MANAGEMENT OF SQUAMOUS CELL CARCINOMA

OF THE FLOOR OF MOUTH

Lawrence W. Rodgers, Jr, MD, Scott P. Stringer, MD, William M. Mendenhall, MD,
James T. Parsons, MD, Nicholas J. Cassisi, DDS, MD, and Rodney R. Milllion, MD

cancers have been managed by primary surgery,

Between 1964 and 1987, 194 patients with previously untreated
squamous cell carcinoma of the floor of mouth were managed at while before that time most were treated by pri-
the University of Florida. A retrospective analysis was under- mary irradiation. This change in treatment se-
taken in order to evaluate the treatment results and associated lection was undertaken in an effort to reduce the
complication rates. Surgery or irradiation alone was found to re- risk of bone exposure andl soft tissue necrosis as-
sult in similar local control rates for stage I and II lesions, sociated with irradiation which may result in
whereas more advanced tumors had better local control rates
with a combination of surgery and irradiation. Radiotherapy had long-term d i ~ a b i l i t y .Although
~,~ the initial risk
a higher incidence of minor and moderate complications, of severe complications from surgery may be
whereas a greater number of severe complications occurred af- higher, these complications are generally of
ter surgery. We recommend surgery for early lesions due to the shorter duration. During both time periods, most
lower overall incidence of associated complications. Despite a patients with advanced primary cancers received
higher risk of severe complications, combination therapy is rec-
ommended for more advanced lesions due to improved local
planned combined irradiation and surgery; in
control as compared to single modality therapy. the majority, the surgical resection was per-
HEAD & NECK 1993;15:16-19 formed first and followed by postoperative irradi-
0 1993 John Wiley & Sons, Inc. ation. An analysis was undertaken to evaluate
treatment results and complication rates with
floor of mouth cancer therapy at the University
Treatment options in the management of floor of of Florida.
mouth carcinoma consist of radiotherapy alone,
surgery alone, or a combination of these two mo- MATERIALS AND METHODS
dalities.'-* Over the last 10 years at the Univer- This is a retrospective ainalysis of 194 patients
sity of Florida nearly all early floor of mouth managed with curative intent for previously un-
treated floor of mouth squamous cell carcinoma
at the University of Florida between October
From the Department of Otolaryngology (Drs. Rodgers, Stringer, and 1964 and December 1987. All patients were fol-
Cassisi) and Radiation Oncology (Drs. Mendenhall, Parsons, and Mil- lowed for a minimum of 2 years; 81% were fol-
lion), University of Florida College of Medicine, Gainesville, Florida.
lowed for at least 5 year!;. Cancers were staged
Presented at the 20th Annual University of Florida Department of Radia-
tion Oncology Clinical Research Seminar, April 28, 1990 according to the Americatn Joint Committee on
Address reprint requests to Dr. Stringer at P.O. Box 100264, UF Health
Cancer staging system. The system was modified
Science Center, Gainesville, FL 32610-0264. such that stage IV was divided into a favorable
Accepted for publication July 3, 1992 subset IVa (Tl-T3, N2A--N3A) and a less favor-
CCC 0 148-64031931010 16 -04
able subset IVb (T4 and/or N3B).7
0 1993 John Wiley & Sons, Inc All patients were included in the analysis of

16 Floor of Mouth Carcinoma HEAD & NECK January/February 1993

survival and complications. Patients were ex- therapy, including hyperbaric oxygen, or surgi-
cluded from analysis of control of disease at the cal wound breakdown requiring outpatient ther-
primary site and/or neck if they died within 2 apy only, such as wound packing with secondary
years of treatment with that site($ continuously healing. Severe complications included postoper-
disease-free. Successful salvage of a local or neck ative problems necessitating prolonged hospital-
failure was defined as continuous disease-free ization (eg, myocardial infarction or pulmonary
status at the site in question for at least 1year. embolism), severe infection, fistula formation, or
The various radiotherapy techniques used for osteoradionecrosis requiring hospitalization or
floor of mouth carcinoma at the University of surgical intervention.
Florida have been previously r e p ~ r t e d . ~In.~,~~~
general, implant alone was used for T1 tumors RESULTS
where the risk of disease in the neck was The initial and ultimate primary control rates by
thought to be less than 20%. External beam and T stage are presented in Table 1. T1 and T2 le-
interstitial implant were used for the majority of sions have similar control rates with irradiation
patients in order to electively irradiate the neck or surgery alone. The T3 and T4 lesions appear
in conjunction with the primary lesion or in pa- to have improved local control rates with com-
tients with clinically positive nodes. External bined therapy. Of those patients whose initial
beam alone was used in a minority of patients; therapy was surgery alone, one of two patients
generally those with tumors too extensive to ad- with T2 lesions were successfully salvaged, and
equately implant and in those patients with one patient with a T4 lesion underwent an un-
medical conditions prohibiting general anesthe- successful salvage procedure. Twenty-one pa-
sia. An additional subset of patients were those tients underwent salvage attempts after irradia-
with early lesions suitable for intraoral cone ir- tion alone failed: three of four patients with T1
radiation where the intraoral cone portion of the lesions were successfully salvaged; six of 10 pa-
irradiation replaced the implant. This was used tients with T2 lesions were salvaged; two of six
only for T1 and early T2 cancers that could be patients with T3 lesions underwent successful
encompassed adequately in the cone. salvage attempts; and attempted salvage of one
Treatment modalities for the primary site T4 lesion was unsuccessful.
were distributed as follows: 117 received irradia- Table 2 outlines the rates of initial and ulti-
tion alone, 36 underwent surgery alone, and 41 mate control above the clavicles (primary and
required combination therapy (surgery and irra- neck control) by stage and by treatment modal-
diation). Ten patients receiving combined ther- ity to the primary site. The same pattern
apy were treated preoperatively, and 31 patients emerges as with local control. Early lesions have
underwent postoperative radiotherapy. Of those similar control rates with irradiation or surgery
receiving irradiation, 62 received CO,, alone, 64
were treated with CO,, and radium implant, 20
received radium implant alone, and 12 under- ~ ~ ~

Table 1. Initial and ultimate local control

went therapy with orthovoltage X-ray via in-
traoral cone with or without GO,,. Twenty-four Stage RT Surgery Surgery and RT
patients received radiotherapy using the split- ~

T1
course technique. This method is no longer used Initial 32/37 (86%) 9/10 (9OYo) 111 (100%)
at the University of Florida because results were Ultimate 35/37 (94%) 9/10 (9OY0) 111 (100%)
noted to be inferior to continuous-course irradia- T2
tion.l0,l1 Management of the neck was individu- Initial 25/36 (69%) 9/12 (75%) 7/7 (100%)
alized and included: no treatment, irradiation Ultimate 31/36 (86%) 10/12 (83%) 7/7 (100%)
T3
alone, neck dissection alone, and neck dissection Initial 11/20 (55%) t 9/9 (100%)
in combination with irradiation either pre- or Ultimate 13/20 (65%) t 9/9 (100%)
postoperatively. T4
Complications were scored as mild, moderate, Initial 2/5 (40%) 112 (50%) 518 (63%)
or severe. Mild complications included bone ex- Ultimate 2/5 (40%) 112 (50%) 5/8 (63%)
posure less than 1.5 cm, soft tissue necrosis RT, radiotherapy
requiring outpatient or no therapy, and minor *Grouped by initial treatment to the primary site Forty-seven patrents
were excluded from local control analysis because they died within 2
infections. The moderate complications includ- years of treatment with the primary site continuously disease-free
ed osteoradionecrosis requiring only outpatient fNo patients in category

Floor of Mouth Carcinoma HEAD & NECK JanuaryIFebruary 1993 17

 Table 2. Initial and u,tirnate control above clavicles.' rate when surgical therapy was used with or
without radiotherapy. The severe complications
Stage RT Surgery Surgery and RT
following combination therapy included: three
I infected titanium trays requiring removal, one
Initial 28/35 (80%) 8/10 (80%) 0/1 (0%) orocutaneous fistula, one bone exposure requir-
Ultimate 33/35 (94%) 9\10 (90%) 0/1 (0%) ing debridement, and one prolonged hospitaliza-
II
tion secondary to adult respiratory distress syn-
Initial 17/26 (65%) 5/7 (71%) 415 (80%)
Ultimate 20126 (77%) 617 (86%) 415 (80%) drome. In the surgery-alone group the severe
111 complications included: three postoperative
Initial 14/23 (61%) 213 (67%) 8/10 (80%) deaths due to medical complications, two orocu-
U timate 18/23 (78%) 313 (100%) 8/10 (80%) taneous fistulas, and one postoperative hemor-
IVA
rhage.
Initial 3/11 (27%) 011 (0%) 314 (75%)
Ultimate 5/11 (45%) 011 (0%) 314 (75%) Analysis of the severe complications in the
IVB subset consisting of stage I or I1 lesions only re-
Initial 216 (33%) 0/4 (0%) 519 (56%) veals that irradiation alone was associated with
U timate 2/6 (33%) 014 (0%) 519 (56%) a severe complication in four of 68 (6%) cases,
RT. radiotherapy whereas three of 21 (14%) lesions treated with
'Grouped by initial treatment /o the primary site Management ot the surgery alone developed severe complications.
neck varied and ncluded no treatment, irradiation neck dissectiofi or
neck dissection and pre- of posfoperafive irradiation Thirty-nine patients
All of the irradiation severe complications were
were excluded from local- regional controi analysis because they died osteoradionecrosis of the mandible, whereas the
wifhin 2 years of 'reatmenf with /he primary site and neck continuodsly surgery-alone group included a postoperative he-
disease-lrec
matoma, an orocutaneous fistula, and a postoper-
ative death due t o cardiopulmonary arrest.
alone to the primary site, and more advanced le-
sions fare better with combined treatment. DISCUSSION
The complication rate by treatment modality When evaluating local and local- regional con-
to the primary site is presented in Table 3. Note trol, difficulties arise in making comparisons
the relatively high rate of mild to moderate com- among institutions due to a lack of uniformity in
plications in the irradiation-alone group. The vast the way the data are analyzed. For instance,
majority of these included soft tissue necrosis or some calculate control rates by stage with all
small bone exposures. Five patients in this sub- treatment modalities combined. Others analyze
group developed mild osteoradionecrosis with control rates by different treatment methods but
two resolving spontaneously and one having per- with all stages combined, or describe lesions as
sistent necrosis but refusing intervention. The early or advanced without stage delineation. Nev-
remaining two patients were treated with hyper- ertheless, control rates at the University of Flor-
baric oxygen and local wound care. The irradia- ida appear to be similar to those represented in
tion-alone group included six patients in the se- the l i t e r a t ~ r e . ~Further
"~ analysis of the litera-
vere complication group due to the need for ture on floor of mouth cancer reveals 5-year sur-
operative debridement or debridement with flap vival rates ranging from 64%-88% for stage I,
coverage to treat bone exposure or severe osteora- 61%- 84% for stage II,28%- 68% for stage 111, and
dionecrosis. 6?&36% for stage IV.1,3,4*'3 These data are com-
Severe complications occurred a t a higher parable to the data presented herein (Table 4).
Initial and ultimate local control rates for ir-
radiation alone or surgery alone for early lesions
did not differ significantly. There was a slight
Table 3. Complications.
improvement in local control in the few patients
Number of Number of patients with early disease managed with combined ther-
lnilial patients with with mild lo apy. The increased morbidity and cost associated
treatment to severe moderate
primary site complications complications
with combined therapy for T1 and T2 lesions
does not seem to be justified, especially because
RT alone 6/117 (5%) 49/117 (42%) this advantage disappears when evaluating con-
Surgery alone 6/36 (1 7%) 3/36 (8%) trol rates above the clavicle. However, combined
Surgery and RT 6/41 (15%) 8/41 (20%)
therapy for T3 and T4 lesions was clearly associ-
RT, radiotherapy. ated with a significant improvement in local con-

18 Floor of Mouth Carcinoma HEAD & NECK JanuaryiFebruary 1993

 ~

Table 4.5-year cause-specific survival ' However, radiotherapy was associated with a
higher overall incidence of complications. There-
Stage RT alone Surgery alone Surgery and RT fore, we recommend surgery for the management
I 22/23 (96%) 416 (67%) t of early lesions. Improved initial local- regional
I1 14/20 (70%) 516 (83%) t control rates were noted with combined treat-
111 14121 (67%) 213 (67%) 518 (63%) ment of advanced lesions. Therefore, despite the
IVA 419 (44%) Oil (0%) 1/3 (33%) higher incidence of severe complications associ-
IVB 1/5 (20%) 114 (25%) 218 (25%)
ated with surgery for these lesions, we recom-
RT, radiotherapy mend combination therapy for stage I11 and IV
'Grouped by initial treatment to the primary site Management of the
neck varied and included no treatment, irradialion, neck dissection, or
cancers.
neck dissection and pie- and postoperative irradiation
jnlo patients in category

REFERENCES
1. Panje WR, Smith B, McCabe BF. Epidermoid carcinoma
trol rates compared to single modality therapy of the floor of mouth: surgical therapy vs combined ther-
even after salvage attempts. apy vs radiation therapy. Otolaryngol Head Neck Surg
1980;88:714-720.
As with local control, initial and ultimate 2. Applebaum EL, Callins WP, Bytell DE. Carcinoma of the
control above the clavicles for stage I and I1 tu- floor of the mouth. Arch Otolaryngol 1980;106:419-421.
mors was similar for radiotherapy or surgery 3. Ildstad ST, Bigelow ME, Remensnyder JP. Intra-oral
cancer a t the Massachusetts General Hospital.
alone to the primary site. Initial local-regional Squamous cell carcinoma of the floor of the mouth. Ann
control rates for stage I11 and IV lesions were Surg 1983;197:34-41.
again enhanced by combined therapy. However, 4. Lehman RH, Cox JD, Belson TP, et al. Recurrence pat-
terns by treatment modality of carcinomas of the floor of
salvage of stage I11 lesions treated with one mo- the mouth and oral tongue. A m J Otolaryngol
dality resulted in local-regional control rates 1982;3:174- 181.
similar to those for combination treatment. Due 5. Mendenhall WM, Van Cise WS, Bova FJ, Million RR.
Analysis of time-dose factors in squamous cell carci-
to variability in the management of the neck, noma of the oral tongue and floor of mouth treated with
any difference in local-regional control rates radiation therapy alone. Int J Radiat Oncol Biol Phys
must be assumed to be associated with the pri- 1981;7:1005-1011.
6. Mendenhall WM, Parsons JT, Mendenhall NM, MIllion
mary site treatment alone. Five-year cause- RR. Brachytherapy in head and neck cancer: selection
specific survival was not significantly different criteria and results a t the University of Florida. Oncol-
for advanced lesions between the combination ogy 1991;5:87-93..
7. Mendenhall WM. Parsons JT. Million RR. Cassisi NJ.
and single modality groups due to the effect of Device JW, Greene BD. A favorable subset of AJCC
distant metastases (Table 4). stage IV squamous cell carcinoma of the head and neck.
When selecting appropriate therapy espe- Znt J Radiat Oncol Biol Phys 1984;10:1841-1843.
8 Levitt SH, Khan FM, Potish RA, eds. Leuitt and Tapley's
cially when control rates are similar as with technological basis of radiation therapy: practical clinical
early lesions, the risk and severity of complica- applications. Philadelphia: Lea & Febiger, 1992:212-
tions must be considered. A relatively high risk 216.
9. Million RR, Cassisi NJ, eds. Management of head and
of minor to moderate bone exposures and/or soft neck cancer: a multidisciplinary approach. Philadelphia:
tissue necrosis was noted in patients treated J B Lippincott, 1984:250-265.
with irradiation alone. Similar complication 10. Million RR, Zimmerman RL. Evaluation of University of
Florida split course technique for various head and neck
rates have been reported p r e v i o ~ s l y . ~ ,Al-
'~ squamous cell carcinomas. Cancer 1975;35:1533- 1536.
though these complications usually do not re- 11. Parsons JT, Bova FJ, Million RR. A re-evaluation of
quire operative intervention, they are often per- split-course technique for squamous cell carcinoma of the
head and neck. Int J Radiat Oncol Biol Phys
sistent and problematic. 1980;6:1645-1652.
Severe complications were more frequently 12. Marks JE, Lee F, Smith PG, Ogura JH. Floor of mouth
associated with surgical procedures with or with- cancer: patient selection and treatment results. Laryngo-
scope 1983;93:475-480.
out irradiation; however, they tended to occur in 13. Nason RW, Sako K, Beecroft WA, Razack MS, Bakam-
more advanced lesions. Evaluation of the 12 se- jian VY, Shedd DP. Surgical management of squamous
vere complications associated with surgery re- cell carcinoma of the floor of the mouth. A m J Surg
1989;158:292-296.
vealed that only three occurred in patients with 14. Larson DL, Lindberg RD, Lane E, Goepfert H. Major
stage I or I1 disease. complications of radiotherapy in cancer of the oral cavity
In summary, early lesions (stages I and 11) and oropharynx: a 10-year retrospective study. A m J
Surg 1983;146:531-536.
were found to have similar local-regional con-
trol rates with surgery or irradiation alone.

Floor of Mouth Carcinoma HEAD & NECK January/February 1993 19