Stage T3 squamous cell carcinoma of the glottic larynx: A comparison of laryngectomy and irradiation

In, J Rodrulron @X’dlJRi~ Bml Ph?:\ Vol. Printed in the U S.A. All nghts reserved.

0360.3016192 $5.00 + .OO ropynght 0 I992 Pergamon Press Ltd.

23, PP. 125-732

??Clinical Original Contribution

STAGE

WILLIAM

T3 SQUAMOUS A COMPARISON

M. MENDENHALL, NICHOLAS

CELL CARCINOMA OF THE GLOTTIC LARYNX: OF LARYNGECTOMY AND IRRADIATION

M.D.,*

J. CASSISI,

JAMES T. PARSONS,

D.D.S.,

M.D.7

M.D.,*

AND RODNEY

SCOTT P. STRINGER, R. MILLION,

M.D.,?

M.D.*

University of Florida College of Medicine, Gainesville, FL One-hundred eighteen patients with previously untreated T3 squamous cell carcinoma of the glottic larynx were treated with curative intent between March 1965 and November 1988 at the University of Florida. All patients were observed for at least 2 years and 83% were observed for 5 or more years. Fifty-three patients were treated with irradiation alone and 65 patients were treated with surgery alone (32) or combined with irradiation (33). Thirty-two patients treated with irradiation alone had twice-daily fractionation and the remainder had oncedaily fractionation. The local-regional control rates, including patients successfully salvaged after a local-regional recurrence, were 81% after irradiation alone and 81% after surgery alone or combined with adjuvant irradiation. The local control rates for patients treated with irradiation alone were 53% after once-daily fractionation and 71% after twice-daily fractionation. There was no relationship between vocal cord mobility at 5000 cCy, at the end of radiotherapy, or at 1 month after treatment and subsequent local control. The 5-year cause-specific survival rates were 74% for patients treated with irradiation alone and 63% for patients treated surgically. The incidence of severe complications, including those associated with salvage procedures, was 15% for both treatment groups. The rates of laryngeal voice preservation were 66% after irradiation alone and 2% after surgery. Irradiation alone for selected patients with T3 glottic cancer resulted in similar rates of local-regional control, survival, and severe complications, with a significantly higher likelihood of voice preservation, compared with surgery. Laryngeal neoplasms,

Radiotherapy.

therapy. Patients were staged according to the recommendations of the American Joint Committee on Cancer (4). All patients had at least 2 years of follow-up and 98 (83%) had 25 years of follow-up. One patient treated with a laryngectomy was lost to follow-up 3 years after surgery and has been coded as dead of intercurrent disease. No other patients were lost to follow-up. Forty patients developed recurrent disease after treatment: 34 (85%) within 2 years, 37 (93%) within 3 years, and 39 (98%) within 5 years. The remaining patient developed an isolated neck recurrence 63 months after irradiation. Sixty-five patients were treated surgically, 63 (97%) with a total laryngectomy, and two with a hemilaryngectomy. A unilateral neck dissection was performed in conjunction with laryngectomy in 42 patients and a bilateral neck dissection was performed in one patient. Twenty-five patients treated with surgery alone or combined with postoperative irradiation had an elective neck dissection; positive nodes were found in four patients (16%). Irradiation was combined with surgery in 33 patients, preoperatively in seven

INTRODUCTION The management of T3 squamous cell carcinoma of the glottic larynx is controversial. The options available to treat the primary lesion include total laryngectomy, extended hemilaryngectomy, and irradiation alone with laryngectomy reserved for local recurrence (5, 9, 10, 15, 18, 19, 27, 35, 36, 37, 40). Discussion of these alternatives by their proponents generates, at times, more heat than light (7). The purpose of this paper is to compare the results of laryngectomy to those obtained with irradiation alone.

METHODS

AND

MATERIALS

One hundred eighteen patients were treated with irradiation and/or surgery with curative intent for previously untreated T3 squamous cell carcinoma of the glottic larynx at the University of Florida between March 1965 and November 1988. No patient received adjuvant chemo-

Reprint requests to: William M. Mendenhall, M.D., Dept. of Radiation Oncology, University of Florida Health Sciences Center, P.O. Box 100385, Gainesville, FL 32610-0385.

Presented at the 33rd Annual Meeting of the American Society for Therapeutic Radiology and Oncology, 4-8 November 199 1, Washington, DC. * Dept. of Radiation Oncology. + Dept. of Otolaryngology.

Accepted

125

for publication

15 November

199 1.

726

1. .I. Radiation

Oncology

0 Biology 0 Physics

patients and postoperatively in 26 patients. All irradiated patients were treated with megavoltage equipment, with ‘(‘CO in 31 patients and 2 MV x-rays in two patients. Irradiation was delivered one fraction per day, 5 days per week. Twenty-seven patients were treated with continuous-course irradiation and six patients were treated with the planned split-course technique (2, 24). Patients treated with preoperative irradiation received 2830-5870 cGy at 180-200 cGy per fraction followed by surgery 4-6 weeks after the completion of irradiation (38). One additional patient received 1130 cGy in one fraction for palliation and, after a good response was noted, had a total laryngectomy with curative intent. Patients receiving postoperative irradiation generally began treatment within 6 weeks of surgery and were irradiated to 4685-6800 cGy at 180-200 cGy per fraction. depending on the suspected or known amount of residual disease (3). One additional patient received only 2380 cGy over 12 fractions postoperatively because treatment was discontinued against medical advice. Fifty-three patients were treated with irradiation alone. Four patients refused laryngectomy and two were medically inoperable; the remaining 47 patients were offered a course of irradiation alone. All patients were treated with megavoltage equipment: h°C~ (46 patients), 2 MV x-ray (two patients), 6 MV x-ray (one patient). and 8 and/ or 17 MV x-ray alone or combined with ‘(‘Co (four patients). Twenty-one patients were treated once-daily, I9 with continuous-course irradiation. and two with the planned split-course technique. Since 1979. 33 patients have been treated with continuous-course twice-daily irradiation (26, 27). The treatment techniques for patients treated with irradiation alone have been described (19. 2 1, 26, 27). Patients treated with twice-daily irradiation received 120 cGy per fraction with a minimum 4-hour interfraction interval, to total doses of 7440-7680 cGy: one patient received 7920 cGy. None of the patients treated with irradiation alone underwent a neck dissection. The method by which patients were selected for laryngectomy or irradiation varied over time. Early in the study, most patients were treated surgically regardless of the extent of their cancer. Later in the study, patients with relatively favorable T3 cancers were offered the alternative of irradiation alone. As we gained experience with treatment by irradiation alone, we have offered this option to a larger proportion of patients presenting with T3 glottic carcinoma. Lesions considered suitable for irradiation alone are unilateral, relatively exophytic. and are associated with an adequate airway. The patient must be reliable and return for frequent follow-up examinations following irradiation. The percentage of patients treated with irradiation alone according to study period is as follows: 1965- 1969, four (20%) of 20 patients: 1970-l 979, I3 (30%) of 44 patients; and 1980-1988, 36 (67%) of 54 patients. It is difficult to compare the extent of the primary lesion in patients treated surgically and patients treated with irradiation alone because of a relative paucity of data

Volume 23, Number

4. 1992

in some of the records of patients included in the early part of the study. The pretreatment characteristics of the patient population are outlined in Table 1. A pretreatment computed tomogram of the head and neck was obtained in 44 patients, 34 (64%) of those treated with irradiation alone and 10 ( 15%) of those treated surgically. Three patients underwent a pretreatment magnetic resonance imaging scan of the head and neck; two were treated with irradiation alone and one had a laryngectomy. Recurrence after treatment with irradiation alone was scored as local, neck, and/or distant (below the clavicles). Because it is difficult to discern a local recurrence from a neck failure after a laryngectomy, recurrence following surgery was coded as local-regional (primary and/or neck) and/or distant (3). All patients were included in the product-limit calculation of local-regional control. Patients were excluded from the absolute calculation of local control and/or neck control if they died within 2 years of treatment with that site continuously disease-free (25). Additionally, two patients treated with irradiation alone were excluded from analysis of local control because one patient had a total laryngectomy 5 months after irradiation for suspected local recurrence (laryngectomy specimen was negative for tumor) and one patient had a total laryngectomy at 8 months for a laryngeal chondronecrosis; both patients were included in the analysis of local-regional control. Patients were coded as successfully salvaged after recurrence ifthey survived for more than 1 year with no evidence of disease at the site in question. All patients were included in the analyses of distant metastasis, survival, and complications. Complications were scored as severe if they necessitated hospitalization or a second operation. The rates of local-regional control, local-regional control with voice preservation, survival, and severe complications were calculated by the product-limit method (14, 31); the significance levels for the differences between

Table 1. Patient population Irradiation alone (53 patients) Parameter AJCC stage (4) T3NO T3N I T3N2-3 Supragtottic extension Present Absent No data Subglottic extension Present Absent No data Pretreatment tracheostomy Present

Surgery + irradiation (65 patients)

No. (%)

No. (%)

40 (75) lO(l9)

49 (75)

3 (6) 46 (87) 7 (13) 0

9 (14) 7(tt) 51 (78) 10 (15) 4 (7)

37 (70) 15 (28)

48 (74) 14 (22)

t (2)

3 (4)

4 (8)

16 (25)

T3 SCC of glottic 0 W. M. MENDENHALL

(66%) of 53 patients treated with irradiation alone and one (2%) of 65 patients treated with surgery alone or combined with irradiation. The probability of local-regional control with voice preservation at 5 years was 49% for all patients; it was 63% for those treated with irradiation alone and 6% for those treated surgically (p = .OOOl). The disease was locally controlled in 29 (64%) of 45 evaluable patients treated with irradiation alone: eight (57%) of 14 patients who developed a local recurrence were successfully salvaged for an ultimate local control rate of 37 (82%) of 45 patients. Five of six patients who were treated with irradiation alone because they refused surgery or were medically inoperable were evaluable for local control; in two of the five patients, the disease was locally controlled with irradiation. Local control with irradiation alone was observed in 23 (64%) of 36 patients presenting with a clinically negative neck and in six of nine patients with clinically positive neck nodes at diagnosis. All patients who had salvage treatment for a local recurrence after irradiation alone received a total laryngectomy. Local control with irradiation alone with respect to treatment technique is outlined in Table 3. Local control rates were higher in patients treated with twice-daily irradiation compared with once-daily irradiation. Only one of four patients requiring a pretreatment tracheostomy was evaluable for local control with irradiation alone: the patient received 7680 cGy with twice-daily fractionation and had a local recurrence. Local control rates after irradiation alone as a function of vocal cord mobility at various points in the treatment course are presented in Table 4. At these points, there is no correlation between local control and return of vocal cord mobility. Of the 11 patients who continued to have vocal cord fixation 1 month after irradiation and who had no local recurrence of tumors, six eventually recovered vocal cord mobility and five continued to have a fixed cord. Only one patient who recovered vocal cord mobility and subsequently developed cord fixation remained free of recurrence. In all other patients who experienced vocal cord fixation after the return of cord mobility, local recurrence was later noted. High-quality computed tomography of the head and neck became available at our institution in 1983. The rates of local control after irradiation alone as a function

curves were calculated using the log-rank test ( 16, 3 1). Multivariate analyses were performed using the forward stepwise log-rank tests of association of covariates (13, 3 1). The parameters that were tested in the multivariate analyses were treatment group (irradiation versus surgery alone or combined with irradiation), gender, pretreatment tracheostomy, and neck stage (clinically negative versus clinically positive).

RESULTS Local-regional control

The local-regional control rates for irradiation alone compared with surgery alone or combined with irradiation are presented in Table 2. For patients treated surgically, the local-regional control rates according to study period were as follows: 1965- 1969, 10 (77%) of 13 patients; 19701979, 20 (74%) of 27 patients; and 1980-1988, nine (75%) of 12 patients. The probability of local-regional control at 5 years for all patients was 69%; it was 62% for patients treated with irradiation alone and 75% for patients treated with surgery alone or combined with irradiation (p = . 10). The local-regional control rate was higher after surgery. However, when patients were included who were salvaged after a local-regional recurrence, the ultimate rates of disease control above the clavicles were practically identical. The probability of ultimate local-regional control at 5 years for all patients was 84%; it was 84% for patients treated with irradiation alone and 82% for patients treated surgically (y = .95). The influence of the various parameters on ultimate local-regional control in the multivariate analysis was as follows: neck stage, p = .093;gender, p = .434; pretreatment tracheostomy, p = .818;and treatment, p = .972.In one of two patients initially treated with a hemilaryngectomy, the disease was locally controlled; the other patient was successfully salvaged with a total laryngectomy after developing a local recurrence. Ten of 16 patients treated with laryngectomy who had a pretreatment tracheostomy were eligible for local-regional control analysis; in eight patients (80%) the disease was controlled above the clavicles. Neither patient with recurrence above the clavicles was salvaged. The rates of local control with laryngeal voice preservation were 35 Table 2. Local-regional

727

d ai.

control

(no. controlled/no.

treated)

Local-regional control Treatment

group

Irradiation Surgery t irradiation

Excluded* 6 13

No. (%) 29147 (62) 39152 (75)

Ultimate No. salvaged/ no. attempts 9115 317

control

No. (%) 38/47 (8 1) 42/52 (8 1)

* Nineteen patients were excluded because they died within 2 years of treatment and were continuously disease-free above the clavicles. Two patients treated with irradiation alone who required a laryngectomy for suspected local recurrence with a negative laryngectomy specimen (one patient) or for a chondronecrosis (one patient) were included in the above analysis because they were evaluable for neck control.

128

I. J. Radiation

Oncology

0 Biology 0 Physics

Table 3. Local control with irradiation alone versus treatment technique (no. controlled/no. treated)* Local control Treatment technique

No. (%) 9/17 (53)+ (53)

Once-daily fractionation Continuous course Split course Twice-daily Total

8/15

l/2 20/28 (7 I)+ 29145 (64)

fractionation

* Excludes six patients who died within 2 years of irradiation with the primary site continuously disease-free, one patient who underwent total laryngectomy at 5 months for suspected local recurrence (laryngectomy specimen negative for tumor) and one patient who underwent a total laryngectomy at 8 months for chondronecrosis. +p = 0.17.

of treatment period and fractionation technique are as follows: 1965- 1982, once-daily fractionation, eight (50%) of 16 patients; 1983- 1988, once-daily fractionation, one patient: 1979- 1982, twice-daily fractionation, four of hve patients: and 1983- 1988, twice-daily fractionation, 16 (70%) of 23 patients. Therefore, the pretreatment evaluation, which varied over the study period, had no obvious impact on the likelihood of local control after irradiation alone. Local control with irradiation alone was obtained in nine (60%) of 15 patients treated once daily with ‘“Co and in 0 of two patients treated once daily with two MV x-ray. For twice-daily fractionation, local control rates were as follows: ‘(‘Co , 18 (78%) of 23 patients; 6 MV x-ray, one of one patient; and 8 MV and/or 17 MV alone or combined with “‘CO, one of four patients.

Table 4. Local control with irradiation versus vocal cord mobility (no. controlled/no. treated)* Local control Cord mobility assessed At 5000 cGy

End of irradiation

I month after RT

Mobility Mobile Impaired Fixed No data Mobile Impaired Fixed No data Mobile Impaired Fixed

No. (%) 719 5/12 l5/21 213 12/18 3/l 13/18 l/2 14/20 417 I l/81

(42) (71)

Volume 23. Number

4. 1992

Twenty-three patients who initially had a clinically negative neck, who were treated with irradiation, and whose primary site was continuously disease-free were eligible for evaluation of the control of disease in the neck. Four received no elective neck irradiation; one experienced a neck failure and was salvaged with further irradiation. Nineteen patients received elective neck irradiation: one (5%) experienced neck failure at the posterior margin of the irradiation field at 63 months. The recurrence was unresectable because of fixation, and the patient died of recurrent tumor in the neck alone. Eight patients who initially had clinically positive neck nodes who were treated with irradiation alone, and whose primary site was continuously disease-free were eligible for evaluation of the control of neck disease. The neck disease was controlled in all patients (six of six patients with N 1 disease and in two of two patients with N2B disease).

Disfanl metastasis The overall incidence of distant metastasis was 15 (13%) of 1 18 patients. Five (9%) of 53 patients treated with irradiation alone subsequently developed distant metastases: two (6%) of 35 patients who remained continuously disease-free above the clavicles and three (17%) of 18 patients who developed recurrent disease above the clavicles. Ten (15%) of 65 patients treated with surgery alone or combined with irradiation developed distant metastasis: seven (13%) of 52 patients who remained continuously disease-free above the clavicles and three (23%) of 13 patients who experienced failure above the clavicles.

The absolute and cause-specific survival rates at 5 years for the two treatment groups are shown in Table 5. The probability of survival at 5 years for all patients was 49%: it was 55% for patients treated with irradiation alone and 45% for those treated with surgery alone or combined with irradiation (p = .119). The likelihood of cause-specific survival at 5 years for all patients was 73%; it was 75% for patients treated with irradiation alone and 71% for patients treated surgically (p = .26). The influence of the various parameters on cause-specific survival in the multivariate analysis was as follows: neck stage, JI = .044; pretreatment tracheostomy, p = .106; gender, p = .37; and treatment, p = .6 13.

(67)

C’omplications (72) (70)

Eight patients alone developed

Table 5. Five-year

(61)

* Excludes six patients who died within 2 years of treatment with the primary site continuously disease-free, one patient who underwent a total laryngectomy at 5 months for suspected local recurrence (Iaryngectomy specimen negative for tumor), and one patient undergoing a total laryngectomy at 8 months for a chondronecrosis.

(15%) initially treated with irradiation severe complications: Three patients in survival

Absolute Treatment

group

Irradiation Surgery f irradiation

Cause-specific

No. (%)

No. (%)

23/40 (58) 26/58 (45)

23/31 (74) 26/41 (63)

T3 SCC of glottic 0 W. M. MENDENHALL

whom the primary tumor was locally controlled by irradiation and five patients after salvage laryngectomy. The complications in the three patients whose tumor was locally controlled by irradiation were laryngeal edema requiring a temporary tracheostomy at 2 months, laryngeal edema necessitating a permanent tracheostomy at 10 months, and laryngeal chondronecrosis requiring a total laryngectomy at 8 months. The last patient had required a pretreatment tracheostomy and, after irradiation, had multiple deep blind biopsies for a suspected local recurrence before developing the chondronecrosis; both the biopsy specimens and the laryngectomy specimen were negative for tumor. Fourteen of 15 salvage procedures after irradiation alone included a total laryngectomy. The complications associated with salvage laryngectomy that were observed in five patients (36%) included a wound dehiscence requiring a split-thickness skin graft (one patient) and pharyngocutaneous fistulae necessitating flap reconstruction (four patients). One of these patients thought to have local recurrence had no cancer in the laryngectomy specimen. There were no fatal complications. Ten patients ( 15%) developed severe complications after surgery alone or combined with irradiation. All of these complications occurred at the time of initial treatment; none were associated with salvage treatment. The severe complications were tracheitis requiring rehospitalization (one patient), bleeding requiring a second operation (three patients), pharyngocutaneous fistula necessitating surgical intervention (one patient), laryngeal edema during preoperative irradiation requiring a tracheostomy (one patient), wound breakdown necessitating a split-thickness skin graft (two patients), airway obstruction 2 weeks after hemilaryngectomy, requiring a temporary tracheostomy (one patient), and upper gastrointestinal bleeding resulting in sepsis and death (one patient). The overall treatmentrelated mortality was one (2%) of 65 patients. The probability of severe complications at 5 years for all patients was 16%: 16% for patients treated with irradiation alone and 15% for those treated with surgery alone or combined with irradiation (p = .558). DISCUSSION There is no randomized study that compares irradiation alone to laryngectomy for T3 carcinoma of the glottic larynx. If advised that either treatment modality offers a similar chance of cure, it is unlikely that most patients would be willing to enter a randomized trial. Because it is improbable that such a study will be available in the near future, one must rely upon nonrandomized data. Local-regional control Razack et al. (30), from Roswell Park Memorial Institute, reported a series of 128 patients with T3 and T4 true vocal cord carcinomas treated with total laryngectomy alone (59 patients) or combined with a neck dissection

et al.

729

(69 patients). At least 115 patients (90%) had T3NO-1 lesions; 13 patients were treated for stage IV disease. Excluded from the series were 48 additional patients who died of intercurrent disease and 2 1 patients who died of a second cancer within 2 years of treatment. Sixty (47%) of the evaluable 128 patients developed recurrent disease above the clavicles; 12 patients were successfully salvaged by radiotherapy so that the ultimate local-regional control rate was 80/128 (63%). Yuen et al. (29), from the M. D. Anderson Hospital, reported a series of 185 patients with T3 glottic cancer treated with surgery alone (155 patients) or combined with postoperative irradiation (30 patients). All patients had a total laryngectomy. Local-regional recurrence was noted in 3 1 ( 17%) of 185 patients. No patients were excluded from the analysis. DeSanto (7), from the Mayo Clinic, reported a series of 106 patients with T3 glottic carcinoma treated with curative intent by surgery (104 patients) or irradiation alone (two patients). Included in the surgically treated group were 12 patients irradiated for earlier-stage lesions that were restaged as T3 cancer at the time of recurrence. Overall, 6% of patients experienced failure at the primary site and 15% had recurrence in the neck. No patients were excluded from the analysis. At the University of Florida, surgery alone or combined with adjuvant irradiation resulted in a localregional control rate of 75%; three patients were salvaged by additional treatment for an ultimate local-regional control rate of 8 1%. Patients were excluded from the localregional control analysis if they died within 2 years of treatment and were continuously disease-free above the clavicles (25). This biases the comparison against the University of Florida data. The local-regional control rate after irradiation alone over a 23-year period at our institution was 62%; 60% of the local-regional failures were successfully treated for an ultimate local-regional control rate of 8 1%. Others, notably Harwood and colleagues at the Princess Margaret Hospital, have reported similar results with irradiation alone (8, 10, 18,23, 33). Local control after irradiation alone was observed to be better after twice-daily irradiation than after once-daily fractionation. This is consistent with the hypothesis that local control will be improved with a similar incidence of late complications if irradiation is hyperfractionated and given over a somewhat shorter overall treatment time (12, 17, 26, 29, 34). In our experience, vocal cord mobility, assessed at various points in the treatment course, was not significantly related to subsequent local control with irradiation alone. Aanesen et al. (1) noted that evaluation of cord mobility at 5000 cGy over 25 fractions was not predictive of local control. It may be difficult to assess vocal cord mobility during treatment because of laryngeal edema and mucositis. Partial laryngectomy Conservation surgery, in the form of the extended hemilaryngectomy and the near-total laryngectomy, offers local-regional control rates that are comparable to, if not

730

I. J. Radiation

Oncology 0 Biology 0 Physics

slightly better than, those obtained with irradiation alone (5, 15, 36). However, most patients initially have lesions that are too advanced to be suitable for conservation surgery. DeSanto (7) reported that only nine (9%) of 104 patients treated surgically at the Mayo Clinic for T3 glottic carcinoma had a partial laryngectomy. Kessler et uf. ( 15) reported that of 585 patients who had a partial or total laryngectomy at the University of California at Los Angeles between 1969 and 1984,27 (4.6%) received a vertical partial laryngectomy for T3 true vocal cord cancer. Razack et al. (30) stated that “it was difficult to find advanced glottic carcinoma restricted to one true cord:” all patients reported in their study had a total laryngectomy. In contrast, 67% of patients presenting with T3 glottic cancer between 1980 and 1988 at the University of Florida were suitable for treatment with irradiation alone. Additionally, many of the patients who have a near total laryngectomy require a permanent tracheostomy, which we would code as a severe complication in a patient treated with irradiation alone. Therefore, although partial laryngectomy remains a viable option for selected patients with T3 vocal cord cancer, it is considerably less applicable than irradiation alone, and only a handful of surgeons report their results.

Survival Razack ef al. (30) reported an overall 5-year survival rate of 53% in their series of 128 surgically treated patients. Yuen et al. (40) noted a 5-year disease-free survival rate of 80% after surgery alone and 90% after surgery and postoperative irradiation. DeSanto (7) reported an 80% cause-specific survival rate with a minimum 2.5-year follow-up. Lundgren et al. (18) reported a 5-year actuarial survival rate of 5 1% in a series of 141 patients treated with irradiation alone at the Princess Margaret Hospital: 28% of their patients died of laryngeal cancer and the remainder from intercurrent disease. The policy at the Princess Margaret Hospital during the time period of the study was to treat essentially all T3 glottic cancers with irradiation alone, so that this represents a relatively unselected series. Meredith et al. (22) from the Royal Marsden Hospital, reported a series of 72 patients treated with irradiation alone (68 patients) or a total laryngectomy (four patients) for T3 glottic carcinoma. For the 68 patients treated with irradiation alone, they noted a 5-year absolute survival rate of 53% and a 5-year cause-specific survival rate of 64%. We noted a 5-year absolute survival rate of 58% and a 5-year cause-specific survival rate of 74% in selected patients treated with irradiation alone at our institution. This compares favorably with the above data and with the survival rates obtained with surgery alone or combined with adjuvant irradiation.

Complications DeSanto (7) stated that irradiation alone with surgery reserved for recurrent disease results in complication rates

Volume

23. Number

4. 1992

that are “about 10 times higher than those obtained with surgery alone.” In the data presented here, the rate of severe complications is similar for both treatment strategies.

C”oiwpreservation Following total laryngectomy, voice rehabilitation may be accomplished by one of the three methods: voice prosthesis (i.e., Singer-Blom prosthesis; Panje button); electrolarynx, or esophageal speech. Esophageal speech is difficult to master and is not used by the majority of patients following total laryngectomy. Approximately 60% of patients will successfully use a voice prosthesis long-term (i.e., 2 years or more following surgery) (39). The remainder are rehabilitated with a device such as an electrolarynx. While a voice prosthesis produces speech similar to that obtained after successful irradiation alone, the results obtained with esophageal speech or an electrolarynx are clearly inferior. Although there is little evidence that adjuvant chemotherapy improves the likelihood of cure in head and neck cancer (32). there is considerable interest in combining chemotherapy and irradiation to preserve the larynx in patients with advanced head and neck cancer who would otherwise require a laryngectomy (6, 28). The Department of Veterans Affairs Laryngeal Cancer Study Group published the only randomized study to date that evaluates the possibility of this treatment strategy (6). Patients with stage III or IV squamous cell carcinoma of the glottic or supraglottic larynx were randomly assigned to receive three cycles of cisplatin and fluorouracil and irradiation or surgery and radiotherapy. Tumor response was evaluated after two cycles of chemotherapy: if a response was observed, the patients received a third cycle of chemotherapy followed by irradiation. If not tumor response was noted, the patient was treated with surgery followed by irradiation. The estimated 2-year survival was 68% for both treatment groups: median follow-up in the study was 33 months. The study raises several issues that deserve further comment. Inclusion of Ti and T2 lesions, regardless of neck stage, in a study where almost all patients in the surgery arm were subjected to a total laryngectomy, is inappropriate. The probability of local control with larynx preservation for supraglottic laryngeal cancer treated with irradiation alone is 95% to 100% for T, lesions and 80% to 85% for T2 lesions (20). Because the probability of involved neck nodes at diagnosis is practically zero for T I and T2 glottic cancer, all of the Tl and T2 cancers involved in the study were presumably supraglottic primaries. Furthermore, the authors conclude that the preliminary results of the study “suggest a new role for chemotherapy in patients with advanced laryngeal cancer and indicate that a treatment strategy involving induction chemotherapy and definitive radiation therapy can be effective in preserving the larynx in high percentage of patients, without compromising survival.” Unfortunately,

T3 SCC of glottic 0 W. M.

as the authors note, because irradiation alone was not compared to induction chemotherapy and irradiation, “the precise contribution of chemotherapy remains uncertain.” In order to compare the data to what one might anticipate with irradiation alone, assume the following primary site and stage distribution for the 166 patients included in the chemotherapy arm of the study, (based on information presented in Table 1 of the paper): (6) T3 glottic (43 patients); T4 glottic (18 patients); Tl-T2 supraglottic ( 16 patients); T3 supraglottic (63 patients); and T4 supraglottic (26 patients). The probability of local control with laryngeal preservation after irradiation alone is estimated to be: T3 glottic, 66%; T4 glottic, 35%; TlT2 supraglottic, 90%; T3 supraglottic, 60%; and T4 supraglottic, 35% (11, 20, 21). Therefore, the estimated overall likelihood of local control with larynx preservation for the 166 patients would be 62% if they had been treated with irradiation alone. The local control rates quoted for irradiation alone are based on substantially longer followup than is available in this preliminary report. In the VA Laryngeal Cancer Study, even patients who died of treatment complications were scored as having had their larynges successfully preserved if they had not undergone a laryngectomy. Therefore, the 64% incidence of laryngeal preservation in the chemotherapy arm of the VA Laryngeal Cancer Study is not substantially different from what one might anticipate with irradiation alone. An interesting question not clearly addressed by the VA Laryngeal Cancer Study is whether induction chemotherapy accurately selects patients, based on a response to chemotherapy, who are more likely to be locally controlled by irradiation. If so, this would subject fewer patients to salvage laryngectomy, which may be associated with significant complications in one-third of the patients operated, and would reduce the overall morbidity of the treatment compared to irradiation alone.

731

MENDENHALLc’t al.

CONCLUSIONS In conclusion, our data indicate that the selective application of irradiation alone provides rates of local-regional control, survival, and severe complications similar to those obtained with surgery alone or combined with adjuvant irradiation. The benefit of irradiation alone is a significantly higher likelihood of voice preservation compared with surgery. One could argue that, had all of our patients been treated surgically, the overall cure rate might have been higher. Although there is no way to definitively counter this hypothesis, comparison of our data with the literature indicates that this is unlikely and that any survival advantage obtained by treating all patients surgically would be small. Irradiation alone is suitable for at least half of patients presenting with T3 glottic carcinoma. The patient who is ideal for this form of treatment has a lesion involving one cord, has an adequate airway, and is willing to have close follow-up after completion of treatment. Currently, all patients treated with irradiation alone at the University of Florida are treated twice daily, in a continuous course, at 120 cGy per fraction to total doses of 7440-7680 cGy with “Co. If the neck is clinically negative, the internal jugular nodes are electively irradiated. After completion of irradiation, patients have follow-up examinations monthly for 1 year, every 2 months for the second year, every 3 months for the third year, every 6 months for the fourth and fifth years, and annually thereafter. Follow-up after irradiation alone for T3 vocal cord cancer may be difficult because of the frequent occurrence of persistent laryngeal edema, distortion of the larynx by the tumor, and/or vocal cord fixation. It may be necessary at times to recommend total laryngectomy without histologic evidence of recurrent disease based on the clinical impression that a recurrence is present.

REFERENCES 1. Aanesen, J. P.; Winther, F. 0.; Tausjo, J. Treatment of T3 laryngeal carcinoma: clinical evaluation after 50 Gy, is it any value? Acta Otolaryngol. 449: 141-144; 1988. 2. Amdur, R. J.; Parsons, J. T.; Mendenhall, W. M.; Million, R. R.; Cassisi, N. J. Split-course versus continuous-course irradiation in the postoperative setting for squamous cell carcinoma of the head and neck. Int. J. Radiat. Oncol. Biol. Phys. 17: 279-285; 1989. 3. Amdur, R. J.; Parsons, J. T.; Mendenhall, W. M.; Million, R. R.; Stringer, S. P.; Cassisi, N. J. Postoperative irradiation for squamous cell carcinoma of the head and neck: an analysis of treatment results and complications. Int. J. Radiat. Oncol. Biol. Phys. 16: 25-36; 1989. 4. American Joint Committee on Cancer. Manual for Staging of Cancer, 2nd edition. Philadelphia: Lippincott; 1983: 3743. 5. Biller, H. F.; Lawson, W. Partial laryngectomy for vocal cord cancer with marked limitation or fixation of the vocal cord. Laryngoscope 96: 6 l-64; 1986. 6. Department of Veterans Affairs Laryngeal Cancer Study Group. Induction chemotherapy plus radiation compared

7. 8.

9.

10.

11.

12.

with surgery plus radiation in patients with advanced laryngeal cancer. N. Engl. J. Med. 324: 168%. 1690, 199 1. DeSanto, L. W. T3 glottic cancer: options and consequences of the options. Laryngoscope 94: 13 1 l- 13 15; 1984. Fletcher, G. H.; Lindberg, R. D.; Jesse, R. H. Radiation therapy for cancer of the larynx and pyriform sinus. EENT Digest 3 1: 58-67; 1969. Foote, R. L.; Buskirk, S. J.; Stanley, R. J.; Grambsch, P. M.; Olsen, K. D.; DeSanto, L. W.; Earle, J. D.; Weiland, L. H. Patterns of failure after total laryngectomy for glottic carcinoma. Cancer 64: 143- 149; 1989. Harwood, A. R.; Beale, F. A.; Cummings, B. J.; Hawkins, N. V.; Keane, T. J.; Rider, W. D. T3 glottic cancer: an analysis of dose time-volume factors. Int. J. Radiat. Oncol. Biol. Phys. 6: 675-680; 1980. Harwood, A. R.; Beale, F. A.; Cummings, B. J.; Keane, T. J.; Payne, D.; Rider, W. D. T4NoMo glottic cancer: an analysis of dose-time-volume fractions. Int. J. Radiat. Oncol. Biol. Phys.: 7: 1507-1512, 1981. Horiot, J. C. The EORTC radiotherapy group experience: phase III trials in hyperfractionation (HF) and accelerated

732

13. 14.

15.

16. 17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

I. J. Radiation Oncology 0 Biology 0 Physics fractionation (AF) in head and neck cancers. (Abstract) Int. J. Radiat. Oncol. Biol. Phys. 21 (Suppl. I): 108, 1991. Kalbfleisch, J. D.; Prentice, R. L. The statistical analysis of failure time data. New York: Wiley; 1980. Kaplan, E. L.; Meier, P. Nonparametric estimation from incomplete observations. J. Am. Stat. Assoc. 53: 457-481: 1958. Kessler, D. J.; Trapp, T. K.; Calcaterra, T. C. The treatment of T3 glottic carcinoma with vertical partial laryngectomy. Arch. Otolaryngol. Head Neck Surg. 113: 1196- I 199; 1987. Lawless, J. E. Statistical models and methods for lifetime data. New York: Wiley; 1982: 420-422. Lindstrom, M. J.; Fowler, J. F. Re-analysis of the time factor in local control by radiotherapy of T3-T4 squamous cell carcinoma of the larynx. Int. J. Radiat. Oncol. Biol. Phys. 21: 813-817; 1991. Lundgren, J. A.; Gilbert. R. W.; van Nostrand, A. W. P.: Harwood, A. R.; Keane, T. J.; Briant, T. D. R. T3NOMO glottic carcinoma-a failure analysis. Clin. Otolaryngol. 13: 455-465; 1988. Mendenhall, W. M.; Million, R. R.; Sharkey, D. E.; Cassisi, N. J. Stage T3 squamous cell carcinoma ofthe glottic larynx treated with surgery and/or radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 10: 357-363; 1984. Mendenhall, W. M.; Parsons, J. T.; Stringer, S. P.; Cassisi. N. J.; Million, R. R. Carcinoma of the supraglottic larynx: a basis for comparing the results of radiotherapy and surgery. Head Neck 12: 204-209: 1990. Mendenhall, W. M.; Parsons, J. T.; Stringer, S. P.; Cassisi. N. J.; Million, R. R. The role of radiation therapy in laryngeal cancer. Cancer J. Clin. 40: 150-165; 1990. Meredith, A. P. D’E.; Randall, C. J.; Shaw, H. J. Advanced laryngeal cancer: a management perspective. J. Laryngol. Otol. 101: 1046-1054; 1987. Mills, E. E. D. Early glottic carcinoma: factors affecting radiation failure, results of treatment and sequelae. Int. J. Radiat. Oncol. Biol. Phys. 5: 81 l-817; 1979. Parsons, J. T.; Bova, F. J.; Million, R. R. A re-evaluation of split-course technique for squamous cell carcinoma of the head and neck. Int. J. Radiat. Oncol. Biol. Phys. 6: 16451652; 1980. Parsons, J. T.; McCarty, P. J.; Rao, P. V.: Mendenhall. W. M.; Million, R. R. On the definition of local control. Int. J. Radiat. Oncol. Biol. Phys. 18: 705-706; 1990. Parsons, J. T.; Mendenhall, W. M.; Cassisi. N. J.; Isaacs, J. H. Jr.; Million, R. R. Hyperfractionation for head and neck cancer. Int. J. Radiat. Oncol. Biol. Phys. 14: 649-658; 1988. Parsons, J. T.; Mendenhall, W. M.; Mancuso, A. A.; Cassisi. N. J.; Stringer, S. P.; Million, R. R. Twice-a-day radiotherapy

Volume 23. Number 4. 1992

28.

29. 30.

3 I.

32. 33.

34.

35.

36.

37.

38.

39.

40.

for T3 squamous cell carcinoma of the glottic larynx. Head Neck 11: 123-128; 1989. Pfister, D. G.; Strong, E.; Harrison, L.; Haines, I. E.; Pfister, D. A.; Sessions, R.; Spiro, R.; Shah, J.; Gerold, F.; McLure, T.; Bikram, B.; Fass, D.; Armstrong, J.; Bose, F. J. Larynx preservation with combined chemotherapy and radiation therapy in advanced but resectable head and neck cancer. J. Clin. Oncol. 9: 850-859; 1991. Peters, L. J.; Ang, K. K. The role of altered fractionation in head and neck cancers. Sem. Radiat. Oncol. (In press). Razack, M. S.; Maipang, T.; Sako, K.; Bakamjian, V.; Shedd, D. P. Management of advanced glottic carcinomas. Am. J. Surg. 158: 318-320; 1989. SAS Institute Inc. SAS technical report P- 179, additional SAS/STAT procedures, release 6.03. Gary, NC: SAS Institute Inc.; 1988: 49-89. Stell, P. M.; Rawson, N. S. B. Adjuvant chemotherapy in head and neck cancer. Br. J. Cancer 61: 779-787; 1990. Stewart, J. G.; Brown, J. R.; Palmer, M. K.; Cooper, A. Panel discussion on glottic tumors. VII. The management of glottic carcinoma by primary irradiation with surgery in reserve. Laryngoscope 85: 1477-1484; 1975. Taylor, J. M. G.; Withers, J. R.; Mendenhall, W. M. Dosetime considerations of head and neck squamous cell carcinomas treated with irradiation. Radiother. Oncol. 17: 95102; 1990. Terhaard, C. H. J.; Karim, A. B. M. F.; Hoogenraad, W. J.: Tjho-Heslinga, R.; Keus, R. B.; Mehta, D. M.; Reichgelt, B. A.: Hordijk, G. J. Local control in T3 laryngeal cancer treated with radical radiotherapy, time-dose relationship: the concept of nominal standard dose and linear quadratic model. Int. J. Radiat. Oncol. Biol. Phys. 20: 12071214; 1991. Tucker, H. M.; Benninger, M. S.; Roberts, J. K.; Wood, B. G.: Levine, H. L. Near-total laryngectomy with epiglottic reconstruction. Long-term results. Arch. Otolaryngol. Head Neck Surg. 115: 1341-1344; 1989. Wang. C. C. Factors influencing the success of radiation therapy for T2 and T3 glottic carcinomas. Am. J. Clin. Oncol. 9: 5 17-520; 1986. Wang, Z.; Million, R. R.; Mendenhall, W. M.; Parsons, J. T.; Cassisi, N. J. Treatment with preoperative irradiation and surgery of squamous cell carcinoma of the head and neck. Cancer 64: 32-38; 1989. Wetmore, S. J.; Krueger, K.; Wesson, K.; Blessing, M. C. Long-term results of the Blom-Singer speech rehabilitation procedure. Arch. Otolaryngol. 111: 106-109; 1985. Yuen, A.; Medina, J. E.; Goepfert, H.; Fletcher, G. H. Management of stage T3 and T4 glottic carcinomas. Am. J. Surg. 148: 467-472; 1984.