Audit in the management of T3 fixed-cord laryngeal cancer

Audit in the Management Laryngeal Cancer Martin J. Porter, FRCS, Nicholas I? M&or, and Andrew C. Hindley, FRACR

of T3 Fixed-Cord

FRACS, Randall l? Morton, FRACS,

Purpose: To determine results of various treatments for T3 fixed-cord lesions and the subset T3 glottic cancer in Auckland from 1979 to 1995. Patients and Methods: Data were collected retrospectively from a departmental database, and the notes were reviewed. Because of the difficulty in determining the subsite of some fixed-cord lesions, the entire group of T3 fixed-cord lesions was examined, and those tumors that were considered to be definitely arising from the glottis were then analyzed as a specific subset. Results: Fixed-cord lesions were diagnosed in 75 patients (21 supraglottic, 54 glottic). Primary surgery (total laryngectomy) was performed on 46 patients, primary radical dose radiotherapy was undertaken on 25 patients, and four patients were treated palliatively. For T3 fixed-cord lesions, disease-specific survival for radiotherapy and surgery was 36% and 66%, respectively, and 32% and 67%, respectively, for T3 glottic lesions. For both T3 fixed-cord and T3 glottic lesions, surgery produced significantly better survival than did radiotherapy (~60 Gy; P= .0157). With radiotherapy greater than 60 Gy, cancer of the larynx has been controlled in seven of 13 patients, although only five patients are alive, with a median follow-up of 24 months (range, 12-49 months). Conclusion: Radiotherapy less than 60 Gy produced markedly inferior results to surgery for T3 fixed-cord lesions and T3 glottis in Auckland. Radiotherapy at more than 60 Gy shows promise, but an ongoing audit is essential to ensure that survival is similar to surgery and to that reported by those promoting organ-preservation protocols. Copyright 0 1998 by W.B. Saunders Company.

The treatment of T3 laryngeal cancer is controversial, and there is considerable variation in reported treatment efficacy.lm7 This disparity of opinion may be due, in part, to variable interpretation of the subsite origin of a fixed-cord lesion (eg, supraglottic, glottic, subglottic, transglottic). In Europe and Canada, there is a tendency to treat T3 glottic lesions with radiotherapy, whereas primary surgery is used in most medical centers in the United States. However, with initial reports of a Veteran Affairs study8 in which survival with

From the Department of Otolaryngology/Head and Neck Surgery, Green Lane Hospital, Auckland, New Zealand, and the Department of Oncology, Auckland Hospital, Auckland, New Zealand. Presented at the 4th International Conference on Head and Neck Cancer, Toronto, Canada, August 1, 1996. Address reprint requests to Nicholas P. Mclvor, FRACS, Department of Head and Neck Surgery, Green Lane Hosoital. Green Lane West, Auckland, New Zealand. Copyright 0 1998 by W.B. Saunders Company 0196-0709/98/l 903-0003$8.00/O

360

American

Journal

of Otolaryngology,

chemo-radiation was reported as not significantly worse than that achieved after surgery and radiotherapy, one might expect a growing trend toward organ-preservation protocols. Before adopting new protocols, it is important for every center, regardless of treatment philosophy, to audit and report its results to ensure that the local expertise compares appropriately with the results reported by major centers in the medical literature. Traditionally, T3 fixed-cord lesions have been treated with primary surgery (total laryngectomy and primary tracheoesophageal puncture). Until 1992, radiation dosages used in Auckland, New Zealand to treat patients with T3 laryngeal cancer were lower than those generally recommended elsewhere, but in 1992 a new schedule of high-dose radiotherapy was instituted with the expectation of improved outcome. This article reviews our experience with surgery and these varying radiotherapy regimens in the treatment of T3 fixed-cord lesions between 1979 and 19%.

Vol 19, No 6 (November-December),

1998:

pp 360-364

MANAGEMENT

OF LARYNGEAL

361

CANCER

PATIENTS AND METHODS Since 1990, all patients presenting with laryngeal cancer at the Auckland Head and Neck Service based at Green Lane Hospital have been entered prospectively on a customized database. Before that date, patients had been entered retrospectively to 1979. For this study, patients who had been categorized as having T3 squamous cell carcinoma of the larynx were extracted. Patient notes were then reviewed to confirm that they fulfilled the diagnosis of T3 laryngeal cancer staged according to the American Joint Committee for Cancer Staging and End Result Reporting.g Dates of cancer recurrence, death, and last follow-up were noted. When follow-up data were not available from hospital records, attempts were made to contact the patient’s general practitioner and/or relatives. When patients with known recurrence were lost to follow up, they were deemed to have been dead from the date of last contact. Patients with supraglottic, glottic, or transglottic lesions that did not clinically or radiologically extend into laryngeal cartilage or beyond the larynx and caused immobility of the vocal cords were considered to be in the T3 fixed-cord group. Glottic and transglottic lesions were considered to be in the T3 glottic subset. There were no T3 subglottic lesions. Surgery in all cases comprised total laryngectomy with primary tracheoesophageal puncture in both the primary and the salvage situation. The radiotherapy dosage regimen for each patient was reviewed and the equivalent tumor dose was calculated. Until 1992, all but two patients receiving primary radiotherapy were treated with less than the 60 Gy equivalent. From 1992, dosages have been greater than 60 Gy. Therefore, these patients were divided into two groups according to dosage: less than or equal to 60 Gy and greater than 60 Gy. Patients who underwent postoperative radiotherapy received it on the basis of histologic features in the laryngectomy specimen and were classified as surgical patients. Parastoma1 or pharyngeal recurrence after laryngectomy was deemed to represent primary recurrence. Survival and adjusted disease-specific sur-

percent survival

30

0 Surgery q Radiotherapy

20

0

1

30 P=O.O089

2

20

3

4

’ 0 5

years Fig 1. Disease-specific survival with T3 fixed-cord laryngeal cancer (total laryngectomy) and radiotherapy.

rates treated

for patients by surgery

viva1 was calculated by using the KaplanMeier method. The chi-square method was used to test significance. RESULTS There were 75 patients with T3 fixed-cord lesions. Of these patients, 63 were men and 12 were women, with a mean age of 64 years for both. Twenty-one fixed-cord lesions were deemed to have arisen in the supraglottic larynx. The remaining 54 had been classified as glottic tumors. Primary surgery was employed in 46 patients, whereas 25 patients were treated with primary radiotherapy (including two patients with synchronous lesions). Four patients received only supportive palliative treatment. Surgical patients had a minimum follow-up time of 2 years with a

percent survival

30

0 Surgery q Radiotherapy

20

0

1

P=O.O077

2

3

4

5

years Fig 2. Disease-specific with T3 glottic cancer treated tomy) and radiotherapy.

survival rates for patients by surgery (total laryngec-

PORTER

median follow-up of 7 years. Patients receiving radiotherapy less than 60 Gy and greater than 60 Gy had a minimum follow-up time of 60 months and 12 months, respectively (median follow-up, 24 months). The disease-specific, 5-year actuarial survival curves for patients with T3 fixed-cord lesions and patients with the subset of T3 glottic lesions are presented in Figs 1 and 2. The actuarial survival rate for patients with T3 fixed-cord lesions that were treated with radiotherapy and surgery was 3.2% and 57%, respectively (P < .005). Disease-specific survival was 36% and 66%, respectively (P < .Ol). The 5-year survival rate for patients with the subset of T3 glottic lesions treated with radiotherapy was Z8%, the rate for treatment with surgery was 64%. The disease-specific survival rate for this group was 32% and 67%, respectively (P < 0.01). The early results for radiotherapy greater than 60 Gy are given in Table 1, but a conclusion has yet to be reached because of short follow-up times. The pattern of failure according to primary site, neck, and systemic metastases is shown in Figs 3 and 4. Surgical salvage after radiotherapy failure was successful in 50% of those attempted (4 out of 8 patients) but only 27% successful overall (4 out of 15 patients] (Table 2).

ET AL

NECK

PRIMARY

METASTASES Fig 3. Site of failure therapy for T3 fixed-cord

for cancer lesions

treatment by radiation (25 patients).

supraglottic lesions extending to the medial wall of the pyriform fossa and/or the preepiglottic space. The figures on surgery of 57% actuarial and 67% disease-specific, 5-year PRIMARY

NECK

DISCUSSION The survival curves for each treatment modality (surgery or radiotherapy) were similar for both the composite group of T3 fixed-cord lesions and the subgroup of T3 glottic tumors. Therefore, we feel it is valid to compare the entire group of fixed-cord lesions with the results reported on treatment for T3 glottic cancer from other medical centers. Other T3 laryngeal lesions without vocal cord fixation have been excluded, because they incorporate TABLE 1. Fixed-Cord No. of Patients 13

Treatment Lesions Dead of Disease 5*

Outcome for Patients Treated by Radiotherapy Dead (Disease-Free) 3

With T3 (>60 Gy)

Alive 5

METASTASES Fig 4. Site of failure for T3 fixed-cord lesions TABLE 2. Fixed-Cord

Follow-up 12-49

mo

‘Two patients had synchronous tumors. Loco-regional control was not achieved with radiotherapy; surgical salvage was not attempted because of uncontrollable second primaries.

by surgery

Fate of the Larynx in Patients With T3 Lesions Treated With Radiotherapy

Dosage

No. of Patients

<60 Gy >60 Gy All patients

12 13 25

*Includes

for cancer treatment (46 patients).

one patient

Lesion Recurrence

Salvage Laryngectomy

9 6* 15 with a synchronous

3 5 8

Successful Salvages 2 2 4

lesion.

MANAGEMENT

TABLE 3.

OF LARYNGEAL

Reported

Patient

363

CANCER

Survival

for T3 Glottic

Cancer

Institution

Author

Series

Harwood et al4 DeSanto6 Yuen et al7

Princess Marqaret Hospital, Toronto, Mayo Clinic, USA MD Anderson Cancer Center, USA

Jones

Royal

et al1

Mendenhall

et al2

Meredith et al5 Potter et al* (T3 Fixed-Cord)

Liverpool

University Royal Green

Abbreviation: RT, radiation *Current study.

Hospital,

of Florida,

Canada

UK

USA

Marsden Hospital, UK Lane Hospital, New Zealand

RT Surgery Surgery Surgery Surgery RT Surgery RT RT Surgery RT

5year Survival Actuarial (%) 55

and RT

? RT

29 37 45 58 53 57 32

5year Survival Adjusted (%) 74 78 80 89 48 70 63 74 74 66 36

therapy.

survival are comparable with those generally found in the literature, given differences not only in the selection criteria but also in the method used for reporting survival.*J In seeking to compare these results with other published works, one should be aware that the limitations inherent in a retrospective study such as this must be recognized. The selection criteria for each of the three treatment regimens cannot be defined accurately for individual patients. Staging was performed clinically by a number of surgeons and computed tomography scans were employed only in the past 10 years. After histologic examination, some patients were upstaged, but remain in the analysis. Thus, if there is any bias, it will act to produce inferior results to those patients who were staged with more recent methods. The inferior survival rate for the patients in our study who were treated with primary radiotherapy seems to be related to poor local control, coupled with an inability to successfully salvage sufficient patients surgically (4 out of 15 patients). Most other groups have found survival after radiotherapy to approximate survival after surgery.2-5 Half of the eight patients (50%) undergoing attempted surgical salvage achieved tumor control (Table 3). This finding is comparable with results reported from other countries3 and shows the value of salvage surgery. Unfortunately, only about 55% of the patients who developed cancer recurrence after radiotherapy went on to have salvage surgery. This is a reflection of attitudes (Maori and Pacific Islander patients, to whom the head is traditionally sacred, often refused

primary and salvage laryngectomy) and poor general condition. However, the results confirm our clinical impression at that time (prior to 1992) and vindicate the former treatment policy of primary surgery. Indeed, it would have been unethical for us to continue to offer that primary radiotherapy regimen as a reasonable alternative to surgery. Although the results for glottic cancer in patients irradiated in Auckland have improved from 1965 to 1979,1° they remain inferior to the local control rates of 60% to 70% reported from a number of other medical centers (Table 3). Our previous radiotherapy results also yielded local control rates in Tl and T2 glottic carcinomas of only 70% and 40%) respectively,lO which are again inferior to many rates quoted in the literature. Further analysis suggests that the major difference has been the relative biologic ineffectiveness of the dose fractionation regimen employed. In the present series, the most commonly used schedule of radiotherapy before 1992 was 54 Gy in 20 fractions given for 42 to 46 days, which is considerably less effective than most regimens used to treat Tl cancer. Most authorities recognize that an even higher dose is necessary to adequately treat T3 tumors with greater numbers of clonogens.2s4 Patients who have been treated surgically in Auckland have received a very acceptable chance of 5year survival and have also achieved a high degree of life satisfaction.ll They are fitted with Blom-Singer valves (In Health, Santa Barbara, CA) after primary tracheoesophageal puncture and have a high rate of speech proficiency. I2 Therefore, it has been

364

with some reservation that we have embarked on an organ-preservation policy when life preservation with excellent life satisfaction might seem more crucial. However, we have taken into account the increasing literature showing that primary high-dose radiotherapy with surgical salvage is able to produce comparable survival to primary surgery. Since 1992, increasing the biologically effective dose of radiation by giving 70 Gy in 35 fractions for 7 weeks has improved loco-regional control rates, although a survival advantage is yet to be seen (Table 1). It has been due to continuing audit, the appointment of an enthusiastic radiation oncologist, and the recognition that dosages were inadequate that the newer regimen has been adopted. Our current policy is to recommend primary radiotherapy for clinically and radiologically staged T3 cancer of the glottic larynx, provided certain criteria are met. These criteria include the ability to regularly follow-up patients and the ability to commence radiotherapy within 6 weeks (the current waiting time for “nonurgent” radiotherapy in Auckland is more than 8 weeks). Patients who are unable to comply with stringent follow-up or default on radiotherapy are treated surgically. We have continued to treat patients with T3 supraglottic fixed-cord lesions with surgery, but these results suggest that we could treat this group in the same way as those patients treated for T3 glottis. Therefore, we will continue to audit the results of treatment for

PORTER

ET AL

laryngeal cancer, and only if the current radiotherapy regimen produces results comparable to surgery will it be reasonable to extend the treatment to all patients with T3 fixed-cord laryngeal cancer. REFERENCES 1. Jones AS, Cook JA, Phillips DE, et al: Treatment of T3 carcinoma of the larynx by surgery or radiotherapy. Clin Otolaryngol 17:433-436,199Z 2. Mendenhall WM, Parsons JT, Stringer SP, et al: Stage T3 squamous cell carcinoma of the glottic larynx: A comparison of laryngectomy and irradiation. Int J Radiat Oncol Biol Phys 23:725-732,1992 3. Sagar SM, McKenna G, Nolan MC: A clinical audit of glottic cancer in Nova Scotia: A paradigm for effectiveness research. Clin Oncol6:14-23,1994 4. Harwood AR, Beale FA, Cummings BJ, et al: T3 glottic cancer: An analysis of dose-time-volume factors. Int J Radiat Oncol Biol Phys 6:675-680,198O 5. Meredith AP, Randall CJ, Shaw HJ: Advanced laryngeal cancer: A management perspective. J Laryngol Otol 101:1046-1054,1987 6. De Santo LW: T3 glottic cancer: Options and consequences of the options. Laryngoscope 94:1311-1315, 1984 7. Yuen A, Medina JE, Goepfert H, et al: Management of stage T3 and T4 glottic carcinomas. Am J Surg 148:467472,1984 8. Department of Veteran Affairs Laryngeal Study Group: Induction chemotherapy plus radiation compared with surgery plus radiation in patients with advanced laryngeal cancer. N Engl J Med 324:1685-1690,199l 9. American Joint Committee for Cancer Staging and End Results Reporting: Manual for Staging of Cancer (ed 3). Philadelphia, PA, Lippincott, 1988. - 10. Morton RP. Chaoman P: The results of treatment of laryngeal cancer.in Auckland New Zealand 1965-1979. Aust NZ J Surg 52:418-423,1982 11. Morton RP: Life-satisfaction in patients with head and neck cancer. Clin Otolaryngol20:499-503,1995 12. McIvor NP, Dorman EB, Morton RP, et al: Primary tracheoesophageal puncture for vocal rehabilitation. J Otolaryngol Sot Aust 6:295-298,199O