they are generally known to show a strong association with a worse outcome.

It has, however, been demonstrated that there are different interactions and synergies between the pathways for SSR and mTOR inhibitors. In particular, several preclinical studies have recently shown that everolimus in conjunction with somatostatin might be acting in synergy on common signaling cascades, such as the PI3K-AKT-mTOR pathway [21]. Another explanation for synergistic effects might be the complementary effect of everolimus and somatostatin analogs on the protein 4E-BP1—a cell-signaling hallmark in NET [22,23]. Those synergies potentiate the inhibitory effect on tumor growth, reduce tumor perfusion, and release endogenous immunomodulation, which directly effects antitumor mechanisms.

It is well known that octreotide, a synthetic somatostatin analog, preferentially binds to the SSR subtype 2 and 5. There have been five SSR subtypes (SSTR 1–5) identified, and they exert direct antitumor effects. The prognostic importance of ASP derived from SSR imaging has been recently demonstrated in pretherapeutic and intratherapeutic SPECT/CT imaging in patients with NET undergoing [¹⁷⁷Lu]Lu-DOTATOC-PRRT [9,10]. Thus, it may be hypothesized that neuroendocrine metastases exhibiting a homogeneous and uniform SSR expression could also respond especially well to everolimus and a concomitant treatment with SSA. High ASP values of SSR-positive lesions represent high tumor heterogeneity, thus hinting at dedifferentiated tumor cells and especially, necrosis (Figure 1). Finally, necrosis is known to be accompanied by disruption of the outer cell membrane integrity, initiating inflammation and edema [24]. Therefore, progressive necrosis causes pruning of tumor vessels that correlates with a significant reduction in locoregional blood flow. Poor perfusion may directly impair the therapeutic efficacy of blood flow-dependent treatments, such as everolimus.

Applying the ASP in pre- and intratherapeutic octreotide scintigraphy, it was recently demonstrated that lesions with low ASP (optimal cut-off <5.5% showed the best response to PRRT with [¹⁷⁷Lu]Lu-DOTA-TATE) [10]. Lesions with homogenous SSR distribution and a regularly shaped SSR-positive lesion volume may facilitate a uniformly high lesional irradiation dose. The optimal cut-off value of 12.9% identified in the current patients was higher. While these two studies may only be compared very cautiously, the combined results might indicate that patients with ASP values between 5.5% and 12.9% might be candidates for treatment with mTor-inhibitors rather than PRRT. According to the North American Neuroendocrine Tumor Society (NANETS) guidelines for the treatment of advanced midgut NETs from 2017 [25], everolimus is also recommended as first-line therapy in patients who are no candidates for SSAs (e.g., weak or absent SSR expression). It is important to notice that the definition ”weak“ is a challenging definition for interdisciplinary consensus. So far, only the Krenning score is used to qualitatively describe the avidity of SSR-positive lesions [12]. If the current results can be confirmed, lesional ASP in SSR imaging may facilitate decision making for or against everolimus.

The current study is limited by its retrospective nature and lack of a matching control group undergoing different treatment. Ultimately, the conclusion that ASP not only shows a prognostic value but also offers a predictive role and can directly guide treatment decisions in patients eligible for several treatment options including everolimus would require comparative investigation of the predictive value of ASP in matched cohorts of NET patients undergoing these different treatments. Further limitations may arise: the cumulative dose (CD) and dose intensity (DI) of everolimus per patient were not assessed. Berardi et al. reported that both CD and DI play a prognostic role for patients with advanced P-NET [26]. A high number of patients had to be excluded from the current analysis because they suffered from everolimus-induced intolerable toxicity or required any dose reduction.

As a further limitation, in the current analysis, eight patients underwent liver-directed treatments including treatment of solitary liver metastases and uni-/bilobar selective internal radiation therapy (SIRT). The presence of pretreated metastases might introduce a bias as these lesions could show especially high ASP due to treatment-induced cellular necrosis, hypoxia-induced angiogenesis, and proliferation. Therefore, in these patients, lesions that had been previously treated with microwave or radiofrequency ablation were excluded from the analysis. Consequently, only two patients with prior SIRT were included, and we believe that this small number does not have a substantial impact on the main results. Mainly due to the remarkable progress and the rising clinical impact of SSR imaging [27], the concept of the ASP will be even transferred to functional PET imaging when enrolling prospective studies. Such prospective studies are required to validate the current explorative results and to ensure a well-selected, homogenous patient collective. Ideally, future studies should use SSR-specific PET/CT instead of SPECT/CT imaging, because the substantially better spatial resolution of PET would allow a more differentiated reflection of the lesions’ irregularity, especially in smaller lesions. Consequently, the currently identified ASP cut-off would not be directly transferable to PET imaging in the same context.

5. Conclusion

High lesional ASP in pretherapeutic SSR imaging as well as no prior performed PRRT independently predicted shorter PFS in patients with metastatic SSR-positive GEP-NET undergoing treatment with everolimus. Considering a broad spectrum of therapeutic strategies in well-differentiated NET, ASP could help to identify patients who are less likely to show favorable PFS with everolimus. Prospective, comparative investigation of patients undergoing different second- or third-line treatments is required to validate the prognostic and predictive value of ASP in this clinical setting.

Author Contributions

Conceived and designed the method: C.W., H.A. Evaluated the data: C.W., P.G. Analyzed the data: C.W., J.R., I.S. Wrote the paper: C.W., M.K., H.J., M.V., C.F. All authors have read and agreed to the published version of the manuscript.

Funding

This research did receive a specific grant as part of the project YING (Young Investigators for Neuroendocrine neoplasia in Germany; grant number: S2SAS-009) Novartis (Novartis International AG, a Swiss multinational pharmaceutical company based in Basel, Switzerland). C.W., J.R., P.G., M.K., H.J., M.V. and H.A. have received a speaker honorarium from Company Novartis.

Conflicts of Interest

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.