In this study, for patients with locally advanced and resectable HNSCC, induction chemotherapy benefited these patients in terms of an 8% lower rate for the occurrence of distant metastases; however, induction chemotherapy did not improve overall survival, disease-free survival or locoregional control. The toxicity of induction chemotherapy was acceptable for further surgery, or radiotherapy or chemoradiotherapy.
Previous meta-analyses of both resectable and unre-sectable HNSCC [4–7] and this meta-analysis of resectable HNSCC confirm the effective decrease in the development of distant metastases in the patients without metastases at baseline (M0) treated with induction chemotherapy and locoregional treatment, compared to locoregional treatment alone. This is reasonable due to the effect of induction chemotherapy on peripheral, potentially metastatic tumor cells.
In this study, with respect to overall survival, we found no significant benefit of induction chemotherapy in the patients with resectable HNSCC, even using the PF protocol, which has been reported to be beneficial to overall survival in other meta-analyses [4, 6, 7]. According to the site of primary lesions, most of patients receiving PF induction chemotherapy in the present study were oral cancer patients; while the previous studies of induction chemotherapy in HNSCC patients included not only oral cancer, but also patients with oropharyngeal and hypopharyngeal cancer. As we know, for oral cancer patients with resectable lesions, radical surgery is considered as the standard of care, followed by post-operative radiotherapy or chemoradiotherapy, depending on the presence of intermediate/high risk features in the surgical specimen; and for patients with resectable locally advanced oropharyngeal and hypopharyngeal cancer, surgery, or radiotherapy or chemoradiotherapy, followed by radiotherapy or chemoradiotherapy is considered the standard of care. The efficiency of adding PF agents to standard care may differ between patients with oral cancer, and those with oropharyngeal or hypopharyngeal cancer. So, the effect of PF induction chemotherapy may differ in the patients with different primary tumor sites, and therefore, the primary tumor site might be considered before adding PF induction chemotherapy. It appears that induction chemotherapy could be more effective in oropharyngeal and hypopharyngeal cancer than in oral cancer. This could be due to various factors, such as high-risk human papilloma virus (HPV) infection, producing virus oncoproteins of E6 and E7, which are necessary for viral replication through their proliferation-stimulating activity, and play a key role in malignant transformation and maintenance; they are also sensitive to cytotoxic chemotherapy and DNA damage-induced apoptosis. Also, as we know, the incidence of HPV in patients with oropharyngeal cancer is higher than in patients with oral cancer. Based on the results of the present study, some factors might be considered to improve the prognosis in future clinical trials, such as insistence on radical surgery, even in patients with clinical response, in order to reduce locoregional failure and to improve survival, or optimization of the induction chemotherapy protocol by adding new or targeted drugs.
Recently, in two randomized phase III trials [3, 28, 29], a new induction chemotherapy protocol of a combination of docetaxel, cisplatin and 5-fluorouracil (TPF) followed by radiotherapy or chemoradiotherapy has been shown to improve survival compared to PF, and it is suggested as the preferred chemotherapy regimen when induction treatment is used for management of HNSCC patients. However, there is still little evidence from large clinical trials that the use of induction TPF prior to locoregional treatment improves survival when compared to locoregional treatment alone. Furthermore, it is unknown whether induction TPF improves outcomes when given prior to surgery in patients with locally advanced and resectable HNSCC. Several clinical trials of epidermal growth factor receptor inhibitors suggest the targeted drugs could improve locoregional control and survival in HNSCC patients with primary, recurrent or distant metastatic lesions; cetuximab, for example, is now firmly established as an active component of treatment for advanced HNSCC, alone and in combination with other modalities, including radiotherapy, platinum-based chemotherapy and induction therapy. Cetuximab has been approved by the Food and Drug Administration (FDA) USA, for HNSCC treatment in combination with radiotherapy for locally advanced, potentially curable disease, and as a single agent for incurable recurrent or metastatic disease [30–33]. Additional clinical trials are warranted to determine the benefit of adding an epidermal growth factor receptor (EGFR)-targeted agent in the setting of locally advanced and resectable HNSCC.
As we know, the different response to induction chemotherapy could lead to different survival, with good response always leading to good survival, bad response leading to poor survival . Some predictive biomarkers reflecting the response to induction chemotherapy could be helpful for the next treatment choice, or in deciding whether induction chemotherapy should be performed, especially for resectable lesions. If the individual biomarkers predict bad response to induction chemotherapy, it should not be performed in those patients; otherwise, induction chemotherapy could benefit both patient response and survival. The biomarkers include DNA sequence mutations, epigenetic changes, and levels of messenger RNA or protein expression. For example, in a prospective study , p53 gene mutations are strongly associated with a poor risk of both objective and major response to PF-based induction chemotherapy, suggesting that patients with HNSCC should first be screened for p53 mutations, before choosing the most appropriate treatment protocol based on the mutations.
For organ preservation, it has been well-recognized that in patients who respond to it, induction chemotherapy, followed by radiotherapy or chemoradiotherapy, instead of radical surgery, could benefit patients with laryngeal cancer by preserving the larynx, without a negative impact on overall survival and disease-free survival. For other organs, although the response rate to induction chemotherapy is relative high (50% to 80%) in the resectable lesions, which provides a better chance to eradicate the locoregional lesions by radical surgery, there is no conclusive evidence that induction chemotherapy confers the benefit of organ preservation. A report by Licitra et al.  revealed that induction chemotherapy can reduce the number of patients requiring mandibulectomy and/or radiation therapy. However, there are no further reports on the differences in survival or locoregional recurrence between patients who do or do not undergo mandibulectomy and/or radiotherapy. In our opinion, in order to remove the tumor completely, emphasis should placed on surgical resection being performed according to the original tumor borders, which are marked with tattoo or ink before treatment initiation, regardless of the response to induction chemotherapy. In this study, there was no evidence of significant differences in relation to benefit for locoregional control between patients receiving or not receiving induction chemotherapy. Further trials are needed to resolve whether induction chemotherapy can lead to organ perseveration of non-laryngeal sites.