Outcomes of patients with metastatic neuroendocrine lung neoplasms: typical versus atypical carcinoids

Neuroendocrine neoplasms (NENs) are a group of neoplasms derived from the endocrine system.(1) Within the NEN group, pulmonary neuroendocrine tumours (NETp) constitute approximately 20% of all primary lung tumours. They are divided into four categories: typical carcinoids (TC), atypical carcinoids (AC), small cell lung carcinomas (SCLC), and large cell neuroendocrine carcinomas (LCNEC).(2 3) These tumours have certain morphologic, ultrastructural, immunohistochemical, and molecular characteristics in common but there are important differences in incidence and survival, and clinical, epidemiologic, histological, and molecular features.(2 4)

NETP diagnosis can be challenging, given the morphological similarities with other tumours.(3) There are limited data on outcomes of systemic treatment for NETP of different cell morphology.(3) For example, mitotic rates distinguish TC from AC, and the 2018 World Health Organization (WHO) criteria defined TC as lacking necrosis, 0 to 1 mitosis per 2mm², while AC demonstrates necrosis and/or 2 to 10 mitosis per 2mm².(1 5)

Among the NETP categories, patients with metastatic TC generally have favourable prognosis.(6 7 8) Treatment options for unresectable/metastatic TC include somatostatin analogue and everolimus, while to date, there is no consensus on the use of chemotherapy for TC patients.(3 9) In contrast, AC appears to be more frequent, with a higher rate of distant and nodal metastases, and inferior five-year survival rate,(10 11 12) even when metastatic disease is present.(13)

Data on real world patients with metastatic NETP are limited. Therefore, given the dearth of information on outcomes of systemic treatment in NETP, our study evaluated the median progression-free survival (mPFS) of metastatic patients, comparing TC and AC subtypes and different therapies.

This study was a retrospective analysis that included consecutive patients diagnosed with metastatic lung neuroendocrine tumors, from June 1995 to October 2017, who were treated and followed up at A.C. Camargo Cancer Center, São Paulo, Brazil, from March 2002 to November 2019. The selected patients had a confirmed diagnosis of metastatic typical or atypical carcinoid. They underwent metastasis resection from December 2004 to August 2018. All patients received at least one treatment modality for a progressive disease (somatostatin analogue [SSA], chemotherapy [ChP], or everolimus [Eve]). Only three of the patients assessed received lutecio, which was not considered in this study

The absolute frequency of each variable was calculated. Quantitative variables were assessed for normality using histograms and the ShapiroWilk test. The chi-square test was used to compare qualitative variables between patients with TC or AC.

To evaluate the median progression-free survival (mPFS), the time each patient needed for treatment, as well as the cause for disrupting treatment (death, reaction to treatment, or disease progression), were considered. The medical records were accessed to define the morphologic response (RECIST 1.1 guidelines) as criteria for disease progression. We considered each treatment line individually, even if the patient had undergone more than one treatment. The period in which the patients received both SSA and Eve was calculated separately to see if they had a different response from that treatment period alone. The mPFS (calculated in months) was analysed using the Kaplan- Meyer method. Statistical tests were considered significant if the two-tailed p -value was <0.05. All analyses were case-complete, with denominators reported for each comparison.

This research, like all studies carried out by A.C. Camargo, was committed to ethics and strict compliance with internal policies and the law.

Twenty-seven patients were included: 12 (44%) with TC and 15 (56%) with AC. TC patients were on average 58 years, 10 (83.3%) were female, and 4 (33.3%) received more than one treatment. AC patients were on average 61 years, 10 (66.7%) were female and 3 (20%) received more than one treatment.

AC patients were numerically more commonly treated with SSA (TC: 75% vs. AC: 80%, p =0. 75), ChP (TC: 33.3% vs. AC: 66.7%, p =0.08) and Eve (TC: 41.7% vs. AC: 80%, p =0.04). The cisplatin and etoposide were the most frequent ChP regimen (TC: 75% vs. AC: 30%, p =0.248) (see more details inTable 1).

There were no significant differences ( p <0.05) between regimens, but we observed that, numerically, patients had disease stabilization with SSA as well as with a combination of SSA-Eve (Table 2). PFS times according to treatments are depicted inFigures 1 and 2 . When all metastatic NETP patients (TC + AC) were evaluated, we observed that most experienced disease stabilization with SSA (Table 2).

AC: Atypical Carcinoids; ChP: Chemotherapy; Eve: Everolimus; SSA: Somatostatin analogue; TC: Typical carcinoids.

Among TC patients, treatments with SSA offered a longer mPFS (mPFS SSA: 14.50 months, Eve: 2.50 months, ChP: 4.0 months, SSA + Eve: 4.50 months) when compared to other regimens (seeFigure 1).

The PFS for AC patients indicated similar mPFS for somatostatin analogues, chemotherapy and somatostatin analogues combined with everolimus (mPFS SSA: 7.50 months, Eve: 4.50 months, ChP: 7.50 months, SSA + Eve: 7.00 months), being inferior mPFS just for everolimus use (seeFigure 2).

The progression-free survival curve for 27 patients indicated higher mPFS for SSA when compared to the other regimens (mPFS SSA: 10.50 months, Eve: 3.00 months, ChP: 6.00 months, SSA + Eve: 6.00 months) and inferior mPFS for everolimus use (seeFigure 3).

Our results indicate that SSA therapy offers tumor control for patients with metastatic NETP. When the groups of AC and TC were assessed separately, disease stabilisation with SSA was observed for both. SSAs that are often recommended for patients with advanced neuroendocrine tumours since they exhibit a high affinity for at least two of the five types of somatostatin receptors.(14 15) SSAs also exert their inhibitory actions against hormone secretion and cell proliferation,(16) promoting stabilisation in 30-70% of patients with well-differentiated NENs of different origins and prolonged mPFS, although, without proven survival gain.(14 15 17 20 21) We demonstrated that these results were observed in our patients with NETP, independent of the carcinoid subtype (typical or atypical).

The effectiveness of SSAs combined with Eve have also been described in different types of functioning NENs,(22 23) including non-functioning NETP, TC, and AC.(9 17 24 25 26) However, patients with metastatic NETP, either TC or AC, were under- represented in these clinical trials. This highlights the importance of generating realworld data of treatment outcomes for NETP patients.

In our study, ChP was used more frequently for AC patients, likely reflecting their worse prognosis, which tends to influence treatment decisions towards more aggressive therapies. The efficacy of ChP is limited for advanced NETP, with data coming mainly from small retrospective studies with cisplatin and etoposide.(17) This combination was found to be appropriate for NETP; although, temozolomidebased regimens have also been investigated in small case series, with heterogeneous results.(27 28 29)

The limitations of our study should be highlighted. There is inherent selection bias because of the nature of the study, which was retrospective and unicentre. Also, given the long-time span covered here, we could not have detailed the information from the medical charts pertaining to treatmentrelated adverse events or proper evaluation of radiological responses. These factors may have interfered in the selection of patients as well as in the calculation of the survival time that was free of disease progression. Nevertheless, this study is relevant for providing oncological clinical results from real-world patients with NETP treated in Brazil, and is the first Brazilian experiment reporting the outcomes of cancer treatment in metastatic disease.

In conclusion, although the statistical analyses did not show a significant difference between progressionfree survival ( p <0.05), numerically, more patients with TC or AC experienced tumor control with SSAs, where the mPFS pairs showed a possible tendency to differentiate themselves from the other regimes (Eve and ChP). Satisfactory results were also achieved in the Eve/SSA combination, with better numbers than the other isolated treatments, which may indicate possible inhibitory effects against cell proliferation and comparatively fewer side effects. Despite the limitations of our study and the low number of patients in the sample, our study may be the basis for future prospective studies.