Neoadjuvant chemotherapy for advanced oral cavity cancer

A Caucasian male aged 51 years presented with tongue pain, odynophagia, right otalgia, and weight loss. Physical examination showed a tender, ulcerative mass of the posterior lateral tongue and crossing the midline. A biopsy confirmed invasive oral tongue squamous cell carcinoma (SCC). Computed tomography (CT) imaging revealed a deeply invasive tumor involving the right oral tongue, extending across the midline and almost to the contralateral tongue border (Figure 1). The tumor measured 3.5 × 4.3 × 4.3 cm with an infiltrative pattern and poorly defined borders. Bilateral pathologic upper cervical lymph nodes were observed. The tumor was classified as clinical stage IVA (T4aN2cM0). Baseline imaging. Representative (A) axial and (B) coronal computed tomography images of a deeply invasive, right oral tongue squamous cell carcinoma (classified as T4a) crossing the midline. Primary surgical resection would require subtotal glossectomy. The patient's past medical history was negative for exposure to risk factors for head and neck cancer, including prior smoking, alcohol, or illicit drug use. The patient works as a SWAT team commander, an occupation in which loss of speech would result in permanent disability. After a multidisciplinary discussion, the decision was made to treat the patient with neoadjuvant chemotherapy followed by surgery, including free-tissue transfer reconstruction and adjuvant radiation therapy (RT). The patient was given three cycles of combined docetaxel, cisplatin, and 5-fluorouracil (TPF). The patient tolerated neoadjuvant chemotherapy very well and noted improvement in pain and eating with treatment. He gained >15 pounds and was able to resume a normal diet from a previously self-restricted diet secondary to pain. Restaging CT imaging showed that the primary tumor had decreased in size to 2.9 × 2.1 × 3.6 cm, corresponding to a 66% decrease in volume (Figure 2). Importantly, the primary tumor had regressed closer to the midline. Enlarged lymph nodes remained present in the neck bilaterally but had similarly decreased in size. The patient was evaluated preoperatively by a speech pathologist, and a modified barium swallow study revealed all phases of swallow within normal limits. The examination was without aspiration/laryngeal penetration or pharyngeal retention, and the motor speech system was functional. Postneoadjuvant imaging. Representative (A, B) axial and (C, D) coronal computed tomography (CT) images (A, C) before and (B, D) after neoadjuvant treatment with three cycles of docetaxel, cisplatin, and 5-fluorouracil. The patient then underwent response-adapted oncologic surgery, including right partial (3 years after surgery. National Comprehensive Cancer Network (NCCN) guidelines for the management of oral cavity SCC indicate that surgery is the preferred initial treatment, including for advanced (T3, T4a) primary oral tongue cancers.1 This is based on data showing improved survival with surgery compared with concurrent chemoradiotherapy as definitive treatment in oral tongue SCC.2, 3 However, primary surgery can be functionally devastating in some patients, even with optimal reconstruction, particularly for cancers that require >50% resection of the tongue. As such, for selected patients who decline or are not candidates for resection, nonsurgical options are suggested alternative treatment options, including definitive RT or concurrent chemoradiation. The recommendation for primary surgery is based on historical precedence and NCCN guidelines. When surgery is performed for oral tongue SCC, there are technical factors that can impact prognosis. Margin status is one of the most important variables associated with survival for oral tongue SCC.4, 5 Traditionally, a 5-mm margin on final pathology is recommended for oral tongue SCC. To achieve this, a 10-mm margin is typically required in surgery to allow for expected shrinkage of the specimen in formalin. More recently, Liao et al. found that 7 mm was the ideal intraoperative resection margin for ensuring an adequate oncologic resection without sacrificing normal tongue tissue, because a planned intraoperative margin smaller than 7 mm was found to decrease local relapse-free survival on multivariate analysis.6 In a retrospective study that included 381 patients with oral cavity SCC who were reviewed by dedicated head and neck pathologists, it was found that local recurrence-free survival was significantly affected only with pathologic surgical margins ≤2.2 mm, considerably less than the traditional 5 mm on final pathology.7 Regardless, it is incumbent upon the surgeon to remove the cancer completely while preserving as much normal tissue as possible, and close collaboration with pathology during surgery is critical for success. Appropriate management of lymph nodes is also critical in the treatment of oral tongue SCC. This was demonstrated in a retrospective analysis of 4341 patients who had pathologic node-negative (pN0) oral cavity SCC in which the number of nodes dissected was a significant predictor of overall survival (OS; ≥22 nodes: hazard ratio HR, 0.854; p = .031). The authors also calculated that, with each additional lymph node removed, survival was increased (HR, 0.995; p = .022).8 A similar analysis of the National Cancer Database demonstrated that patients who presented with clinically node-negative (cN0) oral cavity SCC who had 0.06 was correlated with worse OS.10 However, lymph node density has yet to be incorporated into NCCN guidelines. From a surgical standpoint, the role of neoadjuvant systemic therapy is less well defined. To date, there are no standardized surgical guidelines for patients who are treated with neoadjuvant systemic therapy. In a retrospective analysis comparing perioperative morbidity between patients with oral cavity SCC receiving chemotherapy before surgery (n = 147) or not (n = 687), there was no difference with regard to wound complications (p = .47), return to the operating room (p = .31), or readmission rates (p = .49). The neoadjuvant chemotherapy group received more blood transfusions (p 3 cm, multiple positive lymph nodes, lymphovascular invasion, and perineural invasion (refer to NCCN guidelines). For particularly high-risk patients who have positive margins or extranodal extension, the addition of postoperative chemotherapy to RT is associated with improved local-regional control and survival.35-37 Because the use of neoadjuvant therapy before surgery is less common than upfront surgery, there are less data for guidance on who should receive adjuvant RT, particularly after a complete pathologic response. Consequently, it can be challenging for a clinician to decide whether a patient with locally advanced cancer should receive PORT if they had a major pathologic response and no longer have the standard pathologic indications listed above. In this case, if the patient were to proceed straight to surgery without neoadjuvant therapy, then they would be counseled on surgery followed by PORT based on the anticipated pT4 classification. The RT would be delivered to the initial extent of disease and the at-risk regional lymph nodes. However, if the patient were to receive neoadjuvant systemic therapy, can RT be safely omitted if there is a complete pathologic response such that they no longer have viable tumor in the specimen and lack traditional indications for PORT? There is guidance regarding when induction chemotherapy is given before definitive chemoradiation, as done in seminal phase 3 studies of patients who had tumors deemed unresectable or were candidates for organ preservation.38, 39 In this context, RT is delivered to the initial pretreatment disease extent, even if there is a complete clinical response to induction chemotherapy.40 The rationale for this practice is that tumors do not shrink concentrically, and microscopic disease may exist even when gross disease is no longer visible on examination or imaging. There are published clinical trials in nonoral cavity cancers, such as for nasopharyngeal cancer41 or oropharynx cancer,42 that reduce the dose and extent of RT based on the response to induction chemotherapy, but this is not yet standard practice. The three neoadjuvant studies discussed above from India,1 China,43 and Italy44 provide examples of how clinicians may approach the application of PORT when neoadjuvant chemotherapy precedes surgery. Two of the trials took a conservative approach, requiring that all patients in the neoadjuvant arms receive RT to the initial extent of disease.19, 43 Chaukar et al.,19 in a trial aimed at reducing the extent of surgery, maintained the PORT to avoid increasing the risk of local-regional failure if they were to reduce both surgery and RT at the same time. For Zhong et al.,43 the primary end point was OS. By ensuring all patients received RT to initial sites of disease, they avoided the potentially detrimental effect of local-regional failures on the primary end point. Only the study by Licitra et al.44 omitted RT for patients who did not have prespecified pathologic risk factors, and there was no obvious detriment to local-regional control. A large phase 3 trial, KEYNOTE-689 (ClinicalTrials.gov identifier NCT03765918), is accruing at the time of this writing. It randomly assigns patients who have stage III–IV head and neck SCC to receive either standard surgery with postoperative therapy based on pathology or neoadjuvant and adjuvant pembrolizumab. In that study, all patients who receive neoadjuvant therapy receive postoperative therapy to the initial disease extent, regardless of the treatment response. In summary, after neoadjuvant therapy, PORT is given for standard pathologic indications after surgery and for advanced tumors on initial clinical staging. The challenge is when to give PORT if there is substantial downstaging such that the final pathologic stage is discordant from the initial pretreatment clinical stage. Reducing postoperative treatment is appealing because it is associated with substantial acute toxicity and late effects. Reducing postoperative treatment must be done in a systematic fashion because of the established role of PORT in improving oncologic outcomes of patients with head and neck cancers. For this patient with a very locally advanced tumor, we would anticipate recommending PORT regardless of the pathologic response to neoadjuvant chemotherapy based on the initial extent of disease and the historical precedent described previously.45-47 Ablative surgery of the oral tongue results in alterations to speech intelligibility, articulatory precision, and swallow function that ultimately affect quality of life and performance status. The tongue interacts precisely with nearby structures, such as the teeth, alveoli, and the palate, to perform the motions of articulation.48 Mobility of the oral tongue is critical for speech, mastication, oral health, and swallow function. Decreased lingual range of motion (ROM) heavily affects the oral phase of swallowing, whereby patients have trouble with chewing, bolus formation, and transit of the bolus from the anterior tongue to the posterior tongue to trigger the swallow. This often results in leftover residue on the lingual and/or palatal surfaces. Furthermore, patients may report impaired articulation or phoneme distortions (speech precision).49, 50 Findings of a systematic review after ablation and free flap reconstruction of the tongue revealed that the extent of speech impairment relies on multiple factors, such as tumor size, preservation of tongue tip, method of reconstruction, flap sensation, and adjuvant therapy. Factors that affect postoperative swallowing outcomes include adjuvant therapy and method of reconstruction.51 Additional determinants of speech and swallow function after glossectomy include tumor size and depth (e.g., cross-midline), tumor location (e.g., anterior vs. posterior), extent of resection of adjacent subunits (e.g., floor of mouth, tonsillar pillar), and type of reconstruction (e.g., primary closure, skin graft, free flap, or healing by secondary intention). Poor functional outcomes are associated with reduced postoperative lingual ROM. Recent data identified four clinically distinct glossectomy defects using validated measures of speech intelligibility, quality of life, and functional status. A comparison of functional outcomes by glossectomy defect in 101 patients (37 underwent subtotal hemiglossectomy, 33 underwent hemiglossectomy, 16 underwent extended hemiglossectomy, and 15 underwent oral glossectomy) revealed differences in intelligibility.52 For example, sentence intelligibility differed significantly between the four groups, with a decrease in intelligibility as the defect increased (subtotal hemiglossectomy, 92% intelligibility; hemiglossectomy, 90% intelligibility; extended hemiglossectomy, 83% intelligibility; and oral glossectomy, 70% intelligibility). Clinically, the concept of speech intelligibility differs from articulatory precision. Patients may present as intelligible and yet appreciate the deviations in precise articulation of various speech sounds. In a study comparing both speech (including evaluations of speech rate and overall quality and intelligibility) and swallowing using nonvalidated methods (mean bolus volume ingested, duration of swallow, and mean volume swallowed per second), speech quality was better in patients who underwent primary closure. However, swallowing outcomes were better for those who underwent free flap reconstruction. These results support restoration of volume in the oral cavity, thus improving tongue-to-palate contact, which facilitates oral and pharyngeal clearance. However, overly bulky flaps may limit lingual mobility and ROM, thus negatively affecting speech intelligibility and articulation.53 For example, the inability to approximate the tongue to the hard palate yields diffuse oral cavity residue particularly with solid foods. This is a result of impaired lingual sweeping and/or lack of tongue-to-palate contact that requires a variety of modifications or compensations (e.g., the use of a finger to clear the oral sulci because of limited ROM of the reconstructed tongue, head tilts to facilitate passive movement of a bolus). Functional ROM is critical to achieve optimal speech intelligibility and to advance diet complexity and is related to quality of life. These considerations are relevant to the patient presented because primary closure is typically not feasible in patients who have larger defects. With regard to swallowing function, patients who have larger defects present with significantly worse function on objective swallow evaluations (videofluoroscopy) compared with patients who have smaller defects.21, 54 Huang et al. found that tongue resection >50% was a strong predictor of penetration and aspiration.55 In a cross-sectional study assessing swallowing function in patients with oral tongue SCC who underwent surgery with or without reconstruction, 106 patients were evaluated with videofluoroscopy. The results revealed that the incremental volume of the glossectomy defect was associated with worse outcomes across all videofluoroscopy parameters. On multivariate analysis, this remained as the only independent predictor for oral videofluoroscopy parameters as opposed to method of reconstruction, T stage, and radiotherapy.56 The following testimony was written by our patient, detailing his experience with oral cavity cancer: In early 2020, I began to experience pain in my right ear. Initially, not much thought was given to it as my profession required that I spend much of my time in varying environments, and I believed I simply had an ear infection. With a visit to my health care provider, I learned that the pain was the result of an SCC tumor that originated on my tongue and had developed into a stage IV mass. I met with local oncologists, who informed me that my only option was complete tongue removal. As I continued to search for a plan of care locally, I seemed unable to find a partnership that possessed the same level of commitment for healing that I personally held. Their emphasis was focused solely on cure and not quality of life. That changed when I arrived for my initial visit at MD Anderson with Dr. Neil Gross. As the inquiry of my background began, I quickly emphasized that I was a Law Enforcement Instructor and that speaking, both publicly and privately, was at the center of my daily actions. Dr. Gross began to develop a nontraditional plan through using chemotherapy first to shrink the tumor and surgery to quickly follow. Finally, radiation would be used to cleanse the area. The plan was to save my tongue. I immediately knew I had located someone willing to be equally yoked in my healing. The commitment I had was built from my Christian faith and attributes taught to me by many industry professionals. First, my faith: My faith was and is the cornerstone that allowed me to maneuver through the challenges that cancer treatment posed. My Christian beliefs provided a mental peace and gave endurance when the difficult days would arrive. Second, my family. I am unable to describe the immeasurable devotion my wife had to being my caregiver. She spent countless hours and days keeping count of the plans for care and making sure I was following them. Last, the attributes learned through the years as a SWAT officer: The time spent being taught to endure hardships to accomplish goals and removing quit as an option. I learned to keep my world small; some days, I would work to make it to noon. Once there, I told myself to make it to the evening. Cancer treatment is too much when you consider it in its totality. You have to view it as a series of small goals and work to achieve them. This journey took 50% of my tongue and 77 lymph nodes, but it gave me much more. My tongue was rebuilt by harvesting from my arm. Today, I am 3 years with no evidence of disease and am able to speak. Looking back, strangely, I am thankful for having this experience. It forced me in a new direction. The mental and physical battles make you stronger if you let them. I am a better person today than before cancer. You have to see the benefits in things, even when they bring a short-term unpleasantry. Every profession has its particular phrases. I constantly heard the phrase, "your new normal." Too many take that as a deficit invitation. I learned to view it as possibility. My reply became, "my new normal is discipline and focus." Once home, I began the process of rebuilding myself physically. Learning to complete old tasks in different manners, rebuilding my mobility, and regaining the strength and weight lost to treatment. Along the way, I have learned to not be so rigid in my thinking. I learned quickly that my old routines were not as beneficial as before, and I had to adopt new routines to push myself forward. Today I have returned to my responsibilities and position at work. I instruct various courses of disciplines in our craft. Most importantly, I now teach a class at my church. I found the partnership commitment to my healing at MD Anderson. There are far too many to name, but they are all equally important. To say thank you to them all is so insufficient, but I will not forget their commitment to making cancer a thing of the past. The prognosis for advanced, resectable oral cavity cancer remains grim, with only one half of patients achieving disease-free status 5 years postsurgery. Despite efforts to enhance outcomes through intensified adjuvant treatments and preoperative chemotherapy, there has been little improvement in survival.14, 18, 57 Immunotherapy using immune checkpoint inhibitors has demonstrated efficacy in recurrent or metastatic head and neck mucosal SCC.58, 59 These promising results have prompted investigations into the potential impact of immunotherapy in earlier stages of the disease, including in the neoadjuvant setting. Pathologic responses have been observed after neoadjuvant immunotherapy alone in oral cavity SCC, and a randomized phase 3 trial is underway testing this approach.60, 61 There are also now smaller trials investigating neoadjuvant immunotherapy and chemotherapy in oral cavity SCC. This approach is particularly encouraging given the strength of the data in nonsmall lung cancer and warrants randomized trials. Tanguy Y. Seiwert reports personal/consulting fees from AstraZeneca, EISAI INC., Inate Pharma Inc., iTeos, Merck, Regeneron Pharmaceuticals, Sanofi, and Vir; and service on a Data and Safety Monitoring Board at BioNTech outside the submitted work. Michelle D. Williams reports personal/consulting fees from Bayer Healthcare and support for other professional activities from Springer outside the submitted work. Neil D. Gross reports grants/contracts and personal/consulting fees from Regeneron Pharmaceuticals Inc.; personal/consulting fees from DragonFly Therapeutics Inc., Intuitive Surgical Inc., Merck, Replimune, and Sanofi/Genzyme US Companies; and support for other professional activities from PDS Biotechnology Corporation outside the submitted work. The remaining authors disclosed no conflicts of interest.