CHAPTER 104 Radiotherapy and Chemotherapy of Squamous Cell Carcinomas of the Hypopharynx and Esophagus
The treatment of squamous cell carcinomas (SCCs) of hypopharyngeal and esophageal malignancies remains a real challenge. These tumors are most often diagnosed at advanced stages in rather debilitated patients. They have high locoregional aggressiveness, as well as a high metastatic propensity. Treatment planning must take into account both the locoregional extension and the frequent comorbid illnesses and is too often considered with palliative intent. Surgery with postoperative irradiation has long been the only one treatment performed with curative intent. This surgery was in the vast majority of the cases rather mutilating and had, in particular for esophageal cancer, a notable morbidity. The development of the use of a combination of irradiation and chemotherapy has somewhat modified these therapeutic approaches. If globally, the outcome in terms of disease control and survival has not been tremendously improved, the organ-preserving strategies have allowed, in selected cases, to improve the quality of life without jeopardizing survival. A multidisciplinary decision-making approach that includes the surgeon, medical oncologist, radiation therapist, imaging specialist, and general practitioner is mandatory to better treat these patients.
The first study was a prospective epidemiologic analysis carried out on 339 new successive hypopharynx patients. This population consisted of 329 males (97%) with a mean age of 57 years and 10 females (3%) with a mean age of 55 years. Of the patients, 114 (34%) lived alone at time of diagnosis. Only 95 patients (28%) were employed, 38% were retired, 20% were disabled, 13% were unemployed, and 1% (all females) had never had any occupation. Of the patients, 330 (97%) acknowledged they had both used tobacco and abused alcohol. The mean lifelong consumption of tobacco was 317 kg of tobacco smoked before diagnosis, ranging between 13 and 998 kg. The mean lifelong consumption of alcohol was as high as 1654 kg of equivalent of pure ethanol, ranging between 55 and 6074 kg. Such impressive data explained why more than half the patients presented with comorbid illnesses.
The second study was a retrospective review of the clinical charts of 652 new successive hypopharynx patients. Again, 97% of patients were males. All tumors were SCC. Of the patients, 346 had at least one comorbid illness (e.g., liver cirrhosis in 11% of the patients). Stage I included only 4% of the cases, with a similar figure (3%) for stage II, while stages III and IV made up 40% and 53%, respectively. Metastatic cervical lymph nodes were found on either palpation or computed tomography (CT) scan in 78% of patients.
Surgery, combined with postoperative irradiation, and definitive irradiation (with surgery in reserve for salvage in case of local failure) have long been the only two options for treatment of hypopharyngeal cancers. As for larynx cancers, no randomized comparison of both approaches has been carried out. Comparison of these two approaches is particularly inappropriate for hypopharynx cancers. As most of the patients present with locoregionally advanced diseases and notable comorbidities, surgical series are assessed on the basis of selected cases (resectable diseases in operable patients); however, radiotherapeutic series include cases that are expected to conclude with much more unfavorable results.
Partial surgery of the hypopharynx has been proposed for early diseases. This surgery may consist of partial lateral or posterior hypopharyngectomy or of supraglottic hemilaryngopharyngectomy. The indication of such a surgery is rare because it requires a thorough assessment of patients’ local extension and performance status. When indicated, this surgery is given a local control as high as 90%.1,2 Laser CO2 surgery has also been advocated with similar results.3 This surgery may generate laryngopharyngeal dyssynergia with frequent aspirations, particularly when postoperative irradiation is delivered or when a cricoid resection has been necessary. Adjuvant irradiation is most often required because of the risk of submucosal spread and the frequent nodal involvement. The reported 5-year survival rates range around 50%.
Definitive radiation therapy has also been proposed. Several series have shown similar results, with probably slightly better functional results,4–6 but, in case of failure in about all resectable local recurrences, salvage surgery resulted in a total laryngectomy.
No straightforward rule exists for selection of either partial surgery or irradiation. In general radiation is proposed for exophytic tumors located in the “membranous part” of the hypopharynx (i.e., behind the thyrohyoid membrane). For more infiltrating tumors and tumors located in the “cartilaginous part” of the hypopharynx (i.e., behind the thyroid ala), surgery is the preferable treatment. When irradiation is proposed, a radical neck dissection followed by radiotherapy to the primary tumor and to the neck seemed to improve the outcome in case of T1T2 with enlarged metastatic lymph nodes.7
The majority of hypopharynx cancers, when resectable, are only able to endure a total laryngectomy with a partial pharyngectomy performed with a radical neck dissection. Postoperative irradiation improves the results of surgery.8 This surgery is able to control 85% of the cancers above the clavicles in French series.9–11 Despite this advantageous local control rate, the 5-year survival rate does not exceed 35%, most of deaths being attributed to distant metastases, intercurrent diseases, or, to a lesser extent, second primary tumors. In addition, a major structural and functional alteration to the patient’s upper aerodigestive tract is incurred. All previously untreated patients must receive postoperative irradiation to the surgical bed and both sides of the neck. Comparison between preoperative and postoperative irradiation showed that the postoperative results were superior.12
Endoscopic laser CO2 surgery has also been advocated by some teams for advanced diseases of the hypopharynx. A recent retrospective study reported (5-year local control as high as 75% for pT3 and 57% for pT4a, while 5-year recurrence-free survival was of 59% for stage III and 47% for stage IVa.13 This surgery is indicated for selected cases, although its application widens in the hands of experienced surgeons.
Conventional irradiation often gives poor results in T3 and T4 diseases both in terms of local control and survival far under the locoregional control rates seen after surgery and postoperative irradiation.4,5,14
For a long time, chemotherapy was used primarily for palliative purposes because of its poor outcomes in the treatment of cancer of the head and neck. A new era in the management of cancer of the head and neck was initiated in the early 1980s with the introduction of platinum-based chemotherapeutic regimens. Actually, clinical studies using cisplatin as a single agent or more often in combination with other drugs have demonstrated partial response rates of 40% to 50% and complete response rates of 26% to 50%, for an overall response rate ranging between 78% and 94%.15,16 In the early 1990s it appeared obvious that this approach had not provided any improvement in survival benefit for most patients. This was clearly demonstrated by the major meta-analysis carried out at the Institute Gustave Roussy (MACH-NC analysis)17 including more than 10,000 patients enrolled in randomized trials comparing conventional treatment with the same treatment with chemotherapy (either adjuvant or induction or concomitant chemotherapy). The conclusion of this meta-analysis was that adjuvant chemotherapy, as well as induction chemotherapy (4%), produced an insignificant increase in survival (1%). But when induction chemotherapy consisted of cisplatin and fluorouracil, a significant 5% benefit in survival was achieved. On the contrary, concomitant chemoradiotherapy resulted in a significant increase in survival to 8%, but this group had a mixture of definitive and postoperative radiation trials. However, additional trials of chemoradiation as the first line of treatment, which were published after this meta-analysis, confirmed the trend toward an improved outcome.18–20 This meta-analysis was updated in 200721 including 16,665 patients and confirmed that concurrent chemoradiation provided the highest benefit in survival (hazard ratio: 0.81), particularly for platinum-based chemotherapy (hazard ratio: 0.75), but the effects of chemotherapy decreased with increasing age. It must be underscored that these trials did not address hypopharynx cancers specifically. As a result, chemoradiation remains an important treatment that is still under evaluation for this primary site.
In an attempt to improve the results of radical surgery with postoperative irradiation, chemotherapy has been evaluated. In 1990 the European Organization for Research and Treatment of Cancer (EORTC) Head and Neck Group reported a randomized trial22 comparing this approach with the same treatment preceded by one course of vincristine with bleomycin and methotrexate. No difference was shown for locoregional control, distant metastases, and metachronous cancers. In 1996 the Institute Gustave Roussy23 reported on perioperative chemotherapy in which patients received three cycles of cisplatin-fluorouracil before surgery and two additional cycles between surgery and postoperative irradiation without any improvement in outcome when compared with historical controls without chemotherapy.
A French trial24 assessed the role of induction chemotherapy consisting of three cycles of the conventional regimen cisplatinum-fluorouracil. In one arm this chemotherapy was followed—whatever the response to chemotherapy—by a total laryngectomy with a radical neck dissection and postoperative irradiation. The other arm consisted of the same chemotherapy protocol followed again—whatever the response—by definitive irradiation, keeping the surgery in reserve for salvage. The aim of this trial, to evaluate a larynx preservation approach, is discussed later in this chapter, but the “surgical” arm may be analyzed separately in regard to the impact of chemotherapy. In this arm the 5-year survival was 37%. The fact that induction chemotherapy has not tremendously improved survival has been often underscored, whatever the primary site.
Another way to try to improve the results of radio-surgical protocols is to deliver postoperative concurrent chemotherapy and irradiation25–28 in high-risk patients (e.g., positive margins, extracapsular spread, multiple nodal involvement, perineural invasion). One of these situations is often encountered after hypopharynx resection. Two recent randomized trials (RTOG 95-01 and EORTC 22-981) have explored the same protocol consisting of conventional 70 Gy external beam irradiation with cisplatin on days 1, 22, and 43 at a dosage of 100 mg/m2. Randomized or not, none of these studies have focused on hypopharynx cancer only.
The Radiation Therapy Oncology Group RTOG 95-01 trial29 randomized patients with two or more positive lymph nodes or extracapsular spread or positive margins. This trial failed to find a significant difference. A trend appeared for a slight difference in 2-year estimates for locoregional control in the experimental arm (79% vs. 74%, P = 0.16) and in 2-year estimates for overall survival (63% vs. 57%, P = 0.51), while a borderline significant difference appeared for disease-free survival (54% vs. 43%, P = 0.049). The EORTC 22-951 trial30 included many more clinical situations (stage III and IV or two or more positive lymph nodes or extracapsular spread or positive margins or vascular embolisms or perineural invasion or positive lymph nodes in levels 4 or 5). The interim analysis demonstrated a significant increase in 3-year estimates for disease-free survival (59% vs. 41%, P = 0.0096), for overall survival (65% vs. 49%, P = 0.0057), for time to progression (56% vs. 44%, P = 0.0016), and for local control (83% vs. 64%). Longer-term results and detailed comparison of both trials should help in the future understanding of this discrepancy.
Altered fractionation has been explored in advanced head and neck cancers. The largest randomized trial was a four-arm trial31 carried out by the RTOG (RTOG 90-03) that compared (1) conventional irradiation with (2) a split course accelerated fractionation with (3) a delayed concomitant boost and with (4) a hyperfractionated irradiation. Previously untreated advanced head and neck carcinomas were eligible for this study. Of the 1073 patients enrolled in the study, 13% had hypopharyngeal cancers. Both hyperfractionated irradiation and concomitant boost showed significantly better locoregional control, disease-free survival, disease-specific survival, and overall survival, but also a higher rate of acute and, to a lesser extent, late toxicities. No data specifically studying the hypopharynx have been published. Another large meta-analysis32 pooled together updated individual patient data from randomized trials comparing conventional irradiation with unconventional radiotherapy schedules (hyperfractionation, acceleration, acceleration with dose reduction). A total of 6515 patients had been randomized in 15 trials. With a median follow-up of 6 years there was a significant survival benefit with altered fractionated radiotherapy (+3.4% survival at 5 years); the highest benefit (+8% at 5 years) was achieved with hyperfractionated irradiation. Again, few patients had hypopharyngeal diseases.
As discussed earlier in this chapter, induction chemotherapy using cisplatin and fluorouracil provided impressive response rates.33 Chemosensitive tumors’ radiosensitivity was confirmed in a number of cases and suggested the ability of induction chemotherapy to predict radiosensitivity.34 This led some teams to use a scheme of induction chemotherapy as a selective test for two groups of patients: good responders are candidates for subsequent radiation therapy and poor responders are still candidates for subsequent surgery. The larynx has been the most investigated primary site for this new approach, most often in noncontrolled trials using historical comparisons with surgical series.
Many published series have assessed in a nonrandomized fashion the reliability of induction chemotherapy–based protocols. They usually reported small series of cases mixing laryngeal, hypopharyngeal, and oropharyngeal tumors or tumors amenable to partial surgery and to radical surgery. In addition, the concept of organ preservation or preservation of function was unclear. Finally, larynx preservation rates were presented under different forms: for the overall population at a precise follow-up time, in survivors at that time, or according to a median follow-up. All in all, larynx preservation could be achieved in one third to one half of patients according to the patient selection.
Only two randomized trials focused on cancer of the hypopharynx. The first trial was conducted by the EORTC35 on cancer of the hypopharynx eligible for total laryngectomy with partial pharyngectomy. In this study 202 patients were enrolled in which the standard treatment (surgery and postoperative radiation) was compared with two or three cycles of chemotherapy, followed in clinically complete responders at the primary site by irradiation, or, for other patients, by the conventional treatment. A notable difference appeared in the median survival, favoring the experimental arm (44 months) when compared with the surgery arm (25 months). This difference in median survival was only explained by the fact that distant metastases appeared later after chemotherapy, leading to a better survival at 3 years, which was not shown at 5 years. There was no significant difference in locoregional control or in second primaries’ appearance. However, at 3 and 5 years, half the survivors in the chemotherapy arm had retained their larynx.
The second trial24 compared 92 patients with resectable cancer of the hypopharynx (only amenable to total laryngectomy-pharyngectomy) with the outcome of patients randomly assigned to receive induction chemotherapy. In one arm, chemotherapy was followed by surgery and postoperative radiation therapy. In the other arm, chemotherapy was followed by radiation, with surgery reserved for salvage. In this trial chemosensitivity was not taken into consideration for deciding a subsequent treatment. With a median follow-up of 92 months, researchers observed a better 5-year survival (37%) and a better local control (63%) in the surgery arm when compared with the radiotherapy arm (19% and 39%, respectively). The comparison of this trial with the EORTC trial suggests that the cancer’s sensitivity to chemotherapy must be taken into account before deciding a subsequent radiation therapy instead of surgery.
In 2000 the previously mentioned meta-analysis17 assessed separately three randomized trials on larynx preservation based on induction chemotherapy. There was an insignificant 6% better survival in the surgical arm, but it was balanced by a 56% larynx preservation rate in the induction arm.
Induction chemotherapy has once again gained interest after the publication of two randomized trials comparing the conventional cisplatin 5-fluorouracil regimen with a triplet mixing cisplatin, fluorouracil, and docetaxel either in nonresectable diseases followed by irradiation alone36 or in resectable/nonresectable diseases followed by concurrent chemoradiotherapy.37 In both trials the triplet arm achieved significantly better overall and progression-free survivals.
In 2009 the Groupe d’Oncologie Radiothérapie Tête Et Cou (GORTEC)38 published the results of their induction chemotherapy–based larynx preservation trial for larynx and hypopharynx cancer candidates for a total laryngectomy. Eligible patients were randomly assigned to receive either three cycles of cisplatin and fluorouracil in one arm or, in the other arm, three cycles of cisplatin with 5-fluorouracil and docetaxel. In both arms, following this induction chemotherapy, responders over 50% went to radiation therapy while other patients went to radical surgery and postoperative irradiation. The larynx preservation rate was significantly higher in the triplet chemotherapy arm but without significant difference in survival.
As discussed earlier in this chapter, there are promising data on concurrent chemoradiation. The questions are (1) how many patients in daily community practice are able to undergo such toxic treatments? (2) what is the largest disease stage amenable to larynx-preserving strategies? (are clinically defined T4 or large necrotic lymph nodes acceptable?) (3) what are the most appropriate tools used to detect as soon as possible residual diseases to be treated by salvage surgery? and (4) what would be the postoperative morbidity of this salvage surgery? These concerns are of the utmost importance if the cancer is initially operable.
As far as cancer of the hypopharynx is concerned, it must be underscored that (1) this site is rarely studied separately; (2) in case of mixed sites, the hypopharynx represents just a few cases; (3) little is known about the selection of patients with cancer of the hypopharynx (detailed locoregional extension, performance status); and (4) no information is provided about the possibilities of surgical salvage.
The first randomized trial issuing larynx preservation with concurrent chemoradiotherapy to be published was the RTOG 91-11 trial.39 But it concerned only laryngeal sites. Preliminary results suggested that concurrent chemotherapy was able to significantly increase the rate of larynx preservation when compared with induction chemotherapy and with radiotherapy alone. However, there was no difference in survival (the 5-year, disease-free survival was 39%) and, in particular, there was no difference in laryngectomy-free survival, which was 46% at 5 years.40
In 2007 the EORTC head and neck and radiotherapy groups published the second trial on larynx preservation comparing larynx and hypopharynx cancers candidates for radical surgery including a total laryngectomy. This trial compared the sequential administration of induction chemotherapy and irradiation versus their alternating administration (EORTC 24954 trial).41 No significant difference was found between both arms for either the primary end point survival with a functional larynx in place) or the secondary end points (overall survival, progression-free survival, and larynx preservation) whatever the primary site (larynx or hypopharynx).