Research ArticleClinical Studies

Associations Between Histogram Analysis DCE MRI Parameters and Complex 18F-FDG-PET Values in Head and Neck Squamous Cell Carcinoma

ALEXEY SUROV, LEONARD LEIFELS, HANS JONAS MEYER, KARSTEN WINTER, OSAMA SABRI and SANDRA PURZ
Anticancer Research March 2018, 38 (3) 1637-1642;

Abstract

Aim: To analyze associations between parameters of positron-emission tomography (PET) and histogram analysis values of dynamic contrast-enhanced (DCE) imaging in patients with head and neck squamous cell carcinoma (HNSCC). Patients and Methods: Overall, 28 patients with primary HNSCC of different localizations were involved. 18F-FDG-PET/CT and DCE MR imaging were performed for all patients. Histogram analysis parameters of DCE MRI were calculated. Spearman's correlation coefficient was used to analyze the associations between investigated parameters. Results: In the overall cohort, SUVmax correlated with Kep P10 (ρ=0.43, p=0.027), Kep P25 (0.40, p=0.035), and had a tendency to correlate with median Kep (ρ=0.33, p=0.098). TLG tended to correlate with Kep P25 (0.33, p=0.09) and P10 Ktrans (ρ=0.35, p=0.07). In G1/2 tumors, SUVmax correlated with Kep P10 (ρ=0.645, p=0.032), and tended to correlate with Ktrans mean (ρ=0.54, p=0.089), Ktrans min (ρ=0.58, p=0.06), Ktrans P10 (ρ=0.56, p=0.07), Ktrans P25 (ρ=0.59, p=0.056), and Ktrans median (ρ=0.054, p=0.089), as well with Kep min (ρ=0.56, p=0.07). SUVmean tended to correlate with Kep kurtosis (ρ=0.56, p=0.07). In G3 tumors, no tendencies or statistically significant correlations between the PET and DCE MRI parameters were identified. Conclusion: Tumor metabolism and perfusion show complex associations in HNSCC. These associations depend on tumor grading.

Head and neck squamous cell carcinoma (HNSCC) is the sixth most common cancer worldwide, with an increasing incidence (1). Different imaging techniques, such as computed tomography (CT), positron-emission tomography (PET) and magnetic resonance imaging (MRI), are increasingly used for diagnosis and staging, as well monitoring of treatment of HNSCC (2-5).

Multiparametric imaging is a novel diagnostic trend (4-8). Some reports have indicated that the combination of different imaging modalities can better characterize tumor biology and behavior (4-7). Most frequently, the combination of MRI and PET is used. The combination of diffusion-weighted imaging (DWI) and PET provides information about metabolic activity and tissue architecture in HNSCC (4). Furthermore, retrieved quantified parameters of these modalities, namely standardized uptake value (SUV) and apparent diffusion coefficient (ADC), reflect several histopathological features and can be used as prognostic markers (4). Dynamic contrast-enhanced (DCE) MRI is another imaging modality that can reflect tumor biology (5-9). According to the literature, DCE MRI parameters, namely volume transfer constant (Ktrans), volume of the extravascular extracellular leakage space (Ve) and diffusion of contrast medium back to the plasma (Kep) can quantitatively characterize perfusion and vascularization of tissues (9, 10). It has been reported that DCE MRI can also predict tumor outcome (11, 12).

Some authors also investigated associations between parameters of combined imaging modalities (5-8). Most previous studies analyzed relationships between ADC and SUV. There were only five reports about associations between PET and DCE MRI parameters (5-9). The reported results were controversial (Table I). Most authors did not identify statistically significant correlations between PET and DCE MRI parameters. Only Gawlitza et al. found slight correlations between SUVmean and Ktrans and Kep (7).

We believe that the predominantly negative results of the previous reports may be related to several factors. Firstly, previously, all authors used only mean values of DCE MRI parameters. Secondly, most studies retrieved only two parameters from PET, namely SUVmax and SUVmean. Only one report additionally used total lesion glycolysis (TLG) and metabolic tumor volume (MTV) (5). Thirdly, we hypothesize that associations between DCE MRI and PET parameters may depend on tumor grading.

View this table:
Table I.

Overview of the reported correlations between dynamic contrast-enhanced (DCE) magnetic resonance imaging and positron-emission tomography (PET) parameters.

Therefore, the purpose of this study was to analyze relationships between complex parameters of DCE MRI and PET in patients with HNSCC.

Patients and Methods

This prospective study was approved by the Institutional Review Board (study codes 180-2007, 201-10-12072010, and 341-15-05102015).

Patients. For this study, 28 patients, seven (26%) women and 21 (74%) men, mean age±SD=57.6±10.4 years, range=33-77 years, with different locations of HNSCC were acquired. Tumor localizations were as follows: tongue in seven (25%), tonsil in seven (25%), larynx in five (18.8%), oropharynx in four (14.3%), hypopharynx in four (14.3%), and epipharynx in one (3.6%). Low grade (G1/2) tumors were diagnosed in 11 cases (39%), and high grade (G3) tumor in 17 (61%) patients.

Imaging. PET/CT: In all patients, 18F-FDG-PET/CT (SiemensBiograph 16; Siemens Medical Solutions, Erlangen, Germany) was performed from the skull to the upper thigh after a fasting period of at least 6 hours. Application of 18F-FDG was performed intravenously with a body weight-adapted dose (4MBq/kg, range=168-427MBq, mean±SD=281±61MBq). PET/MRI acquisition started on average 98 minutes (range=60-270 minutes) after 18F-FDG application. In 2/28 patients, a PET/MRI scan was performed prior to PET/CT and in 1/28, a technical defect led to a delayed acquisition start, which explains the late PET/CT image acquisition time in these three patients. Low-dose CT was used for attenuation correction of the PET data.

The acquired PET/CT datasets were evaluated by a Board-certified nuclear medicine and a Board-certified radiologist with substantial PET/CT experience in oncological image interpretation. PET/CT image analysis was performed on a dedicated workstation of Hermes Medical Solutions, Sweden. For each tumor, maximum and mean SUV (SUVmax; SUVmean), TLG and MTV were determined on PET images. Prior to this, tumor margins of the HNSCC were identified on diagnostic CT images and fused PET/CT images and a polygonal volume of interest (VOI), which included the entire lesion in the axial, sagittal and coronal planes, was placed in the PET dataset (SUVmax threshold 40%) (Figure 1). MTV was defined as total tumor volume with an SUV ≥2.5 and was calculated automatically. TLG was also calculated automatically by multiplying the MTV of the primary tumor by its SUVmean (Figure 1).

Figure 1.
Figure 1.

a: 18F-FDG-PET/CT shows a metabolic active hypopharyngeal lesion (arrow). b: A polygonal volume of interest that included the entire lesion in the coronal (b), sagittal (c) and axial (d) planes was placed in the PET dataset (SUVmax threshold 40%). The acquired PET parameters were as follows: SUVmax=22.07, SUVmean=13.92, SUVmin=7.93, metabolic tumor volume=5.74, and total lesion glycolysis=79.97.

DCE: In all patients, dynamic T1-weighted DCE sequence (TR/TE 2.47/0.97 ms, slice thickness 5 mm, flip angle 8°, voxel size 1.2×1.0×5.0 mm) after intravenous application of contrast medium (0.1mmol Gadobutrol per kg of bodyweight; Gadovist®; Bayer Healthcare, Leverkusen, Germany) was performed according to our previous descriptions (10). The following perfusion parameters were acquired: volume transfer constant (Ktrans), volume of the extravascular extracellular leakage space (Ve), and diffusion of contrast medium from the extravascular extracellular leakage space back to the plasma (Kep) (7, 8, 10). The acquired images were transferred to a software module (Tissue 4D; Siemens Medical Systems) and processed offline in DICOM format with custom-made Matlab-based application (The Mathworks, Natick, MA, USA). Thereafter, polygonal regions of interest were automatically drawn on the transferred Ktrans, Ve, and Kep maps on each slice (whole-lesion measure) (13). For every perfusion parameter, mean, maximal, minimal, and median values, as well 10th, 25th, 75th and 90th percentiles (P10, P25, P75, P90) as well histogram-based characteristics (kurtosis, skewness, and entropy) were calculated (Figures 2, 3 and 4).

Statistical analysis. Statistical analysis was performed using SPSS package (IBM SPSS Statistics for Windows, version 22.0; IBM Corporation, Armonk, NY, USA). Collected data were evaluated by means of descriptive statistics. Spearman's correlation coefficient (ρ) was used to analyze associations between investigated parameters. p-Values of less than 0.05 were taken to indicate statistical significance.

Results

Correlation analysis identified several statistically significant correlations between the investigated parameters. For the overall cohort, SUVmax correlated with Kep P10 (ρ=0.43, p=0.027), Kep P25 (0.40, p=0.035), and had a tendency to correlate with median Kep (ρ=0.33, p=0.098). Furthermore, TLG tended to correlate with Kep P25 (ρ=0.33, p=0.09) and Ktrans P10 (ρ=0.35, p=0.07). There were no significant correlations or tendencies for correlation between other parameters.

Figure 2.
Figure 2.

Ktrans image (a) and histogram of Ktrans values (b). The histogram analysis parameters (min-1) were as follows: mean=0.14, min=0.013, max=0.71, P10=0.039, P25=0.065, P75=0.205, P90=0.27, median=0.12, mode=0.042, kurtosis=6.03, skewness=1.34, entropy=3.46.

Figure 3.
Figure 3.

Ve image (a) and histogram of Ve values (b). The histogram analysis parameters (min−1) are as follows: mean=0.33, min=0.031, max=0.99, P10=0.095, P25=0.14, P75=0.5, P90=0.69, median=0.25, mode=0.14, kurtosis=2.87, skewness=0.93, entropy=2.96.

Figure 4.
Figure 4.

Kep image (a) and histogram of Kep values (b). The histogram analysis parameters (min-1) are as follows: mean=0.46, min=0.13, max=0.98, P10=0.28, P25=0.34, P75=0.56, P90=0.65, median=0.45, mode=0.39, kurtosis=2.71, skewness=0.32, entropy=3.52.

In those with G1/2 tumors, SUVmax correlated with Kep P10 (ρ=0.645, p=0.032), and tended to correlate with Ktrans mean (ρ=0.54, p=0.089), Ktrans min (ρ=0.58, p=0.06), Ktrans P10 (ρ=0.56, p=0.07), Ktrans P25 (ρ=0.59, p=0.056), and Ktrans median (ρ=0.54, p=0.089), as well with Kep min (ρ=0.56, p=0.07). SUVmean tended to correlate with Kep kurtosis (ρ=0.56, p=0.07).

In those with G3 tumors, no tendencies or statistically significant correlations between the PET and DCE MRI parameters were identified.

Discussion

To the best of our knowledge, this is the first study to investigate associations between PET and histogram analysis of DCE MRI parameters. As mentioned above, previously, only few studies analyzed these relationships and predominantly did not find significant correlations between the analyzed parameters (5-9). However, it is known that PET can predict tumor aggressiveness in HNSCC (14). Furthermore, PET parameters were significantly associated with expression of p53 and vascular endothelial growth factor (VEGF) (15). On the other hand, DCE MRI parameters have also been reported to reflect several several clinical and histopathological features in HNSCC (16-18). For instance, Bernstein et al. found that pretreatment DCE MRI values predicted response to induction chemotherapy (16). Furthermore, PET and DCE MRI parameters have been reported to be independent prognostic markers predicting overall survival in HNSCC (11, 12). It has also been shown that DCE MRI and PET together predict early distant metastasis in HNSCC (17). Consequently, PET and DCE MRI parameters may be associated with each other. Our results confirmed this assumption. In the present study, several correlations between the analyzed parameters were identified. Here, no significant correlations were identified between mean values of Ktrans, Kep, and Ve and PET parameters. This finding suggests that the mean values of DCE MRI parameters are not sensitive and may explain the predominantly negative results of the previously mentioned reports.

Furthermore, our study identified another phenomenon, namely, associations between PET and DCE MRI parameters depending on tumor grading. This finding is very interestingly and indicates heterogeneity of tissue composition and tumor cell metabolism in different tumor grades. Presumably, low- and high-grade HNSCC may have different relationships between tumor cell count, microvessel density, and metabolic activity. This might have resulted in the different associations between the investigated parameters in the present study.

Additionally, our study showed that only some Ktrans and Kep values correlated with PET parameters but not Ve. Ktrans reflects the permeability of microvessels and its association with metabolic activity would seem to be logical (7, 8, 10). Kep represents the diffusion of contrast medium from the extracellular space back into the plasma (7, 8, 10). Consequently, it can also be associated with different PET parameters. As reported previously, Ve reflects the extravascular extracellular space and, therefore, should be associated with cell density (7, 8, 10). Previously it was shown that PET parameters did not correlate significantly with cell count in HNSCC (4). Therefore, our finding that none of Ve values correlated significantly with PET parameters seems to be logical.

As seen, overall, there were only small number of statistically significant correlations and trends between the investigated parameters. A possible explanation for this finding may be that perfusion and metabolic activity in HNSCC may have non-linear associations. This assumption should be tested in further studies with more patients.

We hypothesize that both DCE MRI and PET values may reflect some histopathological features of HNSCC. The search for these pathological parameters is also a purpose for further research.

In conclusion, our study showed several significant associations between PET parameters and histogram analysis of DCE MRI values in HNSCC. Here, mean values of DCE MRI parameters did not correlate with PET parameters. The identified associations depended on tumor grading.

Footnotes

  • Compliance with Ethical Standards

    All procedures performed in the study were in accordance with the ethical standards of the Institutional Research Committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards

  • Conflicts of Interest

    There was no conflict of interest.

  • Ethical Approval

    The study was approved by the Institutional Review Board of the University of Leipzig (study codes 180-2007, 201-10-12072010, and 341-15-05102015).

  • Informed Consent

    For this type of study, informed consent was obtained from all individual participants included in the study.

  • Received December 2, 2017.
  • Revision received January 14, 2018.
  • Accepted January 17, 2018.

References

  1. ↵
    1. Braakhuis BJ,
    2. Leemans CR,
    3. Visser O
    : Incidence and survival trends of head and neck squamous cell carcinoma in the Netherlands between 1989 and 2011. Oral Oncol 50: 670-675, 2014.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Lai V,
    2. Khong PL
    : Updates on MR imaging and 18F-FDG PET/CT imaging in nasopharyngeal carcinoma. Oral Oncol 50: 539-548, 2014.
    OpenUrlPubMed
    1. De Bree R,
    2. Castelijus JA,
    3. Hoekstra OS,
    4. Leemans CR
    : Advances in imaging in the work-up of head and neck cancer patients. Oral Oncol 45: 930-935, 2009.
    OpenUrlCrossRefPubMed
  4. ↵
    1. Surov A,
    2. Stumpp P,
    3. Meyer HJ,
    4. Gawlitza M,
    5. Höhn AK,
    6. Boehm A,
    7. Sabri O,
    8. Kahn T,
    9. Purz S
    : Simultaneous (18)F-FDG-PET/MRI: Associations between diffusion, glucose metabolism and histopathological parameters in patients with head and neck squamous cell carcinoma. Oral Oncol 58: 14-20, 2016.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Han M,
    2. Kim SY,
    3. Lee SJ,
    4. Choi JW
    : The Correlations Between MRI Perfusion, Diffusion Parameters, and 18F-FDG PET Metabolic Parameters in Primary Head-and-Neck Cancer: A Cross-Sectional Analysis in Single Institute. Medicine (Baltimore) 94: e2141, 2015.
    OpenUrlPubMed
    1. Leifels L,
    2. Purz S,
    3. Stumpp P,
    4. Schob S,
    5. Meyer HJ,
    6. Kahn T,
    7. Sabri O,
    8. Surov A
    : Associations between 18F-FDG-PET, DWI, and DCE parameters in patients with head and neck squamous cell carcinoma depend on tumor grading. Contrast Media Mol Imaging 2017: 5369625, 2017.
    OpenUrlPubMed
  7. ↵
    1. Gawlitza M,
    2. Purz S,
    3. Kubiessa K,
    4. Boehm A,
    5. Barthel H,
    6. Kluge R,
    7. Kahn T,
    8. Sabri O,
    9. Stumpp P
    : In vivo correlation of glucose metabolism, cell density and microcirculatory parameters in patients with head and neck cancer: Initial results using simultaneous PET/MRI. PLOS ONE 10: e0134749, 2015.
    OpenUrlPubMed
  8. ↵
    1. Bisdas S,
    2. Seitz O,
    3. Middendorp M,
    4. Chambron-Pinho N,
    5. Bisdas T,
    6. Vogl TJ,
    7. Hammerstingl R,
    8. Ernemann U,
    9. Mack MG
    : An exploratory pilot study into the association between microcirculatory parameters derived by MRI-based pharmacokinetic analysis and glucose utilization estimated by PET-CT imaging in head and neck cancer. Eur Radiol 20: 2358-2366, 2010.
    OpenUrlPubMed
  9. ↵
    1. Jansen JF,
    2. Carlson DL,
    3. Lu Y,
    4. Stambuk HE,
    5. Moreira AL,
    6. Singh B,
    7. Patel SG,
    8. Kraus DH,
    9. Wong RJ,
    10. Shaha AR,
    11. Shah JP,
    12. Shukla-Dave A
    : Correlation of a priori DCE-MRI and (1)H-MRS data with molecular markers in neck nodal metastases: Initial analysis. Oral Oncol 48: 717-722, 2012.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Surov A,
    2. Meyer HJ,
    3. Gawlitza M,
    4. Höhn AK,
    5. Boehm A,
    6. Kahn T,
    7. Stumpp P
    : Correlations between DCE MRI and histopathological parameters in head and neck squamous cell carcinoma. Transl Oncol 10: 17-21, 2017.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Chan SC,
    2. Cheng NM,
    3. Hsieh CH,
    4. Ng SH,
    5. Lin CY,
    6. Yen TC,
    7. Hsu CL,
    8. Wan HM,
    9. Liao CT,
    10. Chang KP,
    11. Wang JJ
    : Multiparametric imaging using 18F-FDG PET(CT heterogeneity parameters and functional MRI techniques: prognostic significance in patients with primary advanced oropharyngeal squamous cell carcinoma treated with chemotherapy. Oncotarget 8: 62606-62621, 2017.
    OpenUrlPubMed
  12. ↵
    1. Ng SH,
    2. Lin CY,
    3. Chan SC,
    4. Yen TC,
    5. Liao CT,
    6. Chang JT,
    7. Ko SF,
    8. Wang HM,
    9. Huang SF,
    10. Lin YC,
    11. Wang JJ
    : Dynamic contrast-enhanced MR imaging predicts local control in oropharyngeal or hypopharyngeal squamous cell carcinoma treated with chemoradiotherapy. PLOS ONE 8: e72230, 2013.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Meyer HJ,
    2. Leifels L,
    3. Schob S,
    4. Garnov N,
    5. Surov A
    : Histogram analysis parameters identify multiple associations between DWI and DCE MRI in head and neck squamous cell carcinoma. Magn Reson Imaging 45: 72-77, 2017.
    OpenUrlPubMed
  14. ↵
    1. Kitagawa Y,
    2. Sano K,
    3. Nishizawa S,
    4. Nakamura M,
    5. Ogasawara T,
    6. Sadato N,
    7. Yonekura Y
    : FDG-PET for prediction of tumour aggressiveness and response to intra-arterial chemotherapy and radiotherapy in head and neck cancer. Eur J Nucl Med Mol Imaging 30: 63-71, 2003.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Grönroos TJ,
    2. Lehtiö K,
    3. Söderström KO,
    4. Kronqvist P,
    5. Laine J,
    6. Eskola O,
    7. Viljanen T,
    8. Grénman R,
    9. Solin O,
    10. Minn H
    : Hypoxia, blood flow and metabolism in squamous-cell carcinoma of the head and neck: correlations between multiple immunohistochemical parameters and PET. BMC Cancer 14: 876, 2014.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Bernstein JM,
    2. Kershaw LE,
    3. Withey SB,
    4. Lowe NM,
    5. Homer JJ,
    6. Slevin NJ,
    7. Bonington SC,
    8. Carrington BM,
    9. West CM
    : Tumor plasma flow determined by dynamic contrast-enhanced MRI predicts response to induction chemotherapy in head and neck cancer. Oral Oncol 51: 508-513, 2015.
    OpenUrlCrossRefPubMed
  17. ↵
    1. Chin SC,
    2. Lin CY,
    3. Huang BS,
    4. Tsang NM,
    5. Fan KH,
    6. Ku YK,
    7. Hsu CL,
    8. Chan SC,
    9. Huang SF,
    10. Li CH,
    11. Tseng HJ,
    12. Liao CT,
    13. Liu HL,
    14. Sung K
    : Pretreatment dynamic contrast-enhanced mri improves prediction of early distant metastases in patients with nasopharyngeal carcinoma. Medicine 95: e2567, 2016.
    OpenUrlPubMed
  18. ↵
    1. Chawla S,
    2. Kim S,
    3. Dougherty L,
    4. Wang S,
    5. Loevner LA,
    6. Quon H,
    7. Poptani H
    : Pretreatment diffusion-weighted and dynamic contrast-enhanced MRI for prediction of local treatment response in squamous cell carcinomas of the head and neck. Am J Roentgenol 200: 35-43, 2013.
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Reprints and Permissions
Associations Between Histogram Analysis DCE MRI Parameters and Complex 18F-FDG-PET Values in Head and Neck Squamous Cell Carcinoma
ALEXEY SUROV, LEONARD LEIFELS, HANS JONAS MEYER, KARSTEN WINTER, OSAMA SABRI, SANDRA PURZ
Anticancer Research Mar 2018, 38 (3) 1637-1642;
Twitter logo Facebook logo Mendeley logo

Jump to section

Keywords