Multiparametric functional MRI and 18F-FDG-PET for survival prediction in patients with head and neck squamous cell carcinoma treated with (chemo)radiation

Objectives To assess (I) correlations between diffusion-weighted (DWI), intravoxel incoherent motion (IVIM), dynamic contrast-enhanced (DCE) MRI, and 18F-FDG-PET/CT imaging parameters capturing tumor characteristics and (II) their predictive value of locoregional recurrence-free survival (LRFS) and overall survival (OS) in patients with head and neck squamous cell carcinoma (HNSCC) treated with (chemo)radiotherapy. Methods Between 2014 and 2018, patients with histopathologically proven HNSCC, planned for curative (chemo) radiotherapy, were prospectively included. Pretreatment clinical, anatomical, and functional imaging parameters (obtained by DWI/IVIM, DCE-MRI, and 18F-FDG-PET/CT) were extracted for primary tumors (PT) and lymph node metastases. Correlations and differences between parameters were assessed. The predictive value of LRFS and OS was assessed, performing univariable, multivariable Cox and CoxBoost regression analyses. Results In total, 70 patients were included. Significant correlations between 18F-FDG-PET parameters and DWI-/DCE volume parameters were found (r > 0.442, p < 0.002). The combination of HPV (HR = 0.903), intoxications (HR = 1.065), PT ADCGTV (HR = 1.252), Ktrans (HR = 1.223), and Ve (HR = 1.215) was predictive for LRFS (C-index = 0.546; p = 0.023). N-stage (HR = 1.058), HPV positivity (HR = 0.886), hypopharyngeal tumor location (HR = 1.111), ADCGTV (HR = 1.102), ADCmean (HR = 1.137), D* (HR = 0.862), Ktrans (HR = 1.106), Ve (HR = 1.195), SUVmax (HR = 1.094), and TLG (HR = 1.433) were predictive for OS (C-index = 0.664; p = 0.046). Conclusions Functional imaging parameters, performing DWI/IVIM, DCE-MRI, and 18F-FDG-PET/CT, yielded complementary value in capturing tumor characteristics. More specific, intoxications, HPV-negative status, large tumor volume-related parameters, high permeability (Ktrans), and high extravascular extracellular space (Ve) parameters were predictive for adverse locoregional recurrence-free survival and adverse overall survival. Low cellularity (high ADC) and high metabolism (high SUV) were additionally predictive for decreased overall survival. These different predictive factors added to estimated locoregional and overall survival. Key Points • Parameters of DWI/IVIM, DCE-MRI, and 18F-FDG-PET/CT were able to capture complementary tumor characteristics. • Multivariable analysis revealed that intoxications, HPV negativity, large tumor volume and high vascular permeability (Ktrans), and extravascular extracellular space (Ve) were complementary predictive for locoregional recurrence. • In addition to predictive parameters for locoregional recurrence, also high cellularity (low ADC) and high metabolism (high SUV) were complementary predictive for overall survival. Electronic supplementary material The online version of this article (10.1007/s00330-020-07163-3) contains supplementary material, which is available to authorized users.


Introduction
In patients with advanced stage head and neck squamous cell carcinoma (HNSCC), (chemo)radiation is the standard organsparing treatment; however, there is still a 50% (35-65%) recurrence rate [1]. In addition to clinical and histological parameters, other sophisticated biomarkers are needed to stratify patients for optimal therapy (e.g., de-escalation, escalation, or switching to surgery) [2,3]. Being able to correctly identify patients with a favorable prognosis might allow treatment adaptation to reduce long-term toxicity without compromising outcome [4].
Functional imaging techniques capture a variety of biological characteristics, such as cellularity, perfusion, permeability, and glucose metabolism.
Perfusion and vessel permeability can be assessed by dynamic contrast-enhanced (DCE) MRI and quantified by the K trans (transfer rate of contrast agent from plasma to extravascular, extracellular space), V e (fractional volume of extracellular extravascular space), and K ep (contrast agent transfer rate from extravascular, extracellular space to plasma) [8].
The aim of our study was to assess (I) the correlations between diffusion-weighted (DWI), intravoxel incoherent motion (IVIM), dynamic contrast-enhanced (DCE) MRI, and 18F-FDG-PET/CT imaging parameters capturing tumor characteristics and (II) their predictive value of locoregional recurrence-free survival (LRFS) and overall survival (OS) in patients with head and neck squamous cell carcinoma (HNSCC) treated with (chemo)radiotherapy.

Delineation
Whole-lesion delineation was performed manually by two independent observers (J.C. and P.dG., 30 and 15 years of experience in head and neck radiology, respectively) on the ADC map and DCE map. Herewith, T1w, STIR, and T2w maps were used for anatomical correlation, with knowledge of TNM stage and tumor location, but blinded for treatment outcome. Furthermore, the patient largest lymph node metastasis was delineated. DWI/IVIM delineation was performed with VELOCITY software (Varian). To assess the interobserver variability, the correlation (Pearson's r), difference (Wilcoxon signed rank rest), and overlap of delineation (Dice index) were calculated. 18 F-FDG-PET/CT delineation was performed by semiautomatic delineation by a nuclear medicine specialist (B.Z.) using 50% of tumor-specific SUVpeak threshold, corrected for blood glucose level. Details on this method were published previously [29].

Feature extraction
Imaging parameters were extracted from both PT and LNM whole-lesion ROIs of each observer. Anatomical total lesion volume, i.e., gross tumor volume (GTV), was calculated for each ROI on each imaging map (ADC GTV , DCE GTV , and metabolic active tumor volume (MATV)). The following quantitative imaging features were calculated per observer by averaging all voxels included in the whole-lesion ROI.
DCE-MRI analysis was processed with in-house built software (Dynamo; [27]), performing quantitative pharmacokinetic analysis using the 2-compartment Tofts model [8] with patient-specific arterial input function (AIF) obtained from manual delineating the most cranial part of the external carotid artery. The following features were extracted: K trans (transfer rate of contrast agent from plasma to extravascular, extracellular space); V e (fractional volume of extracellular extravascular space); K ep (transfer rate of contrast agent from extravascular, extracellular space to plasma). 18 F-FDG-PET/CT in-house built software (Accurate; [28]) automatically calculated SUV max , SUV mean , and SUV peak (i.e., peak value of 8 highest voxels) based on all included voxels of the ROI, and whole-lesion MATV, and total lesion glycolysis (TLG = SUV mean × MATV).

Statistical analysis
The average of the above written extracted parameters for both observers was used for analyses.
Correlations were assessed between parameters for PT and LNM separately (Pearson correlation coefficient). Differences in imaging parameters among T-stages, N-stages, and intoxications were assessed with the Kruskal-Wallis tests. In order to capture HPV status-specific tumoral characteristics, associations between parameters of patients with locoregional control (LRC) and failure (i.e., recurrence; LRF), and survival and death (univariate Cox regression analysis) were assessed. Bonferroni's correction for multiple testing was applied.
Secondly, multivariable Cox regression analysis was performed of all PT parameters for each modality separately with a backwards Wald test (p value significance threshold of 0.157 according to the Tripod statement [30,31]). All quantitative parameters per modality were corrected for significant clinical parameters (gender, age, T-stage, N-stage, PT location, intoxications) by combination in the backwards Wald elimination analysis.
Thereafter, a CoxBoost analysis was performed to fit a Cox proportional-hazards model by component-wise likelihoodbased boosting, to deal with the amount of features relative to the events. Internal validation was performed, using bootstrap cross-validation with 500 bootstrap samples. Due to lacking of LNM parameters in N0 patients, these LNM parameters were excluded in the multimodality CoxBoost analysis in order to remain statistically robust.
All predictive PT parameters were given a score 1 when they were higher than the parameter's median value, which was based on all included patients. By summing up the points, a risk stratification system was constructed. Thereafter, RFS and OS were assessed, stratified for T-stage, AJCC (7th edition), and risk scores (log-rank test; Kaplan-Meier curves).

Patient characteristics
Between 2013 and 2018, 81 patients were consecutively recruited ( Fig. 1). Nine patients were excluded because of non-curative or surgical treatment and 2 because of significant low image quality.
The final study population consisted of 70 patients (Table 1) with a PT located in the oropharynx (n = 56) or hypopharynx (n = 14). Among the oropharyngeal tumors, the HPV status of 24 patients was positive (43%). Fifty-four patients received concurrent cisplatin-based chemoradiotherapy. Ten patients received weekly cetuximab with concurrent radiotherapy (70Gy). Six patients received radiotherapy only.

Associations of imaging parameters per subgroup
Seventy PT ROIs were drawn and 59 lymph node metastasis ROIs (largest LNM) on each modality ( Table 1). The comparison of both observers resulted in no significant different v a l u e s a n d a h i g h i n t e r o b s e r v e r c o r r e l a t i o n (Supplement 1). A Dice index in primary tumors of 0.88 at the DWI/IVIM and 0.85 at DCE delineation was found (not tabulated). For LNM, a Dice index of 0.97 at DCE and 0.92 at DWI/IVIM delineation was found (not tabulated). Primary tumor ADC GTV , D, f, and D*, DCE G T V , and all 1 8 F-FDG-PET values (Supplement 2) were significantly higher in advanced T-staged tumors (all p ≤ 0.02). In advanced N-staged tumors, PT ADC GTV and V e were significantly higher (p = 0.021 and p = 0.023, respectively). In HPV-negative tumors, ADC mean , D, D*, SUV max , SUV mean , and SUV peak were significantly higher than HPV-positives (p < 0.043). In patients with intoxications, ADC mean , D, and D* were significantly different among the different categories (all p < 0.027).
In LNM (Supplement 3), K ep and all 18 F-FDG-PET parameters were significantly higher in advanced N-stages (p = 0.025, p ≤ 0.016, respectively). In HPV-negative tumors, D was found to be significantly higher (p = 0.002) and D* lower (p = 0.007) than in HPV-positive tumors. In patients with intoxications, f was found to be significantly lower (p = 0.026).

Inter-modality correlations
The inter-modality correlation in PT between 18

Locoregional recurrence-free survival
The univariate analysis (Table 2) showed that HPV-negative status and the combined intoxications were associated with locoregional recurrence (LRF; p = 0.036, p = 0.031, respectively). High ADC GTV , DCE GTV , K trans , V e , and TLG values of primary tumors were significantly associated with LRF (all p ≤ 0.047).
The multivariate analysis per modality (Table 2), corrected for significant clinical parameters (i.e., HPV and intoxications), showed that high primary tumor ADC GTV , DCE GTV , K trans , V e , and MATV remained predictive for LRF (all p ≤ 0.048). For LNM, only DCE GTV remained significantly predictive for LRF (p = 0.018). The subgroup analysis in HPV-negative patients is shown in Supplement 7.
The multivariable CoxBoost analysis (Table 4), combining all modalities and clinical parameters, showed that HPV status, intoxications, ADC GTV , K trans , and V e remained predictive for LRF (C-index of 0.546). The log-rank test (Fig. 2) Fig. 1 The workflow in our prospective study including the inclusion of eligible patients, delineation of the primary tumor and lymph node metastases of the final included patients, the extraction of quantitative imaging parameters, and metastases predictive assessment of locoregional recurrence-free survival and overall survival using the extracted parameters of the primary tumor and lymph node showed that these risk factors were significantly predictive (p = 0.023) for LRF (Fig. 2b), whereas risk stratification per T-stage (Fig. 2a) was not significantly predictive (p = 0.92).

Overall survival
Primary tumor univariate analysis (Table 3) showed that clinical parameters HPV status, PT location, intoxications (p ≤ 0.047), and imaging parameters ADC GTV , ADC mean , D*, D, DCE GTV , V e , MATV, and SUV max, were significantly associated with OS (all p ≤ 0.047). For LNM, SUV max , SUV mean , and SUV peak were associated with OS (all p ≤ 0.015).
In multivariate analysis per single modality (Table 3) The multivariable CoxBoost analysis combining all PT parameters of all modalities, including all clinical parameters (Table 4), shows that N-stage, HPV status, PT location, intoxications, PT ADC GTV , ADC mean , D*, K trans , V e , SUV max , and TLG remain predictive for OS, with a C-index of 0.664.

Discussion
In this study, correlations between pretreatment DWI/IVIM, DCE-MRI, and 18 F-FDG-PET/CT parameters were assessed in order to capture predictive tumor characteristics for LRFS and OS in pharyngeal SCC patients treated with (chemo)radiotherapy.

Tumor characteristics
Advanced stage tumors (high T-stage) and HPV-negative status had significant higher diffusion (high ADC mean , D), higher permeability (K trans , V e ), and lower perfusion (low f and D*), implying different tumor characteristics than early stage and  Univariable and multivariate Cox regression analysis for locoregional recurrence of primary tumor and lymph node metastasis imaging parameters, compared between responders and non-responders. In the multivariate analysis, all parameters per modality were combined, which lead to a loss of intoxications and TLG as remaining predictive parameters for locoregional recurrence HPV-positive tumors. These parameters were also found to be associated with an adverse outcome. This is in line with literature, which described the decrease of cellularity due to apoptosis/necrosis (increased ADC mean and D) to be associated with treatment resistance and thereby with poor prognosis [18]. In contrast, in studies which excluded areas of necrosis in the ROI, lower ADC values were found in high-grade tumors with high cellularity. In the current study, HPV-negative patients had a higher ADC value than HPV-positive patients, which was in line with other studies regardless of including [10] or excluding [32][33][34][35] necrotic areas. An increase of permeability (increased K trans ) is possibly due to tumor neoangiogenesis, which increases immature incompetent vessel leakage, thereby increasing the fraction in the extracellular extravascular space (increased V e ), causing higher interstitial fluid pressure and lower flow [16,36]. The reduced perfusion (low blood flow and volume; low D* and f, respectively) results in worse access to chemotherapeutic drugs and oxygen for radiosensitivity, and is associated with an adverse outcome. This reduced perfusion was found in larger, more advanced stage tumors, and is indicative for low microvessel density, low velocity and hypoxia, due to the incompetent microvessels and increased interstitial pressure [16,18,37,38]. A high/or increased metabolism was also associated with adverse outcome, which might be due to a high/increased glucose demand of advanced staged tumors [39], due to proliferating malignant cells and stromal tissue. In contrast, reduced metabolism in the tumoral center due to diminished access of nutrition and oxygen supply, leading to necrosis with hypoxia, was also associated with adverse outcome [40]. These tumor characteristics might be used to target subvolumes for dose-paint RT [2,3,15].

Recurrence-free survival
In the present study, the combination of HPV status, tumor volume (ADC GTV ), high K trans , and high V e showed more predictive potential for locoregional control than the clinically used risk stratification per T-stage. The more of these adverse factors, the worse the locoregional-free survival was. The previously described predictive value of K trans [21,23] and V e was confirmed in this study [22,23]. In contrast, Ng et al [41]  In a, the recurrence-free survival is shown, which is not significantly predictive. In b, patients were given a risk score by the amount of risk factor points. These risk factors (each with a score of 1 point) were summed up when the predictive quantitative parameters that are higher than the median value of the quantitative parameter or positive clinical parameter (HPV, intoxications, hypopharyngeal PT location or N-stage > 1). The median value of quantitative parameters was calculated based on all included patients. This risk score stratification system is found significantly predictive. In c, the overall survival is shown, which is stratified for T-stage, which is not significant predictive. In d, the overall survival is shown, stratified for the risk score groups, which is found significantly predictive Univariable and multivariable Cox regression analysis of PT and LNM between survivors and patients who died. In the multivariable analysis, all parameters per modality were combined, which lead to a loss of hypopharyngeal PT location, intoxications, ADC mean , D, K trans , SUVmean, SUV peak and TLG, and LNM SUV peak as remaining predictive parameters for OS found K ep , SUV max , TLG, and LNM V e , and ADC mean as predictive parameters for LRFS. However, their chemotherapy scheme was uncommon, delineation was performed on the single-slice largest diameter, and HPV status was not assessed. In this study, ADC mean was not found to be predictive for LRFS, which was also confirmed by King et al [19].
However, in small studies (N = 17 patients, [42], N = 40, [43], N = 32, [18]) with single modality predictive assessment, a low ADC mean was found predictive for LRFS. Furthermore, we found that 18 F-FDG-PET/CT parameters did not remain predictive when combining modalities, which was in line with Ng et al [22]. However, in single modality studies [9,23], SUV parameters were found predictive for LRFS. Such discrepancies in predictive value may be explained by factors such as sample sizes, treatment protocols, and multivariable Cox regression analysis with or without inclusion of important clinical parameters [22].

Overall survival
The combination of all modalities showed that N-stage, HPVnegative status, hypopharyngeal PT location, and intoxication were risk factors for adverse overall survival. This was in line with other studies who found hypopharyngeal PT location [44], alcohol use [23], and HPV status [24] as predictors. Besides, a large tumor volume (ADC GTV , DCE GTV , MATV), K trans , and V e (as were predictive for adverse LRFS), also high ADC mean , D*, SUV max , and TLG, were predictive for adverse OS. The more of these adverse factors, the worse the overall survival was. Previous studies performing multivariable analysis were in line with these findings and found that V e [22] and K trans were predictive for OS. The pretreatment finding of low ADC mean was associated with highly cellular tumors including rapidly dividing cells, which are more sensitive to subsequent chemotherapy and radiotherapy and Multivariable CoxBoost regression analysis of primary tumor imaging parameters to predict locoregional recurrence and overall survival. The C-index (taking the area under the curve over time into account) and hazard ratios (HR) are shown. The adverse locoregional recurrence-free survival is predicted significantly by HPV negativity, intoxications, ADC GTV , K trans , and V e . The adverse overall survival predicted significantly by N-stage, HPV negativity, hypopharyngeal tumor location, intoxications, ADC GTV , ADC mean , D*, K trans , V e , SUV max , and TLG therefore associated with a more favorable prognosis [18,19].
A possible explanation for the extra predictors for OS compared with the predictors of LRFS is that certain tumor tissue architecture, e.g., heterogeneous tissue with low diffusion restriction (high ADC mean ) and aggressive high metabolism (high SUV max and TLG), is less sensitive to (chemo)radiotherapy, which additionally decreases tumor control and survival.
In previous studies, smoking, K trans , K ep [23,41], and a heterogeneity 18 F-FDG-PET/CT parameter (18F-FDG-PET uniformity) were reported to be predictive for OS. In contrast, the significant predictive value of the various clinical parameters combined with significant parameters from the whole spectrum of DWI with IVIM, DCE, and 18 F-FDG-PET/CT was not described previously. The aforementioned risk factors for adverse OS were found significantly predictive, whereas stratification per T-stage was not significantly predictive. This implies an additional predictive value of functional imaging to clinical staging based on morphology.

Complementarity and applicability
In order to improve prediction accuracy, the complementary value of each imaging modality is of importance to capture the whole spectrum of predictive tumoral characteristics, such as tumoral cellularity, necrosis, vascularity, and metabolism [11-13, 16, 24]. Previously described hypothetical overlapping parameters, such as DWI, IVIM (e.g., ADC mean and D, both capturing tissue cellularity indirectly), and DCE (e.g., K trans with K ep ) [11,12,45], correlated not evidently in this study. Different heterogeneous tumor architecture (inflammation, fibrosis, necrosis, and hypoxia) and/or HPV status in more advanced staged tumors might have caused the loss of correlations. Further optimization of protocols and selection and evaluation of qualitative and quantitative parameters is necessary in future studies.
The current study underlines the superiority of combining MRI and 18 F-FDG-PET/CT, which allowed combining significant predictive clinical parameters, such as HPV-negative status, intoxication (smoking/alcohol), hypopharyngeal tumor location, and N-stage with predictive quantitative imaging parameters: ADC mean (DWI) and D* (IVIM), K trans , V e (DCE) and SUV max , and TLG ( 18 F-FDG-PET/CT) for OS. In this way, risk stratification on a patient level was shown to be possible. Furthermore, this might improve patient care and pave the road for personalized treatment options by identification and targeting tumoral subvolumes which are predictive for adverse outcome [7,46].

Limitations
There was a relatively low incidence of events in our cohort; therefore, this study should be considered as hypothesis-generating. Also, selection bias might have occurred by excluding surgical treatment at the prospective selection of patients with curative (chemo)radiotherapy.
Secondly, although T-stage is dependent on the gross tumor volume, they were both included in the predictive analysis, which might have caused confounding bias. Although in this study the GTV was determined on functional imaging maps (ADC and DCE maps), it should be evaluated in future studies whether GTV determined on anatomical MRI sequences is more accurate in the predictive analyses.
Thirdly, we performed pharmacokinetic modeling analysis by using a patient-specific AIF, measured in the external carotid arteries. Flow artifacts as well as high concentrations of contrast agent can result in incorrect amplitude of the arterial concentration. This can affect final calculation of K trans and V e which, as a consequence, are over-estimated (e.g., V e can be larger than 1). We have decided to leave the results as we obtained them, but in future studies, we intend to correct the AIF for flow artifacts.
Finally, the LNM parameters were based on the ROIs of the largest lymph node metastasis, which might falsely ignore the adverse effect of having multiple metastases and consisting of necrotic tumoral areas, which reduced the average tumoral FDG uptake. The LNM parameters were excluded in the multimodality CoxBoost analysis in order to remain statistically robust, which might have limited the predictive value. Moreover, only internal validation by bootstrap crossvalidation was feasible. These limitations were managed by performing a well setup internal validation by bootstrap crossvalidation to test limited parameters repeatedly in a subset.

Conclusion
The combination of clinical parameters, HPV status, with DCE, IVIM-MRI, and 18 F-FDG-PET/CT, provided complementary value in capturing tumor characteristics and improved prediction of locoregional recurrence-free survival and overall survival. HPV-negative status, intoxications, high tumor volume, and permeability and extravascular extracellular space on DCE imaging were predictive for locoregional recurrence and decreased overall survival. Additionally, low cellularity on the ADC map and high metabolism on the 18 F-FDG-PET/CT were additionally predictive for adverse overall survival.
Funding information This research was funded by the Netherlands Organisation for Health Research and Development, grant 10-10400-98-14002. The funding source had no involvement in collection, analysis, data interpretation, writing of the report, nor in the decision to submit the article for publication.

Compliance with ethical standards
Guarantor The scientific guarantor of this publication is Prof. Dr. Jonas Castelijns, MD, PhD.