Treatment of laryngeal cancer

INTRODUCTION The larynx is the most common site of head and neck cancer (HNC) in the United States . In data from the American Cancer Society, larynx cancer accounted for 31 percent of the 30,990 cases of HNC diagnosed annually . The male-to-female ratio of 4 to 5:1 is, at least in part, reflective of differences in tobacco and alcohol use.

The larynx is divided into three anatomic regions: the supraglottic region larynx, glottic larynx (true vocal cords and mucosa of the anterior and posterior commissures), and the subglottic larynx which extends to the inferior border of the cricoid cartilage

Laryngeal cancers are often discovered at an earlier stage than other head and neck cancers, because hoarseness tends to occur relatively early in the course of the disease. However, the symptoms associated with cancer of the larynx, particularly hoarseness, depend upon location. Persistent hoarseness is frequently the initial complaint in glottic cancers; later symptoms may include dysphagia, referred otalgia, chronic cough, hemoptysis, and stridor. Supraglottic cancers are often discovered later and may present with airway obstruction or palpable metastatic lymph nodes; hoarseness is an uncommon presenting sign. Primary subglottic tumors are rare, and affected patients typically present with stridor or complaints of dyspnea on exertion.

Biologic behavior also varies according to location . Supraglottic cancers are richly supplied with lymphatics, leading to a high frequency of cervical nodal metastases. The lymphatic drainage of the right and left sides of the supraglottic larynx and subglottic larynx and those structures below it are separate; this knowledge allows prediction of metastatic behavior and also affects planning of larynx conservation operations. In comparison, the true vocal cords (glottic larynx) possess little or no lymphatic drainage.

Given the fundamental role that the larynx plays in human speech and communication, determining the optimal management of laryngeal cancer must consider both survival and the functional consequences of the given treatment approach. Total laryngectomy is widely recognized as one of the surgical procedures most feared by patients. Although different options for voice rehabilitations exist, many patients express dissatisfaction with the results, and social isolation, job loss, and depression are common sequelae. (

As a result, much effort has focused on larynx-sparing approaches, such as radiation therapy (RT) alone, combined chemotherapy and RT, and function-preserving partial laryngectomy procedures. Treatment guidelines for laryngeal cancer from the American Society of Clinical Oncology, published in 2006, recommend that all patients with T1 or T2 larynx cancer be treated, at least initially, with intent to preserve the larynx

TREATMENT OF EARLY STAGE LARYNGEAL CANCER By definition, patients with stage I and II disease have no evidence of nodal involvement. Specific management of the neck in patients with head and neck cancer and a clinically negative neck examination, including the role for neck dissection, is discussed separately.

General principles

  RT versus surgery  For patients with early stage (T1-2N0) disease, both RT and surgery have the potential to cure a high proportion of patients. There are no randomized trials in which RT has been compared to surgery (laryngectomy or conservation surgery) for treatment of early stage disease. Similarly, there are no randomized data comparing functional outcomes (ie, voice quality and swallowing ability) after surgery or RT for this stage of disease. Treatment recommendations are largely based upon evidence from prospective uncontrolled studies and retrospective cohort series.

In general, because overall outcomes from surgery and RT appear to be similar for early stage disease, the choice of therapy is often dictated by preservation of voice, which favors RT. However, RT may result in other complications that can adversely affect quality of life (QOL).  Furthermore, local tumor recurrence after RT may be amenable to salvage by organ preserving surgery, but total laryngectomy will be necessary for a substantial proportion of such patients, particularly those with T2 tumors.

Another challenge is that limited stage disease represents a wide spectrum of disease, and multiple factors (eg, tumor location, volume, extent of invasion into the vocal cord, involvement of the anterior commissure, nodal metastases, patient age, occupation, compliance, and preference) need to be considered in choosing therapy. These issues can be illustrated by the following examples :

  • For well-defined favorable T1 glottic cancers (ie, superficial tumors located in the middle third of the cord, particularly on its free edge), endoscopic laser resection can result in a good voice outcome, and is preferred over RT, which takes longer to administer. On the other hand, lesions that are indistinct, particularly those arising in the context of widespread, abnormal-appearing mucosa, are more suitable for RT than surgery.
  • RT is preferred by many clinicians for treatment of T2 glottic cancers that have preserved vocal cord mobility and are characterized as superficial on radiographic imaging. Local control rates are high and functional outcomes are good. However, survival may be compromised after failure of RT in such patients , leading some to recommend supracricoid partial laryngectomy with cricohyoidoepiglottopexy for patients who are willing to sacrifice some voice quality in an effort to improve local control.

  Combined modality therapy Single-modality treatment is effective for limited stage, invasive cancer of the larynx. Treatment guidelines from ASCO emphasize that every effort should be made to avoid combining surgery with RT because functional outcomes may be compromised. If undertaken, larynx-preserving surgical excision of the primary tumor should aim to achieve tumor-free margins. Narrow-margin excision followed by postoperative RT is not an acceptable treatment approach.

Induction chemotherapy has been investigated as a treatment for early stage larynx cancer . However, in contrast to the situation with locoregionally advanced disease, the data are insufficient to recommend this approach outside of the context of a clinical trial.
Concurrent chemoradiotherapy may be used as an option for larynx preservation, even if a partial laryngectomy would be possible.

  Tobacco abuse  Continued cigarette smoking is associated with a worse outcome after RT. Patients should be encouraged to abstain from smoking after diagnosis, throughout treatment, and thereafter.

The following section will address the choice of treatment according to the subsite of disease.

Supraglottis Lesions of the supraglottic larynx tend to spread locally and metastasize frequently to the cervical nodes. The supraglottis is considered a midline structure; as a result, there is a substantial incidence (15 percent or greater) of bilateral cervical nodal metastasis . The net effect is that many patients with a clinically negative neck have occult (histologically) positive cervical nodes, and there is a high frequency of relapse and eventual death in patients who do not undergo elective treatment of the neck.

Guidelines from ASCO recommend that all patients with supraglottic cancer should have elective neck treatment, even if clinically N0 . We recommend selective neck dissection or RT for the clinically negative neck, and a neck dissection with or without RT for palpable lymphadenopathy.

Early supraglottic cancers can be effectively treated with a voice-preserving supraglottic laryngectomy or RT, as long as both sides of the neck are also treated. In a retrospective review of 166 consecutive patients with T1-T2 carcinoma of the supraglottic larynx treated between the years 1983 and 1992 , 66 patients underwent conservative surgery and 100 underwent definitive RT. Overall five-year survival for the entire group was 72 percent. Survival for patients in the surgical group was 88.4 percent, whereas for the RT group it was 76.4 percent. Ninety-five percent of patients in the surgery group and 72 percent in the RT group had laryngeal preservation. Selection bias could have affected these outcomes.

  Surgery  Early stage primary disease is highly curable by laryngeal surgery, preferably using procedures that remove only the upper portion of the larynx. This supraglottic laryngectomy is physiologic and permits retention of vocal and swallowing functions. Contraindications to supraglottic laryngectomy include impairment of vocal cord mobility or extension to the true vocal cord, interarytenoid area, piriform sinus apex, post-cricoid region or thyroid cartilage. Resection for tumors that have extended to one arytenoid, tongue base, or medial wall of piriform sinus constitutes "extended" supraglottic laryngectomy. These are still considered conservation surgeries, but require a longer period of swallowing rehabilitation.

Some aspiration is expected postoperatively; therefore, poor pulmonary function is also a contraindication to supraglottic laryngectomy. There are no clearly accepted ways to measure adequate pulmonary function. Some surgeons rely on clinical assessment of a patient's activities (walking, stair climbing), while others rely upon formal pulmonary function testing. An FEV1/FVC less than 50 percent may signify a greater risk of severe aspiration complications

Local control and survival rates for supraglottic laryngectomy are excellent for early stage disease . In one series of 104 patients in which 29 percent of the patients received supraglottic laryngectomy and the remainder total laryngectomy, the five-year survival was 100 percent for T1 and 81 percent for T2 disease .

  Endoscopic laser resection  Several studies have evaluated the use of endoscopic laser resection of early stage supraglottic cancer . This procedure can result in high cure rates if clear surgical margins can be obtained. In a study of 70 patients with cancer of the supraglottic larynx, five-year recurrence-free survival was 85 percent for stage I, and 63 percent for stage II cancers . However, the procedure is not without risk. This was illustrated by a second study of 46 patients with T1/2 cancers of the supraglottic larynx, treated with intraoral laser ablation . The calculated overall survival was 59 percent after five years; among the eight patients who died with disease, four had uncontrollable local or regional recurrences and five (11 percent) failed to relearn swallowing, requiring total laryngectomy.

The criteria for patient selection remain to be established, although most authors agree that early stage suprahyoid lesions are most amenable to endoscopic treatment. Treatment of the clinically negative neck should be strongly considered.

  Radiotherapy  Excellent local control has been achieved for cancers of the supraglottic larynx after external beam (EBRT), approximating 90 percent for T1 lesions and 80 to 89 percent for T2 lesions . Five-year survival in patients treated with EBRT are comparable to the results obtained with primary surgery. As an example, in one study of 166 patients with T1-2N0 cancers of the supraglottic larynx, 66 underwent conservative surgery and 100 had definitive RT . The following outcomes were noted:

  • The five-year overall survival was 72 percent.
  • The five-year disease-free survival with conservative surgery and RT was 88 and 76 percent, respectively.
  • Laryngeal preservation was successful in the majority of patients (82 percent overall, 95 percent with conservative surgery, and 72 percent with RT).
  • Fifty percent of local recurrences after RT could be salvaged with laryngectomy.

Because of the complications associated with persistent swelling and swallowing difficulty, partial laryngectomy is not recommended for patients who have failed RT. The salvage procedure in such patients is a total laryngectomy.

The initial treatment volume for EBRT includes the larynx and subdigastric, midjugular, and low jugular nodes, and uses opposed lateral upper fields matched with a single anterior lower field. The inferior border of the upper fields should be placed at the bottom of the cricoid cartilage or at least 2 cm below the inferior extent of disease when there is subglottic extension. A small cord block should be placed at the superior border of the lower field to avoid overlap of portals.

For T1 lesions, we use 50 Gy/25 fractions by EBRT plus 16 Gy/8 fractions boost by EBRT. For T2 lesions, we use 50 Gy/25 fractions by EBRT plus 20 Gy/10 fractions boost by EBRT. Adverse consequences of RT may include laryngeal edema, laryngeal stenosis, soft tissue necrosis, and cartilage necrosis.

The outcome from EBRT may be enhanced by maneuvers such as accelerated treatment plans (eg, hyperfractionation) , and the use of radiation sensitizers . In one study of hyperfractionation, RT doses of 67.2 to 72 Gy were administered in 1.6 Gy fractions twice daily over six weeks in 169 patients with cancer of the supraglottic larynx . Local control rates were 96 percent for T1 and 86 percent for T2 disease, and the corresponding relapse-free survival rates were 78 and 82 percent at a median follow-up of 56 months. Including surgical salvage, the ultimate local control rates were 96 and 93 percent, respectively. The voice preservation rate was somewhat higher in patients with T1 tumors (96 versus 80 percent with T2 tumors).

  Management of the neck  In contrast to patients with supraglottic cancers, those with early stage (T1-2) glottic cancers and a clinically negative neck do not require elective treatment of the neck nodes . However, patients with advanced (T3-4) lesions of the glottis should have elective treatment of the neck, even if clinically N0.

  Recommendation  We recommend RT as the treatment of choice for T1 lesions and nonbulky, exophytic T2 lesions. Patients who are medically unfit to undergo a voice-preserving supraglottic laryngectomy (eg, inadequate pulmonary function) should also be treated with definitive RT. For patients with T2 N+ disease, concurrent chemoradiotherapy is an appropriate choice for larynx preservation (see below).

Glottis  Glottic or true vocal cord cancer is the most common of all laryngeal cancers in the United States. Most true vocal cord cancers occur on the anterior two-thirds of the cords and a small percentage develop on the anterior commissure.

Although both RT and conservation surgery result in similar rates of survival and local control, patients with early glottic carcinomas usually undergo RT due to superior voice results. Vocal quality following RT is not normal, but is near normal in most patients. In comparison, all patients are hoarse to a variable degree following hemilaryngectomy. Patients who undergo laser excision of T1 cancers may be hoarse post-operatively, depending on the site of the lesion on the vocal cord. Most patients are hoarse to a variable degree following hemilaryngectomy for larger T1 or T2 cancers

These issues were illustrated in a retrospective study that reviewed outcomes of 551 patients treated with conservative surgery or RT for glottic carcinoma. For early stage carcinomas (88 percent of cases), there were no significant differences in five-year survival among patients treated with endoscopic laser resection, VPL or RT. However, there were two advantages to conservative surgery: a lower local tumor recurrence rate (12 versus 27 percent) and a higher voice preservation rate (83 versus 72 percent), but not necessarily improved quality of speech.

For these reasons, RT is usually recommended for T1 lesions and for T2 lesions in which there is no impairment of vocal cord mobility . The more bulky T2 lesions, particularly those associated with impairment of motion, are better treated with voice-preserving hemilaryngectomy, particularly if the patient is obese and is at risk for radiation "geographic miss" . Another mode of therapy receiving increasing attention is endoscopic laser resection, but this procedure is more difficult in these patients because of close proximity to the underlying cartilage .

The necessity for elective neck dissection remains somewhat controversial in glottic carcinomas. Early lesions have a low incidence of occult nodal disease because the vocal cords possess little or no lymphatic drainage . One study, for example, retrospectively reviewed the course of 92 patients who had either undergone neck dissection or been observed for a minimum of two years after primary treatment . The incidence of occult nodal disease was 0 and 19 percent in early and late stage disease, respectively. For this reason, expectant management with observation of the neck is acceptable as long as the patient is reliable for frequent follow-up .

  Surgery  The surgical procedure most often used for patients with early glottic cancer is vertical partial laryngectomy (VPL). In this procedure, the surgeon bisects the larynx and removes a portion or all of the true and false vocal cord along with the ipsilateral half of the thyroid cartilage. Reconstruction can be completed with strap muscles or cervical fascia. This procedure results in acceptable five-year survival rates (90 to 95 percent for T1 and 85 percent for T2), but can be complicated by suboptimal voice quality and the need for salvage laryngectomy due to local recurrence in 2 to 15 percent of patients. If the carcinoma is bulky or associated with impaired mobility, the surgeon may elect to dissect the ipsilateral neck.

Supracricoid partial laryngectomy with cricohyoidoepiglottopexy (SCPL-CHP) is another approach that has been advocated to control selected advanced supraglottic/transglottic tumors classified as T3-T4 for which the conventional surgical alternative would have been total or near-total laryngectomy . The avoidance of a permanent stoma for respiration and the continuity of laryngeal phonation following surgery are significant advantages. (See "Speech and swallowing rehabilitation of the patient with head and neck cancer"). For the majority of patients, these advantages greatly outweigh the temporary but occasionally severe dysphagia during the acute post-operative period, and permanent alterations in vocal quality associated with the procedure. These procedures require adequate pulmonary reserve because of the temporary aspiration that usually occurs following surgery.

  Endoscopic laser resection For appropriate patients, endoscopic laser resection of early glottic cancers allows simultaneous biopsy, staging, and definitive treatment. Earlier studies included only T1 lesions that did not extend to the anterior commissure, periarytenoid, or subglottic areas, while more recent studies include T1b lesions (cross anterior commissure) as well as some T2 lesions (extend to sub- or supraglottis by definition) that are treated with narrow margin laser resection . When less than one-half of the cordal depth is resected, two-thirds of the working thyroarytenoid muscle remains, allowing for a functional voice. Local tumor control and voice quality appear to be acceptable when laser therapy is used for T1 lesions of the vocal cord, as illustrated by the following data:

  • In the largest series, 151 patients with T1 or Tis  glottic carcinoma underwent CO2 laser cordectomy, the three-year local control rates for Tis, T1a, and T1b tumors were 100, 94, and 91 percent, respective .
  • The success of salvage therapy for local recurrence was illustrated in a report in which 40 of 48 patients with Tis or T1 glottic lesions were treated with endoscopic resection; four had disease persistence despite multiple procedures, and four recurred locally. All eight were successfully salvaged with RT, hemilaryngectomy, or total laryngectomy; the four-year disease control rate was 100 percent, with a laryngeal preservation rate of 96 percent.

Although randomized trials are not available, laser therapy seems to result in similar local control, laryngeal preservation, and survival as RT or surgery . One study compared 31 patients with T1N0M0 glottic carcinoma undergoing laser microsurgery with 41 who underwent RT, and 34 who had partial laryngectomy . The five and ten year rates of locoregional control and overall survival were 91 and 87 percent, and 78 and 62 percent, respectively, and did not differ by procedure. There was a trend toward partial laryngectomy being associated with a worse satisfaction, and more hoarseness and breathiness than either radiotherapy or laser microsurgery.

  Radiation  Local control is achieved with RT in approximately 90 percent of patients with T1 lesions and 70 to 80 percent of those with T2 lesions . In one study of 519 such patients, five-year local control rates were 94, 93, 80, and 72 percent in patients with T1A, T1B, T2A (T2 lesion with normal vocal cord mobility), and T2B (T2 lesion with impaired vocal cord mobility) disease, respectively.

Several studies have evaluated the effect of RT dose, treatment duration, and fraction size on tumor control.

  • There appears to be a significant dose-response curve for RT in T1 glottic cancer. There is a significantly lower probability of local tumor control at total doses below 61 to 65 Gy .
  • A shorter total duration of RT appears to improve outcome. In one series of 398 consecutive patients with early glottic cancer, the five-year local control rate for T2 lesions treated within 43 days versus longer than 43 days was 100 and 84 percent, respectively. Overall treatment duration was not a significant factor influencing outcome for T1 lesions in this study, but a second series suggested that the probability of five-year local control for T1N0 glottic cancer decreased from 95 to 79 percent for treatment time of 22 to 29 days versus 40 days or longer.
  • Increasing the dose administered per fraction may also improve local control . In one report, the five-year rate of local control for T2 lesions was higher at fraction sizes 2.25 Gy compared to <1.8 Gy (84 versus 44 percent) .
  • The benefit of accelerated fractionation RT is unclear. In one retrospective series of 85 patients, hyperfractionated RT (1.72 Gy per fraction, two fractions per day, 5 days per week, total 55 or 58 Gy) did not appear to improve local control probability when compared to conventional fractionation (5 daily 2 Gy fractions per week for a total of 64 Gy). The five-year local control probability for T1 tumors with accelerated and conventional fractionation was 87 and 80 percent, respectively, while for T2 tumors, it was 75 and 73 percent, respectively.

In contrast, a second retrospective report of 240 patients did suggest a benefit for hyperfractionation (twice daily to a total 74 to 80 Gy) as compared to once daily therapy (at doses ranging from 32 to 75 Gy) . The five-year local control rates were 79 and 67 percent, respectively. However, interpretation of this study is compromised by the range of doses used for single daily therapy, and the selection of patients for daily treatment who were considered poor candidates for twice daily treatment.

RT to the glottic larynx is usually administered using opposed lateral, wedged, low energy photons (5 x 5 cm or 6 x 6 cm field sizes). Some institutions advocate blocking the arytenoids after 50 Gy in order to reduce the severity of arytenoid edema; the advantage of this approach, however, is uncertain . Based upon the apparent importance of RT dose and duration, we use 66 Gy/33 fractions for T1 lesions and 70 Gy/35 fractions for T2 lesions.

Local recurrence, which occurs in 13 to 24 percent of radiated patients, usually is managed by salvage surgery. Conservative surgery (VPL or subtotal laryngectomy with cricohyoidopexy) can be used in selected cases to preserve laryngeal function .

The prognosis for local control in glottic cancers also appears to be influenced by factors unrelated to treatment. Poor prognostic factors include anterior commissure involvement, particularly in T1 lesions ; subglottic extension in T2 lesions ; bulky tumors (defined as the presence of visible rather than subclinical disease) ; and a low pretreatment hemoglobin concentration .

  Recommendation  We recommend laser resection of T1 carcinomas that are localized to the vocal cord, without extension to the arytenoid or anterior commissure. We usually recommend RT for other T1 lesions, and vertical hemilaryngectomy for appropriate patients with T2 lesions of the larynx. These patients, like those with supraglottic cancers, require adequate pulmonary reserve for successful swallowing and vocal rehabilitation. We usually recommend RT for those with inadequate pulmonary reserve.

  Management of T3N0 glottic carcinoma  The best therapeutic approach for T3N0 tumors of the glottis is uncertain. Although these patients may be considered for chemoradiotherapy approaches as for those with locoregionally advanced disease (see below), selected patients may do well with conservation surgery with or without neoadjuvant chemotherapy [7] or RT alone.

One series reviewed outcomes in 200 patients with T3N0 glottic cancer who were treated by a variety of approaches, including total laryngectomy with or without RT, conservation surgery with and without RT, or RT alone [64] . At five years, the disease-specific and overall survival rates were 67 and 54 percent respectively, and did not differ according to treatment modality. Locoregional control and laryngeal preservation was achieved in 74 and 70 percent, respectively, and also did not vary according to treatment.

Subglottis The subglottic region of the larynx is defined as the cylindrical area bordered inferiorly by the lower margin of the cricoid cartilage and by an imaginary circle 5 mm below the free margin of the true vocal cords. Tumors in this region can be either primary or secondary due to subglottic extension from some other site, usually the glottis. Primary malignant lesions of the subglottis are rare; in one review of 2201 patients diagnosed with laryngeal cancer over a 33-year period at one institution, only 1.8 percent had primary lesions of the subglottis .

Subglottic tumors are usually asymptomatic but can present with hoarseness, dyspnea, or stridor. Direct extralaryngeal extension is common, and the disease is often advanced at presentation . For these reasons, survival figures are lower than those observed at other laryngeal subsites. As an example, in the study described above, 19 patients with T1 or T2 subglottic cancer were treated with total laryngectomy, RT, or combination therapy. Overall and disease-free five-year survival was 86 and 71 percent for T1 and 50 and 42 percent for T2 disease.

In contrast to lesions involving the supraglottic larynx and true glottis, total laryngectomy and appropriate neck surgery, including thyroidectomy, followed by RT is usually recommended. Prophylactic neck dissection has not been shown to improve survival.

For patients who desire laryngeal conservation, primary RT followed by surgical salvage may be considered, although no study has directly compared this approach to upfront surgery followed by RT, and it is not clear how this compares to primary surgery . If this approach is utilized, chemotherapy should probably be added, at least for T3/4 tumors, based upon data from glottic and supraglottic cancers.

Summary and ASCO guidelines  The available evidence supports the use of larynx preservation approaches for the treatment of patients with early stage (T1 or 2) invasive laryngeal cancer. However, no larynx preservation approach offers a survival advantage over total laryngectomy and adjuvant therapy. Treatment guidelines for laryngeal cancer from the American Society of Clinical Oncology, published in 2006, recommend that all patients with T1 or T2 larynx cancer be treated, at least initially, with intent to preserve the larynx

TREATMENT OF LOCOREGIONALLY ADVANCED DISEASE  Locoregionally advanced  squamous cell carcinoma of the larynx represents a difficult management problem. Long-term survival rates range from 20 to 60 percent, depending upon the site, stage, and resectability of the tumor. Traditional therapy for patients whose cancers are resectable has generally consisted of surgery followed by adjuvant RT or RT alone. However, these approaches have potential limitations:

  • One would like to minimize the extent of surgery without sacrificing survival
  • RT may be compromised by factors such as repopulation of tumor cells, tumor hypoxia, and resistance to radiation

Organ-sparing approaches  Because of these limitations, organ sparing approaches, most of which incorporate chemotherapy, have been evaluated in patients with locoregionally advanced laryngeal cancer. Neoadjuvant or induction chemotherapy, followed by RT permits larynx preservation in a high percentage of such patients, but does not improve survival compared to radiation alone, and in the seminal US Intergroup 91-11 study , the rates of larynx preservation were lower than achieved by concomitant chemoradiotherapy. Induction chemotherapy will not be discussed further here, and is presented in detail elsewhere. (See "Organ preservation in locoregionally advanced laryngeal and hypopharyngeal cancer: Neoadjuvant chemotherapy").

Concomitant chemoradiotherapy (CRT) combines RT with chemotherapy in an attempt to take advantage of synergistic effects and provide early eradication of micrometastases. Randomized studies of CRT in patients with unresectable disease have demonstrated survival that is similar to what would be expected in locoregionally advanced, resectable disease. For this reason, a number of studies have been conducted utilizing CRT as a substitute for surgery in patients with otherwise resectable laryngeal cancer, with the major goal of organ preservation. Although there are no controlled data concerning the effect of these regimens on survival compared to surgery alone, numerous phase II and randomized trials support the superiority of CRT over RT alone in terms of locoregional control, but not survival. This topic is discussed in detail elsewhere.

Although randomized trials have not shown a survival benefit for either concomitant CRT or induction chemotherapy followed by RT in patients with locoregionally potentially resectable advanced laryngeal cancer (compared to surgery with or without RT), there is a significant benefit in terms of laryngectomy-free survival and locoregional control in favor of concomitant CRT. This was demonstrated in the phase III United States intergroup trial 91-11, in which 547 patients with stage III or IV potentially resectable laryngeal cancer were randomly assigned to radiation alone, induction chemotherapy followed by radiation, or concomitant chemoradiotherapy utilizing cisplatin every three weeks plus RT .

The primary endpoint was laryngectomy-free survival (LFS). In the initial report, with a median follow-up of 3.8 years, two year LFS rates significantly favored the concomitant chemoradiotherapy group (88 versus 75 and 70 percent for neoadjuvant chemotherapy and radiation alone, respectively), and rates of local control were also significantly better in this group (80 versus 64 and 58 percent, respectively. There were no significant differences when induction chemotherapy followed by radiation was compared to radiation alone, and overall survival was nearly identical among all three groups (two-year and five-year survival rates 74 to 76, and 54 to 56 percent, respectively).

In a later report presented at the 2006 meeting of the American Society of Clinical Oncology (ASCO), there were no longer any differences in LFS between the groups receiving concomitant chemoradiotherapy and induction chemotherapy followed by radiation (47 versus 45 percent at 5 years, respectively). However, the overall laryngeal preservation rate still favored concomitant therapy (84 versus 71 percent respectively, compared to 66 percent with radiation alone), as did locoregional control rates.

Additional results of this trial are described in detail elsewhere. Largely based upon these data, concomitant chemoradiotherapy is preferred for patients with locoregionally advanced laryngeal cancer.

Regimens testing induction chemotherapy followed by concomitant chemoradiotherapy are under investigation, and appear quite promising . At least three randomized studies designed to determine whether the sequential integration of induction chemotherapy followed by concurrent chemoradiotherapy improves outcomes compared to concurrent chemoradiotherapy alone are now in progress. Although none are limited to laryngeal preservation, the implication of their results for nonsurgical approaches to laryngeal and other head and neck tumors are obvious.

  Contraindications There are no validated markers that consistently and reliably predict outcome from organ-preserving therapy for patients with locoregionally advanced laryngeal cancer. However, largely based upon analyses of data from the VA laryngeal cancer study [71] as well as uncontrolled prospective trials and retrospective reports, patients with penetration of tumor through cartilage into soft tissues are considered poor candidates for a larynx-preserving approach . Primary surgery, usually a total laryngectomy, is commonly recommended in this setting.

Management of the neck  Patients with advanced primary lesions should have elective treatment of the neck, even if clinically N0. Treatment guidelines from ASCO recommend that patients with clinically involved regional cervical lymph nodes (N1,  who are treated with definitive RT or chemoradiotherapy, and who have a complete clinical response do not require elective neck dissection. Neck dissection should be performed for patients who do not have a complete clinical response to radiation therapy.

Surgical treatment of the neck is recommended for all patients with N2 or N3 disease who are treated with definitive RT or chemoradiotherapy, regardless of response . The rationale for this recommendation includes the following

  • There is no standard imaging approach in this setting that can reliably predict which patients harbor residual disease.
  • Salvage surgery for recurrent disease in the neck is rarely successful of subsequently required in this setting.

Patients with clinically involved cervical lymph nodes who are treated with surgery for the primary lesion should undergo neck dissection. If there are poor risk features, adjuvant concomitant chemoradiotherapy is indicated.

Summary and ASCO guidelines  Organ-sparing approaches, including induction chemotherapy followed by RT, concomitant chemoradiotherapy with or without induction chemotherapy, and, in some cases, radiation alone, permit larynx preservation in patients with locoregionally advanced laryngeal cancer, but none provides a survival advantage over laryngectomy. Nevertheless, because of the importance of the larynx to speech and swallowing function, all patients with advanced cancer of the larynx or pyriform sinus should be offered the option of organ preservation, unless contraindicated.

In keeping with this general philosophy, clinical practice guidelines from ASCO recommend that for patients with T3 or T4 laryngeal cancers without tumor invasion through cartilage into soft tissue, larynx preservation is an appropriate standard treatment approach In general, concomitant chemoradiotherapy is preferred over induction chemotherapy followed by RT.

POSTTREATMENT FOLLOW-UP  The goal of posttreatment surveillance is improved survival through early detection of recurrent disease and identification of second primary cancers. Principles of treatment for recurrent disease and second primary HNCs is covered in detail elsewhere.

Continued cigarette smoking appears to be associated with a worse outcome after RT. Patients should be encouraged to abstain from smoking after diagnosis, throughout treatment, and thereafter.

Patients with HNC are more likely to develop second primary cancers than any other group of patients with malignancy. This probably reflects the wide distribution of the toxic effects of tobacco and alcohol, the major risk factors for HNC.

The late development of second primary tumors is the most common cause of posttreatment "failure" after 36 months. The major sites are head and neck, lung, and esophagus. Patients with supraglottic laryngeal cancer are at a particularly high risk of developing metachronous lung cancers

Despite the lack of defined survival benefit from any posttreatment surveillance strategy, surveillance protocols are in widespread clinical use after curative-intent therapy for HNC. In general, the intensity of follow-up is greatest in the first two to three years, which is the period of greatest risk for disease recurrence.