Role of ¹⁸F-FDG PET-CT in head and neck squamous cell carcinoma

SUMMARY

The role of PET-CT imaging in head and neck squamous cell carcinoma during pre-treatment staging, radiotherapy planning, treatment response assessment and post-therapy follow-up is reviewed with focus on current evidence, controversial issues and future clinical applications. In staging, the role of ¹⁸F-FDG PET-CT is well recognized for detecting cervical nodal involvement as well as for exclusion of distant metastases and synchronous primary tumours. In the evaluation of treatment response, the high negative predictive value of ¹⁸F-FDG PET-CT performed at least 8 weeks from the end of radio-chemotherapy allows prevention of unnecessary diagnostic invasive procedures and neck dissection in many patients, with a significant impact on clinical outcome. On the other hand, in this setting, the low positive predictive value due to possible post-radiation inflammation findings requires special care before making a clinical decision. Controversial data are currently available on the role of PET imaging during the course of radio-chemotherapy. The prognostic role of ¹⁸F-FDG PET-CT imaging in head and neck squamous cell carcinoma is recently emerging, in addition to the utility of this technique in evaluation of the tumour volume for planning radiation therapy. Additionally, new PET radiopharmaceuticals could provide considerable information on specific tumour characteristics, thus overcoming the limitations of ¹⁸F-FDG.

Introduction

¹⁸F-fluorodeoxy-D-glucose positron emission tomography- computed tomography (¹⁸F-FDG PET-CT) has become an important diagnostic tool for evaluation of head and neck squamous cell carcinomas (HNSCCs), and is applied in various clinical settings, ranging from pre-treatment staging to radiotherapy planning, treatment response assessment and post-therapy follow-up ^{1 2}. Although ¹⁸F-FDG is the most commonly used PET tracer for oncologic purposes, its use in HNSCCs suffers from some limitations due to the complex anatomy of this region and the small size of the anatomical structures, as well as the physiological uptake of ¹⁸F-FDG in normal organs that may influence image interpretation. ¹⁸F-FDG uptake reflects glucose metabolism and can be observed in several normal tissues with wide variability of the normal pattern, including brain, vocal cords, salivary glands, cervical muscles, lymphoid tissue and brown fat, as well as in various benign tumours, such as common Warthin's tumour ³. Moreover, the inflammatory processes that occur in patients submitted to surgery or radiotherapy are a frequent cause of false positive PET results, since the activated inflammatory cells show increased ¹⁸F-FDG uptake ^{4 5}. Finally, artefacts related to patient movement or metal dental prostheses may further limit interpretation of PET images, thus requiring non-attenuation corrected PET data evaluation ^{6 7}. PET-CT scanners currently allow quick and high resolution imaging, and can correlate anatomical location with functional information. The recent introduction of whole-body PET-magnetic resonance imaging (MRI) in clinical practice offers new opportunities for integrated functional-anatomic imaging ^{8 9}. This review will focus on the use of ¹⁸F-FDG PET-CT in various clinical scenarios of HNSCCs with attention to PET radiopharmaceuticals other than ¹⁸F-FDG, and the new perspectives offered by PET-MRI.

Pre-treatment staging

Accurate assessment of disease extension is essential to plan the most appropriate treatment, with important implications for patient outcomes. In clinical practice, the conventional diagnostic strategy of HNSCC patients includes detailed physical examination and endoscopy followed by imaging modalities such as neck ultrasound, neck MRI and neck-chest CT for the assessment of disease extent and diagnosis of synchronous second primary tumours (SPTs). Recent studies have shown that ¹⁸F-FDG PET-CT is more accurate than conventional staging in HNSCCs, thus resulting in a change of therapeutic management in about one-third of patients ^10-12.

Primary tumour assessment

Even though ¹⁸F-FDG PET-CT detects primary HNSCC with high sensitivity (> 95%), primary tumour assessment is generally performed with clinical examination, endoscopy and MRI ^{13 14}. The main limitation of standard PET-CT, if performed with low-dose unenhanced CT, is its inability to accurately assess the site, extent of tumour spread and relationship between the tumour and adjacent structures. This limitation can only partially be overcome by with contrast-enhanced PET-CT ¹². Initial reports on the use of integrated PET-MRI in HNSCCs showed better tumour delineation and good correlation between the metabolic parameters obtained using PET-MRI and PETCT ^{9 15}. In patients with cervical lymph node metastasis from a carcinoma of unknown origin, ¹⁸F-FDG PET-CT represents a useful diagnostic tool to detect the primary tumour, with a detection rate of 25-38.5% ^16-18. Future studies with PET-MRI will determine whether the technique can improve the detection rate of occult primary head and neck tumours ⁹.

Cervical lymph node assessment

The main indication of ¹⁸F-FDG PET-CT in newly diagnosed HNSCCs is detection of cervical lymph node involvement, which is one of the most important prognostic factors. With regards to node involvement, the main limitation of all imaging techniques is the high rate of false negative results, staged as cN0, which turn out to be pN+ after neck dissection. Sentinel lymph node biopsy has several limitations that limit its routine clinical employment in the head and neck ¹⁹.

This awareness is the basis of the indications for prophylactic neck treatment by international guidelines ²⁰; however, this approach may lead clinicians to unnecessarily treat a number of necks with relevant morbidity. Data from the literature support the superiority of PET-CT over morphologic imaging in detecting lymph node involvement ¹². A clear advantage of functional imaging is that an alteration in the size or structure of lymph nodes is not required to detect lesions (Fig. 1). However, it must be stressed that small lymph node lesions may be missed (possible false negative results), and that inflamed lymph nodes may take up the tracer (possible false positive results). In particular, the finite spatial resolution of a PET-CT scanner (4-6 mm) limits its sensitivity for microscopic disease that is detectable only by histopathology after neck dissection. For this reason, a negative ¹⁸F-FDG PET-CT scan for lymph node involvement does not justify a "wait and see" approach in all cases; the decision to perform neck dissection in patients with negative morphological and functional imaging still relies essentially on the evaluation of risk factors and tumour characteristics such as T-stage and histopathological features ^{12 21 22}. Preliminary data with PET-MRI report good sensitivity (85%) and specificity (92%) for nodal staging, higher than PET-CT, but with the same limitation of under-staging a certain number of patients with micrometastases ⁹.

Fig. 1.

¹⁸F-FDG PET-CT performed at staging in a patient with laryngeal carcinoma. Whole body PET-CT (left panel) shows intense ¹⁸F-FDG uptake of the primary tumour as well as two lateral cervical nodes, one of which on the right side demonstrates a short axis less than 10 mm in the transaxial plane of CT (middle and right panels; white arrow). These findings have both prognostic and therapeutic implications: the patient is candidate for radiochemotherapy with bilateral cervical nodal irradiation.

Distant metastasis assessment

The screening for distant metastases is important in patients with advanced disease, especially with nodal involvement, and in naso- and hypopharyngeal carcinomas. The most common sites are lungs, bone and liver (Fig. 2). In this context, a higher accuracy of ¹⁸F-FDG PET-CT than CT for detection of distant metastasis has been clearly demonstrated, except for small lung lesions. Moreover, in approximately 13% of cases the improvement in detection of distant lesions leads to a change of management with important consequences for patient survival ^23-25.

Fig. 2.

¹⁸F-FDG PET-CT performed at staging in a patient with laryngeal carcinoma. Whole body PET-CT (left panel) shows intense ¹⁸F-FDG uptake of the primary tumour and a right lateral cervical lymphadenopathy. In addition, a round solid pulmonary nodule with high ¹⁸F-FDG uptake is evident in the apical segment of the lower lobe of the right lung, suggesting a metastatic lesion (middle and right panels).

Second primary tumour assessment

Second primary tumours (SPTs) can occur in 5-10% of HNSCC patients and are more frequent in the head and neck region, oesophagus and lungs (Fig. 3). SPTs are notably the first cause of death with a decisive impact on overall survival rates of early stage HNSCC patients ²⁶. ¹⁸F-FDG PET-CT is an accurate method to detect second primaries, with a high negative predictive value and a relatively lower positive predictive value as inflammation and benign hyperplasia in the head and neck region, and benign or precancerous intestinal polyps can result in false positive PET findings ^{12 27}. Nonetheless, PET-CT, by detecting second primary lesions, can impact both the treatment choice and, most of all, the overall survival of early stage HNSCCs that already have a very good disease specific survival ²⁸.

Fig. 3.

¹⁸F-FDG PET-CT performed at staging in a patient with laryngeal carcinoma. Whole body PET-CT (left panel) shows intense ¹⁸F-FDG uptake in the primary tumour as well as in a lung mass located in the apical segment of the right lower lobe; mediastinal lymph nodes with increased activity are also associated (left and middle panels). These findings suggest a second primary tumour.

Prognostic significance of pre-treatment PET-CT

Despite the substantial changes in treatment strategies over the past two decades, disease control in HNSCCs is still heterogeneous, since patients with similar clinicopathological features undergoing the same treatment may differ widely in response to treatment and prognosis ²⁹. Thus, the identification of further prognostic factors along with TNM staging could help to identify high-risk patients who respond poorly to conventional therapy, and could benefit from intensification or switch of treatment modality. In this context, the prognostic value of various ¹⁸F-FDG pre-treatment parameters including maximal and mean standardized uptake value (SUV_max and SUV _mean), metabolic tumour volume (MTV) and total lesion glycolysis (TLG) is under investigation ³⁰. The utility of these parameters has been evaluated with conflicting results ^30-32. The main criticism is the variable behaviour of the different metabolic parameters, each showing potential pitfalls in terms of calculation and reproducibility.

Radiotherapy planning

Molecular imaging with ¹⁸F-FDG provides a unique opportunity for radiation treatment planning in HNSCCs in terms of selection and delineation of target volumes as well as dose planning ³³. Looking at target volume selection, the added value of ¹⁸F-FDG PET-CT is identification of the occult primary tumour in patients with cervical lymph node metastasis with the possibility of decreasing the radiotherapy target volume, and consequently the acute and late side effects of radiotherapy, as well as detection of distant metastases with high sensitivity ^16-18. For accurate delineation of target volumes and organs at risk, the goal of using molecular imaging is to optimize the treatment plan thanks to its superior sensitivity and contrast resolution over anatomical imaging techniques ³⁴. In the initial experience of Daisne et al., gross tumour volume (GTV) delineated from ¹⁸F-FDG PET was closest to the pathologic GTV from surgical specimens, and significantly smaller than GTV delineated by CT and MRI ³⁵. However, subsequent data were less consistent ³⁴. The main drawback is the lack of a standardized method for functional volume segmentation, which heavily influences the volume and shape of the resulting GTV ^34-36. Further applications of ¹⁸F-FDG PET-CT for radiotherapy planning are under clinical investigation, and include the possibility of directing dose escalation to ¹⁸F-FDG-avid sub-volumes of the tumour as well as adapting the radiotherapy plan during treatment thanks to the information on the biological and molecular tumour changes induced by therapy ^34-37. Moreover, the possibility of targeting radiation-resistant regions within the tumour on the basis of biologic information of molecular imaging is under investigation, for example the identification of the hypoxic volume within the GTV by using hypoxia-related PET tracers to deliver higher doses to hypoxic cells ³⁸.

Treatment response assessment

An accurate evaluation of response is essential in the management of patients with HNSCC treated with radio- chemotherapy. ¹⁸F-FDG PET-CT is commonly used to assess treatment response, since it identifies viable tumour within residual masses, thus overcoming the known limitations of morphological imaging modalities ^39-42. In particular, ¹⁸F-FDG PET-CT has a high negative predictive value (> 95%) that can spare the patient of unnecessary diagnostic invasive procedures and neck dissection in many cases (about 75%), with a significant impact on patient outcome and morbidity, and a low risk of under-treatment (about 2%) ^{12 41}. Several studies have demonstrated that in patients with a complete metabolic response neck dissection can be avoided, even in the presence of residual node abnormalities by conventional imaging ^{43 44} (Fig. 4); considering the additional morbidity of neck dissections in irradiated necks, this approach can be considered a relevant achievement in head and neck oncology. Unfortunately, the positive predictive value is low, due to the high number of false positive results related to post-radiation inflammation ⁴⁵. To reduce the number of false positive findings, it is crucial to accurately select high risk patients (i.e. HPV negative), to know radiation treatment volume and choose the timing for PET-CT post-treatment assessment. There is general consensus to set the optimum time for ¹⁸F-FDG PET-CT at 8-12 weeks, thereby reducing both false positive and false negative findings, while the the latter is related to the presence of undetectable microscopic residual disease. ¹⁸F-FDG PET-CT was considered to be useful in selective cases to assess response to induction treatment in patients enrolled in larynx preservation trials to assess the risks and benefits of treatment ⁴⁶. Correlating the post-treatment metabolic response with clinical outcomes, the data indicate that ¹⁸F-FDG PETCT performed at the end of radio-chemotherapy provides prognostic information, as it strongly correlates with local and regional control and survival. However, controversial results are reported regarding the role of PET-CT performed early during treatment ^{5 29 47 48}. In a series of 26 patients treated with radio-chemotherapy for HNSCC, our group did not find a significant correlation between the "early" changes of FDG uptake in the primary tumour and lymph node involvement and local and regional control, respectively, or between the overall metabolic response at PET-CT and clinical outcome ²⁹. Our results on the unreliability of ¹⁸F-FDG PET-CT for the early assessment of response to radiochemotherapy in HNSCC are in contrast with the recent study by Hentschel et al. who showed that a decrease of 50% or more of SUVmax from the beginning (0 Gy) to week 1 or 2 of treatment (10 or 20 Gy) is a potential prognostic marker for patients with HNSCC ⁴⁸.

Fig. 4.

¹⁸F-FDG PET-CT at baseline (right panel) and 8 weeks after the end of radio-chemotherapy (left panel). At baseline, intense ¹⁸F-FDG uptake is evident in the left oropharyngeal region (white arrow); after radio-chemotherapy, ¹⁸F-FDG uptake was no longer visible, suggesting complete metabolic response to treatment.

Post-therapy follow-up

Despite initial aggressive treatment, loco-regional or distant recurrence can occur in HNSCC patients, especially within the first year; early detection of loco-regional disease may improve survival by increasing the effectiveness of salvage therapy, which is the most effective modality in this setting; in contrast, the real advantage of early detection of distant metastases in asymptomatic patients is still unclear 12. At morphologic imaging and physical examination, detection of loco-regional recurrence or residual disease may be difficult due to the presence of treatment-induced changes that could not be differentiated from residual/recurrent disease. Several studies involving both patients with suspected recurrence and those without clinical symptoms have demonstrated that PET-CT imaging is superior to physical examination and conventional imaging to detect recurrent loco-regional disease, as well as distant lesions and metachronous primary tumours ^{42 49-53}. For patients with a PET positive result, biopsy is recommended due to a relatively high false positive rate related to post-treatment inflammation ¹². A recent systematic review and meta-analysis of studies assessing the diagnostic performance of ¹⁸F-FDG PET and PET-CT in response assessment and surveillance imaging of HNSCC patients reported a high NPV (> 94%) and a suboptimal PPV (< 60%) both for the primary site and cervical nodes ⁵⁴.

PET radiopharmaceuticals other than ¹⁸F-FDG

It is well known that ¹⁸F-FDG is not a specific marker of malignancy, since ¹⁸F-FDG uptake in normal or active inflammatory tissues may occur, thus limiting image interpretation. To overcome these drawbacks, more specific PET tracers reflecting specific biologic tumour characteristics may be useful to differentiate tumours by inflammation; these tracers, which are under clinical investigation, include ¹⁸F-fluorothymidine for DNA synthesis (¹⁸F-FLT), ¹⁸F-fluoroethyl-L-thyrosine (¹⁸F-FET) and L-methyl-¹¹Cmethionine (¹¹C-MET) for amino acid uptake and protein synthesis ^55-57. Among these tracers, ¹⁸F-FLT seems to be the most promising because of its specificity ³³. Moreover, tracers that allow in vivo detection and quantification of tumour hypoxia such as ¹⁸F-fluoromisonidazole (¹⁸F-FMISO) or ¹⁸F-fluoroazomycin-arabinofluranoside (¹⁸F-FAZA) and ¹⁸F-3-[F]fluoro-2-(4-((2-nitro-1H-imidazol-1-yl)methyl)-1H-1,2,3,-triazol-1-yl)-propan-1-ol (¹⁸F-HX4) are of growing interest; in fact, hypoxia is not a rare event in HNSCCs with important consequences on treatment response since hypoxic cells are more resistant to the cytotoxic effects of ionizing radiation ^{55 57}-⁵⁹ (Table I).

Table I.

PET radiopharmaceuticals other than ¹⁸F-FDG.

PET Radiopharmaceutical	Molecular Target	Indications	Clinical application	References
18F-FLT	Proliferation	Staging-restaging Response evaluation Adaptive radiotherapy	Experimental	56, 60-62
18F-FET 11C-MET	Protein synthesis	Staging-restaging Adaptive radiotherapy	Experimental Early clinical	63, 64
18F-FMISO 18F-FAZA 18F-HX4	Hypoxia	Staging Response Evaluation Adaptive radiotherapy	Experimental	59, 65-70
68Zr/124I-anti EGFR	Targeting of EGFR	Targeted therapies	Experimental	71
18F-Annexin	Apoptosis	Response evaluation	Experimental	72
18F-Galacto-RGD	Neoangiogenesis	Response evaluation Targeted therapies	Experimental	73

FLT = fluorothymidine; FET = fluoroethyl-L-thyrosine; MET = L-methyl-methionine; MISO = fluoromisonidazole; FAZA = fluoroazomycin-arabinofluranoside; HX4 = 3-[F]fluoro- 2-(4-((2-nitro-1H-imidazol-1-yl)methyl)-1H-1,2,3,-triazol-1-yl)-propan-1-ol; EGFR = epidermal growth factor receptor; RGD = arginine-glycine-aspartate.

Our attitude in clinical practice

In our Centre, we usually perform pre-treatment 18F-FDG PET-CT evaluation in all patients with HNSCC who are candidates for primary irradiation (± chemotherapy), with the exception of early glottic lesions, and in all cases with high risk of distant metastases (hypopharyngeal and rhinopharyngeal primary tumours, cN3 cases). The most relevant clinical information coming from this pre-treatment scan is refinement of neck staging and evaluation of distant metastases or second primaries; both inquiries have a substantial impact on treatment planning. We also prescribe ¹⁸F-FDG PET-CT in patients with cervical metastases and no primary site at conventional assessment.

For treatment response evaluation, we usually perform a PET-CT scan 8-12 weeks after the end of radio-chemotherapy. Patients with a complete metabolic response are referred to periodic follow-up, whereas patients showing a positive scan after treatment undergo subsequent evaluation including, when indicated, biopsy.

During follow-up, especially in the first year when most recurrences occur, we perform ¹⁸F-FDG PET-CT when a recurrence is suspected and a confirmatory biopsy would be associated with significant morbidity, as in case of neck node enlargement, tissue deep under a reconstructive flap, significant persistent post-irradiation toxicity and, in general, difficult exposure requiring general anaesthesia for an adequate biopsy sampling.

Conclusions

¹⁸F-FDG PET-CT is particularly useful for staging, restaging and radiotherapy planning as well as for assessment of treatment response in HNSCC patients, due to its superior accuracy over clinical examination and conventional anatomic imaging. The main limitations, especially in the post-treatment setting, are possible false positive results due to inflammation and the inability to detect microscopic disease. In the future, new tracers other than ¹⁸F-FDG, as well as PET-MRI imaging, will provide clear advantages in several clinical scenarios.