Effectiveness of FDG-PET/CT for evaluating early response to induction chemotherapy in head and neck squamous cell carcinoma: A systematic review.

BACKGROUND: F-Fluoro-Deoxy-Glucose Positron Emission Tomography with Computed Tomography (F-FDG PET/CT) may be a powerful tool to predict treatment outcome. We aimed to review the effectiveness of F-FDG PET/CT in the assessment of early response to induction chemotherapy (IC) in patients with advanced Head and Neck Squamous Cell Cancer (HNSCC) without previous treatment.
METHODS: PubMed, Cochrane Library, Science Direct and Web of Science were searched to May 2016. Reference lists of the included articles and additional studies identified by one nuclear medicine expert were screened for potential relevant studies that investigated the effectiveness of F-FDG PET/CT performed before and after IC. Three authors independently screened all retrieved articles, selected studies that met inclusion criteria and extracted data. The methodology of the selected studies was evaluated by using the risk of bias checklist of the Agency for Healthcare Research and Quality (AHRQ).
RESULTS: Seven out of 170 eligible studies met our inclusion criteria. A total of 207 advanced HNSCC patients were evaluated with F-FDG PET/CT at baseline and after IC in the selected articles. Six from seven studies concluded that F-FDG PET/CT allowed early evaluation response to IC and predicted survival outcomes.
CONCLUSION: The present systematic review confirms the potential value of F-FDG PET/CT as a diagnostic tool for early IV response assessment in HNSCC patients. However, the lack of standard definitions for response criteria and heterogeneous IC protocols indicate the need to further studies in order to better define the role of F-FDG PET/CT in these patients.

Introduction

Head and neck cancers are a group of neoplasias that arise from the oral cavity, pharynx (nasopharynx, oropharynx, and hypopharynx), and larynx. The most common histologic type is squamous cell carcinoma (HNSCC). HNSCC is the sixth most common malignancy worldwide and account for approximately 4% of all diagnosed malignancies. The incidence is rapidly increasing due to tobacco and alcohol consumption, which are the most important risk factors. The human papillomavirus (HPV) in commonly related to oropharynx tumors, which shows a distinct response pattern to chemotherapy and radiotherapy.[ ]

The assessment of the tumor extension is of paramount importance to provide cost-effective treatments. The standard conventional imaging modalities for evaluating patients with HNSCC are computed tomography (CT) and magnetic resonance imaging (MRI). However, these methods are based on morphologic criteria and do not allow information of disease activity such as 18F-fluoro-deoxy-glucose positron emission tomography with CT (18F-FDG PET/CT). In this scenario, FDG PET/CT has been used for initial staging of head and neck cancer, restaging, detection of metastases, detection of unknown primary tumors presented with cervical metastases, radiotherapy planning, and on the assessment of response to chemotherapy and radiotherapy. [ ]

Treatment strategies for locally and advanced HNSCC have greatly changed in the past 20 years with an increase in the use of organ-preserving protocols, which combine radiotherapy with chemotherapy (chemoradiotherapy, CRT) and/or biological molecules in advanced stages. When compared with radiotherapy alone, the multimodality approach has shown higher tumor response rates and has significantly improved local control and outcome.

HNSCC is a highly responsive malignancy at initial presentation and induction chemotherapy (IC) before definitive CRT may reduce distant failure rates. Several reports have revealed that HNSCC patients achieving a clinical complete response (CR) or partial response (PR) after IC had better survival rates than those with residual disease. IC has been increasingly used, especially in cases in which a delay occurs between the definitive diagnosis and the beginning of CRT or surgery. Furthermore, IC before CRT could lead to a better local control of advanced HNSCC stages (III and IV) but with an increased risk of acute toxicity. Early assessment of therapeutic efficacy is a key issue when considering the benefit of escalation in a nonresponder population or to avoid unnecessary toxicity and costs of ineffective treatment.

Conventional diagnostic imaging modalities, such as CT and MRI, have been used for response evaluation of IC, and the RECIST Criteria (Response Evaluation Criteria in Solid Tumors) by anatomical parameters are widely accepted. Some authors highlighted the limitations of anatomic imaging using RECIST and noted the benefits of metabolic tumor response assessment with 18F-FDG PET/CT. [ ] The early assessment of tumor response by 18F-FDG PET/CT performed during therapy has been established for some tumors and has been recently proposed for others solid tumors. Several studies have shown that predictions of survival prognosis might be possible using the FDG PET/CT-based IC response evaluation in carcinoma of the esophagus, stomach, lung, and ovarian. The aim is to avoid overtreatment or ineffective treatment, to reduce acute and late side effects, and finally to improve the outcome. Therefore, the evaluation of IC response by 18F-FDG PET/CT has an important clinical use in predicting survival prognosis in HNSCC patients.

Despite of the advances in multimodality therapy and technical delivery of radiotherapy for locally advanced HNSCC, outcomes remain suboptimal with survival rates of 50% to 60% in 5 years. Despite aggressive combined modality treatment regimens (surgery, radiotherapy, and chemotherapy), HNSCC still shows high rates of recurrence, particularly because the majority of patients harbor advanced disease at presentation, commonly involving regional lymph node metastasis. Therefore, the purpose of this systematic review is to determine the locally advanced HNSCC patient's response to IC by 18F-FDG PET/CT.

Methods

Protocol

This systematic review follows the Preferred Reporting Items for Systematic Reviews (PRISMA) checklist. Approval of the ethics committee was not required, as the study is a systematic review, not involving human participants.

Study design

A systematic review that evaluated the value of 18F-FDG PET/CT imaging at baseline and after IC in patients with locally advanced HNSCC was performed to answer the following question: “Is 18F-FDG PET/CT effective the assessment of early response of IC in patients with HNSCC?”

Eligibility criteria

The aim of this study was to analyze studies that report the use of 18F-FDG PET/CT to early IC response assessment concerning patients with HNSCC. No language or time restrictions were set. The study design included diagnostic tests, cohort, and cross-sectional studies.

Studies were excluded for the following reasons: studies in which 18F-FDG PET/CT was used to evaluate other subtypes of head and neck cancer (not HNSCC); studies in which patients underwent previous CRT, radiotherapy, or surgery; reviews, letters, personal opinions, book chapters, conference abstracts, patents, case reports; studies in which FDG-PET was performed without integrated CT (not PET/CT) or studies in which FDG-PET/CT was not performed at baseline; studies in which patients had distant metastases; and studies with others radiopharmaceuticals (not 18F-FDG).

Information sources and search strategy

The Cochrane Library, Science Direct, Web of Science, and PubMed electronic databases were comprehensively searched with the following keywords

(1) “PET-CT” OR “PET/CT” AND (2) “head and neck cancer” AND (3) “neoadjuvant chemotherapy” OR “induction chemotherapy” AND (4) “survival” OR “outcome response” OR “prognostic value.” All databases were searched up to May 4, 2016. In addition, reference lists of selected articles were hand screened for potential relevant studies that could have been missed during the electronic database search. Field experts were also consulted during research process. Duplicate references were removed from reference manager software (EndNote® X7 Thomson Reuters, Philadelphia, PA).

Study selection

Eligibility of the selected articles was determined in 2 phases. In phase 1, both authors (RF and DLV) independently screened titles and abstracts identified in all electronic databases. The authors selected articles that appeared to meet the inclusion criteria on the basis of their abstracts. In phase 2, the same authors (RF and DLV) read the full text of all selected articles and excluded studies that did not meet the inclusion criteria. Disagreements between authors were solved by consensus and, when a consensus was not reached, a third author (NSM) was involved to make a final decision.

Data extraction

One of the authors (RF) collected all key information included in every article such as authors, year of publication, country, samples, median ages, study design, IC drugs and number of cycles, time of 18F-FDG PET/CT at the baseline and after IC, FDG dose, response criteria, PET/CT parameters and interpreters, time between FDG administration and scanning, reference standard, methods, results, and main conclusions. The second author (DLV) crosschecked all the collected data. Once again, disagreements between them were solved by consensus and the third author (NSM) became involved, when required, to make a final decision.

Risk of bias in individual studies

There is currently no validated tool clear indicated for risk of bias assessment among cross-sectional studies. The Agency for Healthcare Research and Quality (AHRQ) elaborated and validated a checklist for general observational studies. [ ] The AHRQ quality assessment was applied to all the selected articles. Two reviewers (RF and DLV) scored each item with “yes,” “no,” “unclear,” and “not applicable” and assessed independently each included study. Disagreements between both reviewers were solved by consensus and the opinion of a third reviewer (NSM).

Summary measure

Maximum standard uptake value (SUVmax), metabolic tumoral volume (MTV), total lesion glycolysis (TLG), and gross tumoral volume (GTV) were the parameters analyzed to assess the response to IC by 18F-FDG PET/CT imaging in patients with locally advanced HNSCC. For this purpose, all metabolic and volumetric parameters were compared at baseline and after IC.

Synthesis of results

A meta-analysis was planned if the data from the included studies were considered relatively homogeneous.

Risk of bias across studies

The risk has only to be applied if meta-analysis was possible.

Results

Literature search strategy

The literature search yielded 170 citations from The Cochrane Library, Science Direct, Web of Science, and PubMed. After discarding duplicates, 105 articles remained and were screened on title for eligibility. Afterward, 84 articles were discarded and the remaining 21 articles were considered for abstract screening. Reviewing titles and abstracts, 14 were excluded following the exclusion criteria. Thus, the 7 remaining studies added a total of 207 patients with advanced untreated HNSCC. While screening the references of these 7 articles, no new articles were found. A flow chart of the process of identification, screening, eligibility, and inclusion of studies is shown in Fig. 1.

Figure 1

Flow diagram illustrating the literature search protocol and selection criteria adapted from PRISMA.

Study characteristics

All the selected articles were published in medical journals.[ ] Samples size ranged from 15 to 62 participants. Two studies were conducted in France,[ ] 2 in Japan,[ ] 1 in the United Kingdom, 1 in Germany, and 1 in the USA. Three studies were prospective[ ] and four retrospective. [ ] The majority of patients were staged III or IV HNSCC and were male. The tumor location was distributed as follows: 74 from oral cavity/oropharynx, 69 hypopharynx, 3 nasopharynx, and 56 from larynx. One study included 4 tumors from sites that were not specified and other study included 1 case of cervical esophagus carcinoma. The characteristics of the included studies are summarized in Table 1.

Table 1

Main characteristics of the selected studies.

All the included studies were evaluated by 2 18F-FDG PET/CT scans: at baseline and after IC. The characteristics of drug types, IC cycle duration, number of IC, reference standard, and endpoint were not randomized between the studies. Five from seven selected studies used SUVmax as the main parameter of FDG PET/CT to evaluate response of IC.[ ] One study used MTV and TLG , and other study used GTV. Two studies selected the same SUVmax thresholds.[ ] The other selected studies have used different SUVmax thresholds.[ ] The reference standards also varied among the selected studies. Four studies used patients’ follow-up[ ] and two used endoscopy as the reference standard.[ ] Only 1 study selected histopathology response as the reference standard. Regarding the interpretation of FDG PET/CT, in 4 studies, the observers were not described. Only 1 study evaluated HPV status. A summary of the FDG PET/CT parameters, response criteria, endpoint, and main conclusion can be found in Table 2.

Table 2

FDG PET/CT dose, parameters, interpreters, response criteria, reference standards, endpoints, and main conclusions of included studies.

Methodological quality assessment

All studies, except one, were evaluated as having moderate risk of bias based on the sum of the 9 applicable items of the AHRQ. The main methodological limitations representing potential risk of bias were related to the incomplete explanation of the confounding variables, the absence of follow-up in some studies, [ ] and the source of information was not described.[ ] Details about each one of the AHRQ items and the evaluation criteria are described in Table 3.

Table 3

Quality assessment criteria for the included studies (AHRQ).

The heterogeneity of the selected studies concerning different SUVmax cut-offs, response criteria, and endpoints precluded a meta-analysis. Despite of the differences among the selected studies, almost all of them highlighted the effectiveness of FDG PET/CT for early prediction of IC response in advanced HNSCC patients.[ ]

In 1 study , metabolic tumor response was assessed by the measurement criteria of the European Organization for Research and Treatment of Cancer (EORTC) after the second cycle of docetaxel, cisplatin, and 5-fluouracil. The referred study divided the patients in 2 groups (Group 1 = responders; Group 0 = nonresponders) and showed statistically significant differences between the 2 groups regarding the reference standard (P = 0.0014). From the group 1, composed by 10 responders on FDG PET/CT, none had relapsed after a median follow-up of 18.9 months.

In other study, 13 of 21 patients were considered as responders with ≥70% tumor reduction on control endoscopy after a single cycle of TPF. From these 13 endoscopic responders, 9 patients were considered responders by 18F-FDG PET/CT criteria (SUVmax decrease >30%). Among 8 nonresponders on endoscopy, 3 were responders by 18F-FDG PET/CT criteria. Nine patients were considered nonresponders on 18F-FDG PET/CT.

Some authors compared histopathologic response of 26 resected specimens (16 primary lesions and 11 lymph node) with 18F-FDG PET/CT and MRI after 1 cycle of IC and divided into 2 groups: histopathology responders (n = 7; 3 grade 1a and 4 grade 1b) and nonresponders (n = 19; 3 grade 2 and 16 grade 3). The grading of histopathologic regression in tumor beds was performed according to a previous study : grade 1a, complete tumor regression; grade 1b, less than 10% vital tumor seen in tumor bed; grade 2, 10% to 50% residual tumor seen in tumor bed; grade 3, more than 50% remaining residual tumor seen in tumor bed. Specimens with grade 1a or grade 1b responses were regarded as histopathologic responders, while those with grade 2 or 3 were considered histopathologic nonresponders. This classification was made before data analysis. On the MRI parameters, no significant differences were found between the 2 groups. Between PET/CT parameters (pre-chemotherapy SUVmax, post-chemotherapy SUVmax, and % decrease SUVmax), only post-chemotherapy SUVmax and % decreased in SUVmax showed significant differences (3.6 ± 2.3 vs 11 ± 5.6, P < 0.001 and 61 ± 15% vs 18 ± 22, P < 0.001, respectively). The authors concluded that post-chemotherapy SUVmax and % decreased in SUVmax in FDG PET/CT could predict histopathology response to IC more accurately than MRI. Using a SUVmax decrease ≥55.5% as cut-off, they achieved a sensibility of 86%, specificity of 95%, a positive predictive value of 86%, and a negative predictive value of 95%.

Another study has also used the SUVmax decrease ≥55.5% as a threshold for responders on 18F-FDG PET/CT and RECIST criteria on CT and MRI. 18F-FDG PET/CT revealed 27 responders and 30 nonresponders. For 46 lymph node evaluations, 28 patients were considered responders and 18 nonresponders. These patients were also evaluated by RECIST criteria. Thirty patients were considered responders, while 27 patients were nonresponders. Of 57 patients, PET/CT and RECIST criteria matched in 42. Nonetheless, only the nonresponders revealed by 18F-FDG PET/CT were significantly linked to a poor local control rate and the disease-specific survival rate at 2 years after the completion IC (P = 0.03 and P = 0.02, respectively). IC response evaluation of the primary lesion by 18F-FDG PET/CT was the only independent prognostic factor in the disease-specific survival rate (nonresponder vs responder hazard ratio 4.9, 95% confidence interval: 1.4–17.1, P = 0.01).

Some authors compared endoscopic response and early metabolic response by FDG PET/CT with different parameters. The response parameters used on 18F-FDG PET/CT were a decrease >20% in SUVmax (residualSUVmax < 0.8), a decrease >50% in SUVmax (residualSUVmax < 0.5), or a postSUVmax ≤10 after IC. They concluded that these were the best predictors of long-term prognosis. These authors showed that post-IC SUVmax mean values were lower in endoscopic responders (6.0 ± 4.0: mean ± SD) than in nonresponders (14.5 ± 6.2). The residual SUVmax percentage was also lower in endoscopic responders than in nonresponders (34% ± 19% vs 81% ± 37%, respectively). Regarding the correlation with endoscopy, 98% of responders also had a metabolic response, while only 50% of nonresponders achieved metabolic response.

Only 2 studies did not use SUVmax as the 18F-FDG PET/CT parameter for evaluating early response to IC. The former study compared changes in functional imaging parameters delineating the ROIs (regions of interest) containing the primary tumor or lymph node by using specific methods, previously related for PET/CT and MRI through GTV. GTV is commonly used in radiotherapy planning to determine radiation target. After IC, all patients had a greater reduction in 18F-FDG-avid volume and 7 of the 9 included patients had a complete metabolic response on 18F-FDG PET/CT (impossible to create a GTV using the threshold technique). The preliminary data of this study suggested that 18F-FDG PET/CT has limited utility in guiding image-guided radiotherapy following IC. In this respect, functional MRI had an advantage over PET/CT by virtue of its higher resolution and specificity.

In the second study that did not use SUVmax to assess the early response to IC by FDG PET/CT, the following parameters were analyzed: MTV and TLG. The authors showed a reduction of 42% of MTVG and 55% in TLGG that could predict event-free survival (EFS). The values of sensitivities and specificities were 67% and 90%, respectively, for gradient MTV (MTVG) and 62.5% and 90% for gradient TLG (TLGG). The areas under the curve (AUC) for MTVG, MTV with 38% threshold (MTV38), and the MTV with 50% threshold (MTV50) were 0.76, 0.77, and 0.76, respectively (P = 0.03). The AUC was 0.82 to TLGG (P = 0.009) and 0.84 to both TLG38 and TLG50 (P = 0.006). Although these authors did not use SUVmax to evaluate outcomes, this parameter was measured and the median SUVmax decreases were 100% for patients who had no events and 51.88% for patients who had progressive disease or died. Still, these values were not statistically significant (P > 0.05).

Discussion

18F-FDG PET/CT has dramatically changed the management of HNSCC patients in whom treatment is often expensive and associated with a significant morbidity. A recent trial showed that PET/CT-guided surveillance may reduce the number of surgeries and it was more cost-effective than planned neck dissection, although the survival rates were similar in both approaches. The high diagnostic performance of F-FDG PET/CT in detecting recurrence in curatively treated patients with head and neck cancer has also been well established. Nevertheless, the role of 18F-FDG PET/CT concerning the early assessment of IC response deserves further investigation. Several authors have stated that the early assessment of IC response by means of 18F-FDG PET/CT would be a powerful tool to predict treatment outcome.[ ] However, some previous studies have found no survival benefit for HNSCC patients undergoing IC.[ ] To our knowledge, this is the first systematic review that assessed the effectiveness of 18F-FDG PET/CT for evaluating early response of IC in HNSCC patients. The literature regarding this topic is scarce, as only 7 studies met the inclusion criteria.[ ]

In our systematic review, the response to IC has been evaluated on 18F-FDG PET/CT in 207 patients. From the 7 selected studies, 6 demonstrated that early therapeutic response can be evaluated by 18F-FDG PET/CT after IC in patients with advanced staging HNSCC.[ ] Unfortunately, the selected studies were heterogeneous concerning response criteria, reference standards, chemotherapy strategy, and endpoints. For this reason, a meta-analysis was not possible.

In the selected studies, the most used PET/CT parameter was SUVmax.[ ] In addition, the aforementioned studies have used different SUV cut-offs. Therefore, the risk of bias related to the lack of standardization should be mentioned, providing different response measurements after therapy. Only 1 study has verified limitations in 18F-FDG PET/CT findings after IC and problems related to the specificity and spatial resolution of this imaging modality for early response of IC. However, the authors have used GTV as the parameter for evaluating the response, which may have influenced the results. Some previous studies have demonstrated that GTV has not been incorporated into clinical algorithms due to difficulties in reproducibility and absence of prospective data validating the previous reports. The aforementioned authors concluded that MTV is a parameter independently correlated with locoregional control and overall survival in oropharyngeal patients undergoing CRT.

The measurement of “response” is critical in cancer therapy. It is an important determinant of outcome, predicts survival, and guides treatment decisions. For some imaging methods, such as CT and MRI, the RECIST can be considered as a currently available and reproducible method for measuring target lesions. However, these response criteria are based only on morphologic characteristics, such as tumor size, nodal size, and contrast-enhancement patterns.

Anatomical imaging biomarkers do not allow the evaluation of the disease activity and are of limited value for local restaging after IC and CRT due to difficulty in differentiating fibrotic or reactive tissue from viable tumor. Furthermore, these anatomical imaging methods, regardless of methodology, consider a reduction in tumor size as an early sign of clinical benefit. Tumors may have irregular margins and show heterogeneous contrast enhancement, resulting in potential reader variability. It may take 2 or 3 months to detect any shrinkage, whereas functional imaging methods such as 18F-FDG PET/CT can show changes as early as 8 days following the beginning of treatment, and these changes may have been predictive of a longer progression-free survival.[ ] Early metabolism changes in 18F-FDG in squamous-cell carcinoma during chemotherapy have been demonstrated in a previous experimental study. This study showed a rapid decrease in 18F-FDG uptake with subsequent tumor regression after an early transient increase in uptake. Glucose decreases after cytotoxic therapy may be an early sign of a response therapy following tissue necrotic changes before any decrease in tumor size. The authors concluded that all viable tumor cells areas such as hypermetabolic and that an early decrease in 18F-FDG uptake after cisplatin therapy correspond to regressive changes and a reduction in the number of viable tumor cells.

Changes in tumor size do not necessarily reflect the killing of malignant cells. Traditional chemotherapeutic agents are cytotoxic in nature and act primarily by eliminating neoplastic cells. Some novel molecularly targeted therapies are predominantly cytostatic and act primarily by halting tumor growth rather than causing significant tumor cell death. These agents may stabilize or potentially increase tumor size despite excellent clinical response. Therefore, for the targeted therapies, the size criteria based on anatomical imaging modalities are of limited value for evaluating treatment response. On the contrary, 18F-FDG PET/CT may be a useful endpoint for assessing response to targeted therapies. The biologic basis of changes in 18F-FDG uptake may be more complex than those for traditional cytoreductive therapies. In the last version of RECIST, a section on detection of new lesions, including the interpretation of 18F-FDG PET, was added.

Due to the limitations of anatomic tumor response assessed on RECIST, the PERCIST (PET Tumor Response) has been proposed, aiming to serve as a starting point for use in clinical trial. However, PERCIST has not yet been well established, and so further revisions and enhancements would be necessary for the adequate validation method. The European Organization for Research and Treatment of Cancer had proposed some useful and foresight response criteria on PET, available at that time, by using SUVmax as the main parameter. In our systematic review, only 1 study considered the EORTC criteria.

Our systematic review has also shown differences regarding chemotherapy strategy, the number of cycles of IC, and the time of performing 18F-FDG PET/CT after this neoadjuvant therapy. All these differences may have influenced the results of our work, reinforcing the need of further standardized parameters. As images interpretation in the post-therapeutic setting with 18F-FDG PET/CT is challenging (due to anatomical distortions, inflammation, edema, fibrosis, asymmetry), an interval of 12 weeks has been recommended for performing this imaging modality after CRT and/or radiotherapy. Nonetheless, the optimal timing of the first response assessment after IC or CRT is not yet known.

Regarding the reference standards, most of the selected studies have used patients’ follow-up.[ ] Follow-up requires a long time to serve as the reference standard and it may impair the early assessment of therapeutic efficacy using a metabolic response. Therefore, for this purpose, the histopathology response is the ideal reference standard. However, from the 7 selected studies, only 1 used the histopathology response as the reference standard. If a more precise prediction of the histopathology response to IC could be achieved by a noninvasive modality (18F-FDG PET/CT), the response to subsequent therapy may also be more precise, thus contributing to organ preservation. For this reason, further studies are required to evaluate the early IC response assessment by 18F-FDG PET/CT using the histopathology response as the reference standard in order to reduce this source of bias. Some authors have developed a 5-point interpretation criteria for therapy response assessment (Hopkins Criteria) on PET/CT scans. This subjective method has demonstrated a substantial interobserver agreement, a negative predictive value of 92%, and predicted overall survival and progression-free survival in HNSCC patients. The method should be further investigated for evaluating early therapy response of IC.

The inclusion criteria can be considered as a limitation of our study, as we decided to only select studies that have used 18F-FDG as the radiotracer on PET/CT. Other less studied radiopharmaceutical tracers have been tested for HNSCC patients.[ ] Nowadays, the cost-effectiveness and availability of 18F-FDG justify the current use, despite the lack of large and randomized studies. The main limitation refers to the heterogeneity of the selected studies concerning response criteria, reference standards, chemotherapy strategy, and endpoints. Nonetheless, all of the studies, [ ] except one, have demonstrated the effectiveness of 18F-FDG PET/CT in advanced HNSCC patients.

In conclusion, our systematic review shows that 18F-FDG PET/CT may assess early response of IC in HNSCC patients. However, as the 7 selected studies were heterogeneous regarding response criteria, reference standards, chemotherapy strategy, and the endpoints, further standardized studies would clarify the role of PET/CT in the context of predicting IC response in HNSCC patients. Moreover, the assessment of early response of IC by PET/CT may influence treatment outcome. The success of surgical salvage depends on early and accurate detection of IC response.