Diagnostic performances of ⁶⁸Ga-DOTATOC versus ¹⁸Fluorodeoxyglucose positron emission tomography in pulmonary carcinoid tumours and interrelationship with histological features

Abstract

Few data are available about radio-receptorial positron emission tomography (PET) results by the use of ⁶⁸Ga-DOTA-peptides in pulmonary carcinoid tumours. In this study, we retrospectively analysed ⁶⁸Ga-DOTATOC/PET and ¹⁸Fluorodeoxyglucose (FDG) PET diagnostic performances in 62 pulmonary carcinoids (occurring in 57 patients) and interrelationship with histological features. All patients underwent at least 1 PET/computed tomography: 26 patients received ⁶⁸Ga-DOTATOC, 52 patients had ¹⁸F-FDG and 20 patients received both techniques. There were 55 typical carcinoids and 7 atypical carcinoids. ⁶⁸Ga-DOTATOC/PET recorded an 88.4% overall detection rate (DR) (meanSUV_max: 15.5); ¹⁸F-FDG/PET a DR of 53.8% (meanSUV_max: 3.2). When adopted a maximum standardized uptake value-threshold of 1.5, DRs of ⁶⁸Ga-DOTATOC and ¹⁸F-FDG/PET increased to 100% and 80.8%, respectively. Moreover, DRs in both techniques vary considerably according to histology with ⁶⁸Ga-DOTATOC/PET having better performances in typical carcinoids than in atypical carcinoids (DR: 91.7% vs 50.0%, P = 0.076). We also observed a significant correlation between a low number of mitoses (<2/10 high-power field) and ⁶⁸Ga-DOTATOC/PET-positivity (P = 0.004), and an association trend (P = 0.076) between necrosis and ⁶⁸Ga-DOTATOC/PET-negativity. In conclusion, ⁶⁸Ga-DOTATOC had better diagnostic performances than ¹⁸F-FDG/PET in detecting pulmonary carcinoids. DRs of both techniques remarkably varied according to histology with ⁶⁸Ga-DOTATOC/PET performing at its best in typical carcinoids, whereas ¹⁸F-FDG/PET did the same in atypical carcinoids.⁶⁸Ga-DOTATOC/PET results were presumably associated with the number of mitoses and the presence of necrosis.

INTRODUCTION

Broncho-pulmonary carcinoids (PCs) are rare neoplasms (2–5% of all pulmonary tumours [1]) that might be classified into 2 groups: typical carcinoid tumours (TCs) and atypical carcinoid ones (ACs) with ACs showing a more aggressive clinical course. Thus, PC diagnosis and differentiation between TC and AC could achieve clinical relevance if performed in a preoperative setting, with both therapeutic and prognostic implications [2]. Unfortunately, TCs and ACs share structural radiological findings [3], and also a preoperative histological confirmation might be challenging, especially in small peripheral PCs. In this framework, availability of further data on the biological behaviour of these neoplasms would be of great help to surgeons before intervention. Although of common use in metabolic characterization of lung nodules, ¹⁸Fluorodeoxyglucose positron emission tomography (PET)/computed tomography (CT) functional imaging was considered poorly useful in PCs, above all in TCs [3]. Unlikely to this, ⁶⁸Ga-labelled somatostatin analogues (DOTANOC, DOTATOC, DOTATATE) PET/CTs are proving to be effective assessment tools for patients affected by neuroendocrine tumours, even PCs [4], as they are characterized by a high expression of somatostatin receptors.

Concerning ⁶⁸Ga-DOTA-peptide combined ¹⁸Fluorodeoxyglucose (¹⁸F-FDG)-PET/CT dual-tracer evaluation in PC patients, our preliminary data (<30 cases) [5] suggested that combined PET information might increase the sensitivity and specificity, in accordance with other analyses [6, 7].

The aims of this study are (i) to confirm the diagnostic performance of both ⁶⁸Ga-DOTA-peptide and ¹⁸F-FDG-PET/CT in a larger monocentric cohort of PC patients (62 cases), (ii) to investigate DRs of both PET-techniques according to PC histology (TCs versus ACs) and (iii) to explore interrelationships between pathological features and PET results.

MATERIALS AND METHODS

We retrospectively reviewed the clinical information of all consecutive PC patients (57 cases) undergoing ¹⁸F-FDG and/or ⁶⁸Ga-DOTATOC/PET followed by surgical resection between January 2009 and August 2017. Moreover, to calculate the sensitivity (SN), specificity (SP), positive and negative predictive values (PPV and NPV) of ⁶⁸Ga-DOTATOC/PET scan in patients with suspected PCs, we explored our Institutional databases and reviewed the clinical records of (i) those patients who tested ‘positive’ at ⁶⁸Ga-DOTATE/PET scan but with no pathological evidence of PC after pulmonary resection (‘False-Positive’) and (ii) those cases resulted ‘negative’ at ⁶⁸Ga-DOTATE/PET scan and with no evidence of PC after radiological surveillance or surgical resection (‘True-Negative’). Prior approval to the scientific use of data related to standard clinical practice was given by the Institutional Review Board.

Indication to positron emission tomography/computed tomography

In the line of extreme simplification, ¹⁸F-FDG was generally the PET tracer of choice in the initial evaluation of patients with a solitary pulmonary nodule. ⁶⁸Ga-DOTATOC/PET scan was indicated after a negative result at ¹⁸F-FDG-PET when the histological diagnosis was not available but the pulmonary lesion resulted to be radiologically suspected of PC. Moreover, in few cases ⁶⁸Ga-DOTATOC/PET scan was performed for the staging purpose when the histological diagnosis of PC was already achieved (endoscopic biopsy) before scheduling ¹⁸F-FDG/PET.

Positron emission tomography/computed tomography protocols and image analysis

All PET/CTs were performed at the same Nuclear Medicine Center on a dedicated hybrid scanner (Discovery STE; GE Healthcare, Chalfont St. Giles, UK) as already described [5]. PET/CT-findings were independently reviewed by 2 experienced nuclear medicine physicians (A.V./A.F.), who reread all studies blinded to the original clinical reports and reached a consensus. DRs for both tracers were determined on per-patient basis, taking into account the number of positive studies and the number of PCs, according to the histological diagnosis. For each lesion, the maximum standardized uptake value (SUV_max) was calculated as the highest tumour voxel value for the primary lung tumour.

Surgery

Lateral thoracotomy was performed in 42 cases (68%), whereas the mini-invasive thoracoscopic approach was used in 20 cases (32%). Sublobar resection (wedge resection/segmentectomy) was indicated in patients with poor pulmonary function or multiple PCs. A lymph node dissection was indicated in all cases.

Pathological evaluation

In agreement with WHO criteria, all pathological specimens were reviewed and classified as TCs or ACs based on the presence of necrosis and the mitotic rate [number of mitoses per 10 high-power field (HPF)]. Cases were evaluated for the presence of adjacent organ invasion and vascular invasion. Mitoses were calculated by the mitotic count and mitotic indexes by immunohistochemistry with Ki-67. The presence or absence of necrosis (focal/diffuse extension) was also evaluated. All cases were investigated by immunohistochemistry by using an automated immunostainer (Benchmark XT Ventana Medical System, Tucson, AZ, USA) with thyroid-transcription-factor-1 (Ventana, clone-8G7G3/1) and markers for neuroendocrine cell differentiation: chromogranin-A (Ventana, clone-LK2H10), synaptophysin (Ventana polyclonal) and CD56 (Ventana, clone-123C3D5).

Statistical analysis

Absolute and relative frequencies were used to analyse qualitative variables, whereas median and mean frequencies were employed to descriptively analyse the quantitative variables. The χ² test was used to test dissimilarities in the prevalence of categorical variables. SN, SP, NPV and PPV were calculated on the entire population. The analysis was carried out through SPSS software (version_16, SPSS Inc., Chicago, IL, USA).

RESULTS

The mean age and female/male ratio were recorded as 68.2 ± 12.3 years and 3:1. Table 1 summarizes the population’s main features: there were 55 TCs and 7 ACs; all patients underwent at least 1 PET/CT scan: 26 patients received ⁶⁸Ga-DOTATOC, 52 patients had ¹⁸F-FDG and 20 patients underwent both techniques. Most tumours were classified as Stage-I (∼80%), with a mean 1.9 cm tumour size.

Diagnostic performances of ⁶⁸Ga-DOTATOC and ¹⁸Fluorodeoxyglucose/positron emission tomography

Table 2 reports DRs of both techniques, according to PC histology (TCs versus ACs) and according to SUV_max-threshold (1.5 vs 2.5). In the overall population (all PCs), ⁶⁸Ga-DOTATOC/PET showed a significantly better diagnostic performance compared to ¹⁸F-FDG/PET. More precisely, by adopting an SUV_max-threshold of 2.5, the ⁶⁸Ga-DOTATOC/PET recorded a 88.4% DR (meanSUV_max: 15.5), whereas ¹⁸F-FDG/PET reached 53.8% (meanSUV_max: 3.2; P = 0.0025). By using the SUV_max-threshold of 1.5, DRs increased to 100% and 80.8% (P = 0.0166), respectively.

The superiority of ⁶⁸Ga-DOTATOC/PET was evident in TCs but not in ACs, where ¹⁸F-FDG/PET showed a better diagnostic performance compared to ⁶⁸Ga-DOTATOC/PET (see Table 2).

Sensitivity, specificity, negative predictive value and positive predictive value of ⁶⁸Ga-DOTATOC/positron emission tomography

We did not find any case with a ‘False-Positive’ result at ⁶⁸Ga-DOTATOC/PET, whereas a total of 16 ‘True-Negative’ cases were observed in the same study period. In detail, among 16 True-Negative cases (meanSUV_max: 0.2; range: 0–0.6) the benignity was confirmed after pulmonary resection in 9 cases (hamartochondroma in 8 cases and intrapulmonary lymphadenopathy in 1 case) or a long-term radiological surveillance (at least >3years) in the remaining 7 cases.

Adopting 2.5 as the SUV_max-threshold, the SN, SP, NPV and PPV of ⁶⁸Ga-DOTATOC/PET were 88%, 100%, 84.2% and 100%, respectively.

Positron emission tomography/computed tomography results and histological features

Exploring the relationship between the number of mitoses and the presence of necrosis and both PET-findings, we found that a <2/10 HPF number of mitoses significantly correlates with ⁶⁸Ga-DOTATOC/PET-positivity (P = 0.004). A trend of association (P = 0.076) was also found between the presence of necrosis and ⁶⁸Ga-DOTATOC/PET-negativity.

DISCUSSION

Surgery represents the gold standard in PCs treatment, whereas the extension of surgical resection (lobar versus sublobar) is almost debated. Indeed, some authors [2] stated that the parenchymal-sparing approach in TCs could be justified from the oncological perspective. On the contrary, recent analyses of large multicentric datasets [8] reported better overall survivals when performing anatomical surgical resection, even for Stage-I TCs. Therefore, accurate preoperative PET/CT scans and correlations with PC histology may help physicians in setting up treatment plans and deciding the best surgical approach.

In the present analysis, ⁶⁸Ga-DOTATOC-PET/CT in diagnosing PCs presented very rewarding results in terms of SP, PPV and, in a lesser manner, SN and NPV. Moreover, in accordance with previous studies on smaller series [3, 6], ⁶⁸Ga-DOTATOC-PET/CT scan leads to better results compared to ¹⁸F-FDG-PET/CT in PCs, especially in TCs (DR 91.7% vs 50.0%; P = 0.076). Unlikely, ¹⁸F-FDG-PET/CT seems to be more accurate (DR 71% vs 50%; P = ns) than ⁶⁸Ga-DOTATOC-PET/CT, although limited data did not allow us to draw any final conclusion. Different results in ⁶⁸Ga-DOTA-peptides and ¹⁸F-FDG-PET/CT scans can be explained by their different uptake mechanisms, correspondent to distinctive biological qualities (different metabolic pathways of tumour cells) between TC and AC. As already underlined [5], ¹⁸F-FDG uptake mirrors glucose metabolism and peculiarly accumulates in high-grade tumours.⁶⁸Ga-DOTA-peptides-PET relates to the presence of somatostatin receptors, which are expressed by well/moderate-differentiated neuroendocrine tumours, as PCs usually are. In line with previous studies [3, 8], we underline that by adopting the ‘standard’ 2.5 SUV_max cut-off, the ¹⁸F-FDG-PET/CT leads to a suboptimal DR in PCs, especially in TCs (54% in PCs; 51% in TCs). Findings vary considerably when setting the SUV_max-threshold at 1.5, as suggested above [9] leading to an 81% DR in all PCs, TCs too. Considering that TCs generally show a low metabolism although still present, we should evaluate ¹⁸F-FDG-PET/CT results as there are no rigid axioms regarding SUV_max-threshold. On the other hand, a different SUV_max-threshold might increase the number of false-positive cases (inflammatory diseases or lung benign tumours). Moreover, we correlated PET/CT results with histological features, aiming at better identifying different cancer patterns. Possibly due to scarcity of data, we failed in distinguishing clearly different patterns of disease, although we preliminarily had found a correlation between the number of mitoses and ⁶⁸Ga-DOTATOC/PET results. Our result is not so impressive, as TCs are marked by few mitoses and, in general, they show a more avid behaviour at ⁶⁸Ga-DOTATOC/PET. An association between necrosis and ⁶⁸Ga-DOTATOC/PET-negativity may be much more interesting, even if it was detected in just few cases. We could assume that fewer viable cells had fewer somatostatin receptors available for the ⁶⁸Ga-labelled somatostatin analogues, this meaning a low uptake at ⁶⁸Ga-DOTATOC/PET, but such preliminary remarks need confirmation by further studies.

CONCLUSIONS

In the detection of PCs, ⁶⁸Ga-DOTATOC/PET ensures better diagnostic performances compared to ¹⁸F-FDG/PET. DRs by both techniques remarkably vary according to histology: ⁶⁸Ga-DOTATOC/PET performs at its best in TCs, whereas ¹⁸F-FDG/PET does it in ACs. ⁶⁸Ga-DOTATOC/PET results could be associated with a number of mitoses and the presence of necrosis.

Presented at the 26th European Conference on General Thoracic Surgery, Ljubljana, Slovenia, 27–30 May 2018.

ACKNOWLEDGEMENTS

The authors thank Daniela Masi (AUSL-IRCCS, Reggio Emilia) for her support in English editing.

Conflict of interest: none declared.

REFERENCES