Prognostic value of interim PET-CT in follicular lymphoma

Abstract

Background

Interim positron emission tomography-computed tomography (iPET) has shown prognostic value in high-avidity lymphomas, like diffuse large B-cell lymphoma and Hodgkin lymphoma. However, its role in follicular lymphoma (FL) remains unclear.

Methods

We conducted a retrospective observational study including patients with grades 1–3a FL treated with frontline chemoimmunotherapy between 2012 and 2014 at our institution, who underwent iPET after the fourth treatment cycle. All PET-CT were re-evaluated to assign Deauville 5-point scale (D5PS) and we considered complete metabolic response (CMR) when D5PS was 1–3. Primary endpoints were progression-free survival (PFS) and overall survival (OS).

Results

Of the 38 patients included, 63.2% (24/38) achieved CMR on iPET. All the patients who were in CMR on iPET, were also in CMR at the end of treatment PET-CT. However, among the 14 patients without CMR on iPET, only 4 patients achieved CMR at end of treatment PET-CT and 5 patients had progressive disease. After a median follow-up of 87.6 months (range 11.0–129.8), achieving CMR was associated with longer PFS (p < 0.001; HR 0.08; 95% CI 0.03–0.24) and longer OS (p = 0.012; HR 0.24; 95% CI 0.07–0.80).

Conclusions

In this single-center study with long follow-up, iPET evaluated by D5PS predicted PFS and OS in FL.

Resumen

Introducción

La tomografía por emisión de positrones con tomografía computarizada intermedia (iPET) ha demostrado tener un alto valor pronóstico en linfomas con alta avidez por la glucosa, como el linfoma difuso de células grandes B y el linfoma de Hodgkin. Sin embargo, su papel en el linfoma folicular (LF) aún no está claro.

Métodos

Realizamos un estudio observacional y retrospectivo en el que incluimos pacientes con LF grado 1-3a que iniciaron tratamiento de primera línea entre 2012 y 2014 en nuestro centro y a los que se realizó un iPET tras el cuarto ciclo. Todos los PET-TC fueron reevaluados acorde a la Deauville 5-point scale (D5PS) y consideramos la Respuesta Completa Metabólica (RCM) cuando la D5PS estaba entre 1 y 3. Los endpoints primarios de pronóstico fueron la supervivencia libre de progresión (SLP) y la supervivencia global (SG).

Resultados

De los 38 pacientes incluidos en el estudio, un 63.2% (24/38) alcanzaron una RCM en la iPET. Todos los pacientes que estaban en RCM en la iPET, también lo estaban en el PET-TC de final de tratamiento. Sin embargo, de los 14 pacientes que no estaban en RCM en la iPET, solo 4 alcanzaron una RCM al final de tratamiento y 5 pacientes tenían progresión de la enfermedad. Con una mediana de seguimiento de 87,6 meses (rango 11,0-129,8), encontramos que alcanzar una RCM en la iPET se asociaba con una mejora significativa de la SLP (p < 0,001; HR 0,08; 95%CI 0,03-0,24) y de la SG (p = 0,012; HR 0,24; 95%CI 0,07-0,80).

Conclusiones

En este estudio realizado en un solo centro y con largo seguimiento, demostramos que el iPET evaluado por la D5PS fue predictor de la SLP y de la SG de los pacientes.

Introduction

Follicular lymphoma (FL) is the most common indolent non-Hodgkin lymphoma, representing 15–20% of all lymphomas.¹ Prognosis of FL is heterogeneous. Most patients achieve good responses with frontline treatment and experience prolonged survival. However, a small subset of patients do not respond well to frontline chemoimmunotherapy and experience early progression, leading to poorer survival outcomes.^2,3 Several scores at diagnosis aim to identify patients with poor prognosis and reduced survival, such as the Follicular Lymphoma International Prognostic Index (FLIPI), Follicular Lymphoma International Prognostic Index 2 (FLIPI-2), and the PRIMA-prognostic index (PRIMA-PI).^4–6 These established prognostic tools stratify patients based on various parameters at diagnosis. More recent models, like m7-FLIPI and bio-FLIPI, incorporate molecular or biological factors in an attempt to enhance prognostic accuracy.^7,8 However, a significant proportion of patients with poor outcomes are not detected by these scores, highlighting the need for improved prognostic tools to identify such patients.

Both high-grade and low-grade FL have sufficient glucose avidity,⁹ and 18F-FDG PET-CT is recommended in several clinical scenarios. Baseline PET-CT is recommended for initial staging, while end-of-treatment PET-CT is used for response evaluation.¹⁰ Persistence of positive end-of-treatment PET-CT results is associated with worse prognosis in FL patients.¹¹ In high-avidity lymphomas, such as diffuse large B-cell lymphoma and Hodgkin lymphoma, interim PET-CT (iPET) analysis during first-line treatment has shown high prognostic value.¹² However, the role of iPET in FL remains unclear, and there is no standard recommendation for its use. Multiple studies have assessed the prognostic value of iPET, but results remain controversial.^13–24 Nearly all of these studies are retrospective, with most focusing on progression-free survival (PFS) as the primary endpoint. Furthermore, some of these studies predate the development of the Deauville 5-point scale (D5PS), leading to non-standardized interpretation of iPET results. The number of treatment cycles before conducting PET-CT varies widely across studies, and this variability may contribute to the differing results. Two systematic reviews have addressed this issue. The first one, which included three studies, did not use the D5PS for interpretation and concluded that the available evidence (as of 2016) did not support the use of iPET in FL.²⁵ The more recent systematic review,²⁶ which included only studies utilizing the D5PS, analyzed four studies with a total of 427 patients. This review found significant differences in both PFS and overall survival (OS), concluding that positive iPET results were strongly associated with worse outcomes and that iPET could be a valuable tool for early prognostic risk stratification.

At our institution, a subset of FL patients undergoes iPET evaluation, although the criteria for its use are not standardized and are largely based on the treating physician's discretion. Therefore, we decided to investigate the role of iPET in FL at our center. Given the long survival times of most FL patients, we opted for a study period with an extended follow-up.

Methods

This is a retrospective observational study aimed at evaluating the prognostic value of iPET in FL at our institution. The inclusion criteria were: diagnosis of FL grades 1–3a, treatment with frontline chemoimmunotherapy in our department between 2012 and 2014, and evaluation with iPET after the fourth cycle of frontline treatment. All patients were required to have both pre-treatment and end-of-treatment PET-CT scans. Exclusion criteria included patients without follow-up at our institution, those evaluated with imaging tests other than PET-CT to assess response, and patients who started treatment after 2014 despite being diagnosed before that year.

The primary endpoint of the study was to determine whether a positive or negative iPET result was associated with PFS and OS. The secondary endpoint was to assess whether iPET could improve the prognostic value of the FLIPI and FLIPI-2 indices at diagnosis.

All PET-CT scans were re-evaluated to assign a D5PS score and classify responses according to the Lugano classification.^9,10 For interim PET-CT, complete metabolic response (CMR) was considered when the D5PS score was 1–3, while no CMR was assigned to scores of 4 or 5, which corresponds to partial response (PR), stable disease (SD), or progressive disease (PD).

Demographic, clinical, biological, and follow-up data were collected from electronic medical records using a predefined form. A comprehensive list of the variables recorded is shown in Table S1 of the supplementary data.

PFS was defined as the time from the start of treatment until progression, relapse, or death. OS was defined as the time from the start of treatment until death or the last visit. Progression of disease within 24 months (POD24) was defined as progression, relapse, or death from any cause occurring within 24 months from the start of first-line treatment and was recorded as a binary variable (yes/no) for each patient.

Descriptive statistics for continuous variables were reported as mean ± standard deviation (SD), while categorical variables were expressed as numbers and percentages. Categorical variables were compared using the chi-square test, with Fisher's exact test applied when appropriate. Survival analysis for both PFS and OS was conducted using the Kaplan–Meier method, with differences between survival curves tested by the log-rank test. Univariate and multivariate analyses were performed using Cox proportional hazards regression models, with the number of variables restricted based on the number of events. Statistical significance was defined as p < 0.05. All statistical analyses were performed using IBM SPSS Statistics software (version 25.0).

The study was designed and conducted in accordance with the Declaration of Helsinki, approved by the local Ethics Committee, and informed consent was obtained from all patients who were alive at the time the study was conducted.

Results

A total of 38 patients met the inclusion criteria during the study period. The main reason for exclusion was the absence of an iPET scan for early treatment evaluation. Baseline patient characteristics are summarized in Table 1. Notably, 22 patients (58.7%) had a high-risk FLIPI score, while only 12 patients (31.5%) had a high-risk FLIPI-2 score. Most patients (35; 92.1%) received R-CHOP as first-line therapy, whereas one patient was treated with R-bendamustine (BR) and two patients with rituximab plus lenalidomide. The median PFS for the entire cohort was 91.9 months. After a median follow-up of 87.6 months (range 11.0–129.8), the median OS had not been reached; 68.4% of patients were still alive.

Table 1.

Main characteristics of patients at diagnosis.

Gender
Male	21 (55.3%)
Female	17 (44.7%)
Age, years	60.04 (±11.03)
Grade of FL
Grade 1 or 2	24 (63.2%)
Grade 3a	14 (36.8%)
Ann Arbor staging
I–II	5 (13.1%)
III–IV	33 (86.9%)
Substage A	20 (52.6%)
Substage B	18 (47.4%)
Bulky disease
Yes	19 (50%)
No	19 (50%)
Bone marrow involvement
Yes	18 (47.4%)
No	20 (52.6%)
Splenic involvement
Yes	17 (44.7%)
No	21 (55.3%)
Blood expression
Yes	14 (36.8%)
No	24 (63.2%)
LDH
Normal	25 (65.8%)
High	13 (34.2%)
β2-Microglobulin
Normal	7 (18.4%)
High	31 (81.6%)
Hemoglobin (g/dL)	13.2 (±2.31)
White cell count (×10^9/L)	9.98 (±8.60)
FLIPI score
Low (0–1)	6 (15.8%)
Intermediate (2)	10 (26.3%)
High (≥3)	22 (58.7%)
FLIPI-2 score
Low (0–1)	15 (39.5%)
Intermediate (2)	11 (28.9%)
High (≥3)	12 (31.5%)
First line treatment
R-CHOP	35 (92.1%)
R-bendamustine	1 (2.6%)
R-lenalidomide	2 (5.2%)

Abbreviations: FL: follicular lymphoma; LDH: lactate dehydrogenase; FLIPI: Follicular Lymphoma International Prognostic Index; R-CHOP: rituximab, cyclophosphamide, vincristine, doxorubicin and prednisone; R-: rituximab.

On iPET, 24 patients (63.2%) achieved CMR (D5PS ≤3), while 14 patients did not (13 in PR and 1 with SD). All patients with CMR on iPET also achieved CMR at the end-of-treatment PET-CT. Among the 13 patients with PR on iPET, 4 converted to CMR, 4 remained in PR, and 5 progressed to PD at end-of-treatment evaluation. The patient with SD on iPET maintained SD at the end of treatment (Fig. 1).

Fig. 1.

Flowchart of patients in the study according to interim PET-CT and end of treatment PET-CT evaluation. Abbreviations: FL: follicular lymphoma; PET-CT: positron emission tomography with computed tomography; CMR: complete metabolic response; PR: partial response; SD: stable disease; PD: progressive disease.

Nine patients (23.7%) received consolidation radiotherapy, mainly due to bulky disease at diagnosis. Five of these patients achieved CMR on iPET (20.8%) and four patients without CMR on iPET (28.6%; p = 0.699). Rituximab maintenance (every two months for two years) was administered to all patients achieving at least PR, comprising all patients with CMR on iPET as well as eight patients without CMR on iPET (p = 0.001). During follow-up, only 4 of 24 patients (16.7%) with CMR on iPET required further treatment, compared with 13 of 14 patients (92.8%) without CMR on iPET (p < 0.001). Salvage therapies consisted of BR (n = 5), R-FND (rituximab, fludarabine, mitoxantrone, dexamethasone; n = 5), R-ESHAP (rituximab, etoposide, methylprednisolone, cytarabine, cisplatin; n = 5), and R-GEMOX (rituximab, gemcitabine, oxaliplatin; n = 2). In addition, 8 patients underwent consolidation with autologous stem cell transplantation. Of the 17 patients who experienced disease progression or relapse during follow-up, histological transformation to diffuse large B-cell lymphoma was confirmed by biopsy in only one case, which occurred in a patient without CMR on iPET.

Patients achieving CMR on iPET had significantly longer PFS than those without CMR (p < 0.001; Fig. 2A), with a hazard ratio (HR) of 0.08 (95% CI, 0.03–0.24). Median PFS was 12.2 months in the non-CMR group and not reached in the CMR group after a median follow-up of 87.6 months (range 11.0–129.8). Achieving CMR on iPET was also associated with significantly better OS (p = 0.012; Fig. 2B), with an HR of 0.24 (95% CI, 0.07–0.80). Median OS was 67 months in the non-CMR group and not reached in the CMR group.

Fig. 2.

(A) Progression free survival according to interim PET-CT response. (B) Overall survival according to interim PET-CT response. Abbreviations: PET-CT: positron emission tomography with computed tomography; CMR: complete metabolic response.

Additional exploratory analyses according to individual D5PS scores were performed. Patients with a D5PS score of 3 on iPET showed no significant differences in PFS or OS compared with those with scores of 1–2. Similarly, no significant differences in PFS or OS were observed between patients with D5PS scores of 4 and 5 (Fig. S1 on supplementary data). Given the limited number of patients within each category, these analyses should be interpreted with caution.

Kaplan–Meier curves for PFS according to FLIPI and FLIPI-2 showed similar patterns. The low-risk group had significantly better PFS than intermediate- or high-risk groups, with no difference between the latter two (Fig. 3A and B). In univariate analysis, only iPET and FLIPI-2 (low vs intermediate/high) were significantly associated with PFS. FLIPI could not be assessed in the Cox model because no events occurred in the low-risk group; however, differences between low- and moderate/high-risk groups were significant by log-rank test (p = 0.029). In multivariate analysis, iPET retained independent prognostic value, whereas FLIPI-2 did not (Table 2).

Fig. 3.

(A) Progression free survival according to FLIPI score at diagnosis. (B) Progression free survival according to FLIPI-2 score at diagnosis. Abbreviations: FLIPI: Follicular Lymphoma International Prognostic Index; FLIPI-2: Follicular Lymphoma International Prognostic Index 2.

Table 2.

Univariate and multivariate analysis for progression free survival.

Variables	Univariate			Multivariate
	p value	HR	95% CI	p value	HR	95% CI
Age	0.357	1.02	0.98–1.07
Staging(I–II vs III–IV)	0.230	0.36	0.01–8.14
Grade(1–2 vs 3a)	0.426	0.45	0.08–3.22
Bulky disease	0.262	0.57	0.22–1.51
LDH	0.413	0.67	0.25–1.76
β2-Microglobuline	0.184	0.49	0.17–1.40
FLIPI-2(low vs moderate and high risk)	0.021	0.23	0.06–0.80	0.149	0.37	0.09–1.42
Interim PET-CT	<0.001	0.09	0.03–0.27	<0.001	0.10	0.03–0.34

Bold values indicate statistically significant results (p<0.05). Abbreviations: LDH: lactate dehydrogenase; FLIPI: Follicular Lymphoma International Prognostic Index; PET-CT: positron emission tomography with computed tomography.

We found no significant association between FLIPI or FLIPI-2 and OS, except for a difference between low-risk and high-risk FLIPI groups (p = 0.042; Fig. S2A and S2B on the supplementary data). In univariate analysis for OS, iPET was the only variable with statistical significance.

We also analyzed POD24, which has emerged as an early marker of poor survival.² In our cohort, 10 of 14 patients without CMR on iPET were POD24-positive, compared with only 3 of 24 patients with CMR (p < 0.001).

Discussion

In this retrospective study, we found that iPET during frontline treatment for FL has independent prognostic value for both PFS and OS. These results are consistent with several previous reports, though not all studies agree. Table 3 summarizes the available studies evaluating the prognostic value of iPET in FL.^13–24 Almost all are retrospective, except for the prospective study by Dupuis et al.¹⁴ Most assess PFS as the primary prognostic endpoint, with only a subset also evaluating OS or POD24. On global, nine of the twelve published studies found a significant association between iPET results and PFS, while three did not.

Table 3.

Summary of published studies analyzing the prognostic value of interim PET-CT in follicular lymphoma.

Author (year)	Patients(n)	Study type	Treatment of patients	iPET after how many cycles	iPET positive criteria	Median follow-up(months)	Prognostic variables	Prognostic value of iPET	Multivariate analysis
Bishu^13 (2007)	11	RetrospectiveSingle center	R-CHOPCNOPCVPRICE	4 cycles(2 cycles for RICE)	Subjective assessment	26	PFS: p = 0.13	No	No
Dupuis^14 (2012)	111	ProspectiveMulticentric	R-CHOP	4 cycles	D5PS ≥4	23	PFS: p = 0.046OS: NS	Yes	No
Lu^15 (2014)	44	RetrospectiveMulticentric	R-CHOP	3 cycles	Subjective assessment	58	PFS: p = 0.31OS: p = 0.92	No	No
Boo^16 (2019)	29	RetrospectiveSingle center	R-CVPR-CHOP	3 or 4 cycles	D5PS ≥4	35	PFS: p = 0.005	Yes	PFS negative
Liang^24 (2019)	22	RetrospectiveSingle center	R-CHOP	3 or 4 cycles	ΔTMTVΔTLG	35	PFS: p < 0.001OS: p < 0.001	Yes	PFS positiveOS positive
Zhou^17 (2019)	24	RetrospectiveSingle center	R-CHOPR-CVPR-FM	3 or 4 cycles	D5PSIHPΔSUVmax	34	PFS: p = 0.056 (D5PS)p = 0.186 (IHP)p = 0.148 (ΔSUVmax)	No	No
Laverdure^22 (2022)	143	Retrospective Single center	R-CHOPR-CVPBR	2–4 cycles	D5PS ≥4(D5PS ≥3)	67	PFS: p < 0.001OS: p < 0.001	Yes	PFS positiveOS positive
Merriman^19 (2023)	128	RetrospectiveTwo centers	R-CHOPBRR-CVP	2–4 cycles	D5PS ≥3ΔSUVmax	59	PFS: p = 0.010 (D5PS)p < 0.001 (ΔSUVmax)OS: NS	Yes	PFS positive (D5PS and ΔSUVmax)
Sun^18 (2023)	30	RetrospectiveSingle center	R-CHOP	3–5 cycles	D5PS ΔSUVmax	55	PFS: p > 0.05 (D5PS)p = 0.040 (ΔSUVmax)	Yes, only ΔSUVmax	No
Durmo^21 (2024)	123	RetrospectiveMulticentric	R-CHOPBR	4 or 5 cycles	D5PS ≥4	68	PFS: p < 0.01OS: NS	Yes	PFS positive
Kiss^23 (2024)	103	RetrospectiveSingle center	B+AntiCD20R-CHOPR-CVP	3 or 4 cycles	D5PSSUVmaxΔSUVmax	40	PFS: p < 0.0001 D5PSp < 0.0001 (SUVmax)	Yes	No
Poza^20 (2025)	117	Retrospective Multicentric	R-CHOPBR	4 cycles	D5PS ≥4	72	PFS: p < 0.001, POD24: p = 0.005	Yes	PFS positive

Abbreviations: iPET: interim PET-CT; R-CHOP: rituximab, cyclophosphamide, vincristine, doxorubicin and prednisone; CVP: cyclophosphamide, vincristine, and prednisone; CNOP: cyclophosphamide, vincristine, mitoxantrone and prednisone; RICE: rituximab, ifosfamide, carboplatin and etoposide; R-CVP: rituximab, cyclophosphamide, vincristine, and prednisone; R-FM: rituximab, fludarabine and mitoxantrone; BR: bendamustine and rituximab; B: bendamustine; D5PS: Deauville 5-point scale; ΔTMTV: change in total metabolic tumor volume; ΔTLG: change in total lesion glycolysis; SUVmax: maximum standardized uptake value; ΔSUVmax: reduction of SUVmax from basal PET-CT to interim-CT; IHP: International Harmonization Project criteria; PFS: progression free survival; OS: overall survival; NS: not significant.

One of the key aspects influencing these findings is the method used to interpret PET-CT results. Over the past decade, the standard tool for PET-CT response assessment has been the D5PS. Most studies using D5PS report a significant association between iPET positivity, defined as D5PS ≥4, and worse PFS. Only two studies did not find an association between D5PS ≥4 on iPET and PFS, but these studies had very small sample size, which may have prevented detection of statistically significant differences.^17,18 In addition, there are two studies that have suggested lowering the cutoff to D5PS ≥3, as patients scoring 3 have a poorer prognosis than those with scores of 1–2, and outcome of score 3 patients is closer to those with score 4.^19,22 Other metabolic parameters have also been associated with PFS, including the reduction in maximum standardized uptake value (SUVmax) between baseline PET-CT and iPET,^17–19,23 the absolute SUVmax value on iPET,²³ and changes in total metabolic tumor volume (TMTV) and total lesion glycolysis (TLG).^18,24

Differences are also evident between studies – and even among patients within the same study – regarding the timing of iPET. In most reports, iPET is performed after the third or fourth cycle of treatment, whether with R-CHOP or bendamustine–rituximab (BR). One study demonstrated that the later the iPET is performed, the more likely it is to be negative, as expected.¹⁹

Only a few studies have investigated the association between iPET and OS. Of the six studies that evaluated this, only two found a significant correlation.^22,24 A key factor in detecting OS differences in FL is follow-up duration, as life expectancy is long for most patients. In some studies, shorter follow-up may have limited the ability to detect differences in OS. To our knowledge, our study has the longest median follow-up reported to date (87.6 months). Furthermore, the introduction of novel therapies for relapsed/refractory FL – such as bispecific antibodies and chimeric antigen receptor (CAR) T-cell therapy – has substantially improved patient outcomes,^27,28 making it increasingly challenging to detect OS differences in more recent series.

A recently published meta-analysis examined the prognostic value of iPET assessed by D5PS in FL.²⁶ Only four studies, comprising a total of 427 patients, met inclusion criteria. The authors concluded that iPET assessed by D5PS is a prognostic factor for both PFS and OS.

The main prognostic indices in FL remain FLIPI and FLIPI-2. Several studies have confirmed through multivariate analyses that iPET retains prognostic value independent of these scores, which assess only baseline variables.^19–22,24 A proportion of patients with good or intermediate FLIPI/FLIPI-2 scores will nonetheless experience early relapse or progression. In our study, more than half of the patients who failed to achieve CMR at end-of-treatment PET-CT had low- or intermediate-risk FLIPI/FLIPI-2 scores. Early identification of patients with worse prognosis could be important for identifying possible candidates for novel therapies in the first relapse setting, such as bispecific antibodies CAR T-cell therapy. Incorporating iPET results into FLIPI or FLIPI-2 might therefore enhance their prognostic accuracy.

Considering all available evidence, predominantly from retrospective studies, we believe there is sufficient justification to conclude that iPET in FL has important prognostic value, at least for PFS and likely for OS as well. Therefore, performing iPET during frontline treatment appears advisable to refine prognostic assessment beyond that provided by FLIPI and FLIPI-2 alone.

Our study has several important limitations. It is a single-center retrospective study with a relatively small sample size, which may have limited the statistical power of some analyses. Not all patients treated during the study period were included, mainly because some did not undergo iPET for early evaluation; therefore, a potential selection bias cannot be ruled out. PET-CT scans were re-evaluated by a single reviewer without blinding to previous results, introducing the possibility of interpretation bias. Furthermore, not all patients received the same induction regimen, and not all were given rituximab maintenance, which could have influenced outcomes. Salvage therapy was also heterogeneous and autologous stem cell transplantation was performed only in some patients as consolidation second line treatment and that could influence OS results. Moreover, this study was conducted in a period preceding the availability of novel highly effective therapies for relapsed or refractory follicular lymphoma, such as bispecific antibodies and CAR T-cell therapy. As these treatments have substantially improved outcomes after relapse, the differences in overall survival observed in our cohort should be interpreted with caution and ideally confirmed in more contemporary series treated in the current therapeutic era.

Conclusion

In this single-center retrospective study with long follow-up, iPET assessed by the D5PS was associated with longer PFS and OS in patients with FL treated with first-line chemoimmunotherapy. Our findings are consistent with most of the published literature and suggest that iPET may provide additional prognostic information beyond established baseline risk scores such as FLIPI and FLIPI-2, potentially allowing earlier identification of patients at higher risk of treatment failure. However, given the retrospective design, limited sample size, and the therapeutic context in which the study was conducted, these results should be considered hypothesis-generating and warrant confirmation in prospective studies before being incorporated into routine clinical decision-making.

CRediT authorship contribution statement

S.L., M.-T.H. conceived and designed the study. M.D.-L. recorded the data. D.C.-G. reviewed PET-CT. S.L., M.D.-L., C.B.-B., M.-J.R.-S., B.P.-P., A.M.-M. and M.-T.H. analyzed the data, performed the statistical analysis and interpreted the results. S.L., M.D.-L. wrote the paper. All authors reviewed and approved the final manuscript.

Ethics approval and patient consent statement

The study was approved by the local Ethics Committee (“Comité de Ética de la Investigación con Medicamentos (CEIm) del Complejo Hospitalario Universitario de Canarias”) and informed consent was obtained from all the patients who were alive when this study was performed.

Funding

The present study has not received any funding.

Conflicts of interest

None of the authors has any personal or financial conflicts of interest that could influence the results of this study.

Acknowledgments and Data availability

The authors would like to thank all patients and caregivers of the patients included in this study. The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.