staging and grading as prognostic factors in maxillary squamous cell carcinoma

Smroatdc

SURGICAL ONCOLOGY AND RECONSTRUCTION

J Oral Maxillofac Surg69:3038-3044, 2011

Staging and Grading as PrognosticFactors in Maxillary Squamous

Cell CarcinomaPaul W. Poeschl, MD, DMD,* Guenter Russmueller, MD, DMD,†

Rudolf Seemann, MD, DMD,‡ Clemens Klug, MD, DMD, PhD,§

Ellen Poeschl, MD, DMD,� Irene Sulzbacher, MD, PhD,¶ and

Rolf Ewers, MD, DMD, PhD#

Purpose: This retrospective study was performed to present our long-term results in the treatment ofmaxillary squamous cell carcinoma and evaluate especially the influence of T staging and grading onpatients’ survival.

Patients and Methods: We performed a retrospective analysis of 93 consecutive patients withalveolar, gingival, or palatal maxillary SCC treated at our clinic with surgical resection and/orradiation therapy. Data were obtained from chart review and patients’ records and were analyzedstatistically using the log-rank test and Kaplan-Meier survival curves. The male:female ratio was 2:1and the mean age was 63 years (range 35 to 94 yrs). Most patients showed a T4 stage (66%) and themost frequent staging was T4N0M0 (42%). The most common histopathological grading was G2(57%), followed by G3 (22%) and G1 (21%). The 5-year overall survival rate was 71%, and therecurrence rate was 37%. Advanced T stage (T4) and grading did not significantly influence thecumulative survival rates.

Conclusions: T-stage and grading do not have a significant impact on patients’ long-term survival. Themost crucial factor for recurrence prevention and therefore survival are free resection margins.© 2011 American Association of Oral and Maxillofacial Surgeons
J Oral Maxillofac Surg 69:3038-3044, 2011
0

quamous cell carcinoma (SCC) is the most frequentalignant tumor in the oropharyngeal region and

epresents more than 90% of all malignancies in theral cavity. The most common locations for oral SCCre the tongue and the floor of the mouth, whereashe maxillary region is affected quite rarely by theisease. In Germany oral SCC is still the sixth mostommon carcinoma in males.1 It is typical of maxillary

SCC that most of the patients are looking for treat-

*Senior Consultant, University Hospital for Cranio, Maxillofacial

and Oral Surgery, Medical University of Vienna, Vienna, Austria.

†Resident, University Hospital for Cranio, Maxillofacial and Oral

Surgery, Medical University of Vienna, Vienna, Austria.

‡Resident, University Hospital for Cranio, Maxillofacial and Oral

Surgery, Medical University of Vienna, Vienna, Austria.

§Associate Professor and Vice Chairman, University Hospital for

Cranio, Maxillofacial and Oral Surgery, Medical University of Vi-

enna, Vienna, Austria.

�Resident, Institute of Oral and Maxillofacial Surgery and Den-

tistry, Krankenhaus Hietzing, Vienna, Austria. d

3038

ment when the tumor has already reached an ad-vanced stage (ie, T4 tumor stage). This is mainlybecause the tumor can grow within the maxillary orother paranasal sinuses in a quite unhindered andunperceived way (Fig 1). Considering this back-ground, the treatment of maxillary SCC today stillposes a challenge for the head and neck surgeon. Fortreatment planning, the TNM staging and grading aremostly seen as very important factors in terms of

¶Associate Professor, Clinical Institute for Pathology, Medical

University of Vienna, Vienna, Austria.

#Professor and Chairman, University Hospital for Cranio, Maxillofa-

cial and Oral Surgery, Medical University of Vienna, Vienna, Austria.

Address correspondence and reprint requests to Dr Poeschl:

University Hospital for Cranio, Maxillofacial and Oral Surgery, Med-

ical University Vienna, Währinger Gürtel 18-20, A-1090 Vienna,

Austria; e-mail: [email protected]

© 2011 American Association of Oral and Maxillofacial Surgeons

278-2391/11/6912-0034$36.00/0

oi:10.1016/j.joms.2011.02.064

mailto:[email protected]

l

aw

Poeschl et al. Staging and Grading as Prognostic Factors inMaxillary SCC. J Oral Maxillofac Surg 2011.

POESCHL ET AL 3039

poorer or better prognosis for the patient and seem tohave a strong influence on the decision process. Atthe University Hospital for Cranio-, Maxillofacial-, andOral Surgery of the Medical University of Vienna thetherapeutic strategy is usually determined on an indi-vidual basis for each patient. In the present study wewant to present our long-term results in the treatmentof alveolar, gingival, and palatal maxillary SCC as asingle academic institution’s experience and espe-cially describe the influence of T staging and gradingon survival.

Patients and Methods

A total of 93 patients suffering from alveolar, gingi-val, or palatal maxillary SCC were treated at our clinicbetween 1992 and 2007. Data were obtained fromchart review and compiled database of Vienna Gen-eral Hospital, comprising all necessary documentsand protocols. Approval was obtained from the localethics committee. The parameters included were tu-mor location, TNM staging, grading, age at the time ofdiagnosis, gender, mode of primary treatment, statusof resection margins, and eventual tumor recurrence,including its treatment protocol. Statistical analysiswas performed with the SPSS for Windows, version15 (SPSS Inc, Chicago, IL).

Long-term survival was analyzed by Kaplan-Meiersurvival curves and a P value less than .05 in theog-rank test was considered statistically significant.

TREATMENT PROTOCOL

TNM staging was determined according to the lat-est Union Internationale Contre le Cancer classifica-tion2 before any treatment. A biopsy of the suspicioustissue was taken in every case to confirm the clinicalpresumptive diagnosis and the grading was then de-termined by histopathological examination. When alldiagnostic data had been obtained, the treatment pro-tocol was then discussed with the patient. Wheneverthe SCC was resectable (ie, no intracranial invasion,no invasion into other organs indispensable to life[�T4a]), and the patient’s general condition allowedfor the surgical procedure, resection as primary treat-ment, including neck dissection according to the Nstage (only when positive lymph nodes were pres-ent), was proposed to the patient. In cases of unfa-vorable disease stage (T4b), poor general condition,or refusal of the surgical resection by the patient, theprimary treatment protocol comprised radiation therapyand/or chemotherapy (routinely cisplatin and 5-fluorou-racil), depending on the patient’s comorbidities. Radia-tion therapy was applied as adjuvant treatment in casesof R1 resection according to Batsakis3 (involved margin)nd in cases where the status of the resection margins

FIGURE 1. Intraoral view (A) and coronal (B) and axial (C) CTscans of a maxillary SCC in a single patient.

as unclear. Also in cases of R0 resection with free

(

tsT

(sFetw

fimcwr

b

3040 STAGING AND GRADING AS PROGNOSTIC FACTORS IN MAXILLARY SCC

margins less than 5 mm (close margin3), an adjuvantradiotherapy was proposed.

Results

Due to incomplete or missing follow-up data, thenumber of valid cases had to be adapted in thelog-rank tests, Kaplan-Meier curves, and tables.Therefore, a reduction of the total number caneventually be seen in the tables (explanation in thelegends).

The location of the tumors is shown in Figure 2multiple nominations possible).

The male:female ratio was 2:1 (m � 60, f � 33) andthe mean age at the time of diagnosis was 63 years(range 35 to 94 yrs). The overall 5-year survival ratewas 71%. The 10-year survival rate for males was 71%and for females was 60% (Fig 3). This difference wasstatistically not significant (P � .938).

FIGURE 2. Tumor location (absolute numbers, multiple nomina-tions possible).

Most patients showed a T4 stage (66%, T4a and T4b
together) and the most frequent staging was T4N0M0(42%, Table 1). Neither the cumulative overall sur-vival rates (P � .718) concerning the TNM staging norhe cumulative 5-year survival rates (P � .131) weretatistically significant as can be seen in Figure 4 andable 2.The most common histopathological grading was G2

57%) followed by G3 (22%) and G1 (21%). The differenturvival rates depending on the grading can be seen inigure 5. The survival rates showed no significant differ-nce concerning the grading (P � .376). Neverthelesshere is a tendency of a poorer prognosis for G1 SCC,hich is quite interesting and surprising.Eighty-six patients (92%) were treated surgically as

rst-line treatment and the status of the resectionargins and its influence on recurrence and survival

an be deduced from Figure 6 and Table 3. Patientsith R1 resections show significantly lower survival

ates (P � .000).Seven patients did not receive any surgical treatment

ecause of the above-mentioned limitations. All 7 pa-

FIGURE 3. Kaplan-Meier curve,gender distribution, and cumula-tive survival, P � .938.

Poeschl et al. Staging and Grad-ing as Prognostic Factors inMaxillary SCC. J Oral Maxillo-fac Surg 2011.

Table 1. TNM STAGING (ACCORDING TO UNIONINTERNATIONALE CONTRE LE CANCER2)

Staging N0 N1 N2b N2c Total

T1 8 1 0 0 9T2 11 0 3 0 14T3 8 0 0 1 9T4a 34 7 9 3 55T4b 3 1 1 1 6Total 64 9 13 5 93

There were no N2a, N3, or M1 stages.


a

tasar

illary S

POESCHL ET AL 3041

tients died subsequently because of tumor-related rea-sons.

The overall recurrence rate was 37%. The cumula-tive survival rates of patients with tumor recurrencewere significantly lower than for patients withoutrecurrence (P � .000), but there was no differencebetween the type of recurrence (regional or local,P � .778, Figs 7 and 8).

The modality of recurrence treatment was deter-mined according to the patient’s general condition,resectability of the tumor, and the patient’s decision.Due to the small number and the variety of treatmentmodalities, it is not possible to draw serious andsignificant conclusions.

FIGURE 4. Kaplan-Meier curve, T stage, and cumulative s

Poeschl et al. Staging and Grading as Prognostic Factors in Max

Table 2. T STAGE AND TUMOR-RELATED DEATHS

T-Stage Number

Number ofTumor-Related

Deaths (%)

T1 8 1 (12)T2 13 3 (23)T3 7 2 (29)T4 (Only T4a) 55 18 (33)Total 83 24 (29)

Only 83 (out of 93) patients could be evaluated; T4b stagenot included.

Poeschl et al. Staging and Grading as Prognostic Factors in
Maxillary SCC. J Oral Maxillofac Surg 2011.
Discussion

Carcinomas of the maxillary complex account forabout 0.2% of all malignancies in the United Stateswith an incidence of about 1:100,000 per year.4 Thus,they can be considered as quite rare tumors. Com-pared with other tumors of the head and neck regionand especially to other locations within the oral cav-ity, maxillary SCCs usually show 2 characteristics:first, the tumor stage is quite advanced at the time offirst presentation in the hospital5,6; second, the over-ll prognosis is poor.5,7 In our series, 66% of the

patients showed a T4 stage. Due to the proximity ofthe maxillary region to vital structures like the brain,big vessels, and cranial nerves at the skull base andthe orbit, the treatment of tumors in this region stillposes a challenge to the head and neck surgeon.

INFLUENCE OF SURGERY

Among the various treatment options, the combina-tion of primary surgical resection followed by adjuvantradiation therapy is widely accepted as the most effec-tive concept in resectable cases.4 Whenever surgicalreatment is recommended to the patient (ie, in resect-ble cases), the primary and most important goal for theurgeon is to prevent local recurrence. The best way tochieve this aim is to obtain free resection margins, thuseducing the risk of tumor recurrence,8,9 no matter

what T staging the tumor shows. Our results support

, P � .131 (5 yrs), P � .718 (cumulative overall survival).

CC. J Oral Maxillofac Surg 2011.

urvival

these findings and emphasize the importance of nega-

fcppiptsl

sr

rt

apr


tive margins as a crucial prognostic factor (Fig 6 andTable 3). However, other reports show different resultsand do not identify free resection margins as a significantfactor concerning the patients’ survival.10 To achieveree margins, the surgical procedure was adapted ac-ording to the tumor size and location, ranging fromartial maxillectomy through a simple intraoral ap-roach to a Dieffenbach-Weber-Fergusson incision mod-

fied by Zange et al11 combined with a bicoronal ap-roach in order to perform a total maxillectomyogether with a frontal skull base resection. Our surgicaltrategy usually was geared to the principles of maxil-ectomy described by Brown et al.12

There exists strong evidence in the internationalliterature that a combined surgical and radiothera-peutic strategy leads to the best results.4 In oureries we did not routinely recommend adjuvantadiotherapy in all cases. Only in cases with R1

FIGURE 6. Kaplan-Meier curve,status of resection margins, andcumulative survival, P � .000.


esections (involved margin3) or free margins lesshan 5 mm (close margin3) postoperative radiother-

apy (66 to 70 Gy delivered in fractions of 2 Gydaily) was applied. This concept was based on anindividual decision for each patient and led to anoverall 5-year survival rate of 71% and a 10-yearsurvival rate of 71% for males and 60% for females.These results are very encouraging and support thetherapeutic strategy applied in our clinic.

INFLUENCE OF T STAGING

The most common tumor stage was T4 (66%, T4a andT4b together; Table 1). Interestingly, the T stage did nothave a significant impact on the overall survival rates(Fig 4 and Table 2). This stands in contrast to otheruthors,4 who found a lower overall survival rate foratients with advanced tumor stages (T4). However, ouresults show that even advanced tumors, as far as they

FIGURE 5. Kaplan-Meier curve,grading, and cumulative sur-vival, P � .376.


trioTia

INT

O

POESCHL ET AL 3043

are resectable (T4a), can be treated successfully, al-though the esthetic and functional outcome of thesepatients, including their quality of life, which was notevaluated in this study, should be addressed too. Thus,the T stage itself cannot be considered a reliableprognostic parameter in resectable cases as far asthe surgical principles of tumor removal are re-spected. The lack of statistical significance may bedue to a different attitude regarding the extent ofthe surgical resection in smaller tumors, resulting intheir possible underestimation. Based on our re-sults, we strongly recommend not depriving surgi-cal treatment from any patient with large T4a tu-mors, because their prognosis regarding cumulativesurvival does not seem to be any worse.

INFLUENCE OF GRADING

The most frequent tumor grading was G2 (57%), butthe grading also did not have any significant influenceon the patient’s overall survival (Fig 5). In contrast, thereis even a tendency for a poorer prognosis for G1 tumors,which is quite surprising. Hence the prognostic value of

Table 3. STATUS OF RESECTION MARGINS,RECURRENCE RATES, AND TUMOR-RELATED DEATHS

Margin Status NumberRecurrence

(%)

Number DiedWithin

Group (%)

n Sano (RO) 62 17 (27) 11 (18)on in Sano (R1) 16 12 (75) 10 (62)otal 78 29 (37) 21 (27)

nly 78 (out of 93) patients could be evaluated.


FIGURE 7. Kaplan-Meier curve, tumor r

Poeschl et al. Staging and Grading as Prognostic Factors in Maxillary S

the grading is questionable. These findings partly agreewith the literature concerning tongue and floor of themouth SCC,13 but to our knowledge there are no de-tailed reports about the prognostic value of the histo-pathological grading of maxillary SCC in the literature.We will try to highlight and clarify this point in furtherinvestigations. Our data confute the presumption thatSCC with high differentiation (G1) have a better prog-nosis in terms of cumulative survival than tumors withlower differentiation (G3).

RECURRENCE

The overall recurrence rate in this series was 37%and the 5-year survival rate with recurrence was 40%.The 10-year survival rate was 27% (Figs 7 and 8 andTable 3). Compared with these patients the percent-age of tumor-related deaths in the group withouttumor recurrence was only 4%, resulting in a 10-yearsurvival rate of 96%. These findings are of course ofhigh significance and agree with the literature byidentifying tumor recurrence as the most crucial fac-tor for the patient’s survival.9 Similarly the status ofhe resection margin was a crucial factor for tumorecurrence (Table 3). These findings also support themportance of free margins in the surgical treatmentf maxillary SCC independent of T stage and grading.he grading and the T stage do not have a significant

mpact on recurrence and patients’ long-term survivalnd are therefore unreliable prognostic parameters.

PERSPECTIVES

During the last few years the application of immuno-therapy with monoclonal antibodies (ie, cetuximab) hasbeen introduced to head and neck oncology and the firstresults are encouraging.14,15 Maybe these therapeutic

ce, and cumulative survival, P � .000.
ecurren

1

1

1

1

1

1

P illary S


agents can help to further improve the results in thetreatment of maxillary SCC, especially in patients suffer-ing from recurrence, although detailed results for thisgroup have not been published yet.

The outcome data of our investigation regarding Tstage and grading identify these 2 parameters as unreli-able prognostic factors. In the clinical daily routine theinfluence of T stage and grading regarding survival isoften overestimated and may have a negative effect onthe decision-making process in terms of planning aproper treatment. Our data also show that the prognosisfor the patients is very poor without a surgical resectionof the tumors. The most crucial factor for recurrenceand therefore survival are free resection margins afterthe surgical procedure. Grading and T stage do not havea significant impact on patients’ long-term survival.

References1. Howaldt HP, Vorast H, Blecher JC, et al: Results of the

DOSAK tumor register. Mund Kiefer Gesichtschir 4(Suppl1):216, 2000

2. Sobin LH, Wittekind C: TNM Classification of Malignant Tu-mours (ed 6). International Union Against Cancer. New York,John Wiley & Sons, Inc, 2002

3. Batsakis JG: Surgical excision margins: A pathologist’s perspec-tive. Adv Anat Pathol 6:140, 1999 [Review]

4. Stavrianos SD, Camilleri IG, McLean NR, et al: Malignant tu-

FIGURE 8. Kaplan-Meier curve, location of tu

oeschl et al. Staging and Grading as Prognostic Factors in Max

mours of the maxillary complex: An 18-year review. Br J PlastSurg 51:584, 1998

5. Myers LL, Nussenbaum B, Bradford CR, et al: Paranasal sinusmalignancies: An 18-year single institution experience. Laryn-goscope 112:1964, 2002

6. Kermer C, Poeschl PW, Wutzl A, et al: Surgical treatment ofsquamous cell carcinoma of the maxilla and nasal sinuses.J Oral Maxillofac Surg 66:2449, 2008

7. Giri SP, Reddy EK, Gemer LS, et al: Management of advancedsquamous cell carcinomas of the maxillary sinus. Cancer 1:657,1992

8. Jacobs JR, Ahmad K, Casiano R, et al: Implications of positivesurgical margins. Laryngoscope 103:64, 1993

9. Ravasz LA, Slootweg PJ, Hordijk GJ, et al: The status of theresection margin as a prognostic factor in the treatment ofhead and neck carcinoma. J Craniomaxillofac Surg 19:314,1991

0. Suarez C, Llorente JL, Fernandez De Leon R, et al: Prognosticfactors in sinonasal tumors involving the anterior skull base.Head Neck 26:136, 2004

1. Zange J, Schuchardt K: Rhinologische und plastische Opera-tionen auf Grenzgebieten mit der Ophthalmologie und Chiru-rgie, in Thiel R (ed): Ophthalmologische Operationslehre.Leipzig, Germany, Thieme, 1950, pp 1302-1334

2. Brown JS, Rogers SN, McNally DN, et al: A modified classi-fication for the maxillectomy defect. Head Neck 22:17, 2000

3. Keski-Säntti H, Atula T, Tikka J, et al: Predictive value ofhistopathologic parameters in early squamous cell carcinomaof oral tongue. Oral Oncol 43:1007, 2007

4. Vermorken JB, Mesia R, Rivera F, Remenar E, Kawecki A,Rottey S, et al: Platinum-based chemotherapy plus cetux-imab in head and neck cancer. N Engl J Med 359:1116, 2008

5. Vermorken JB, Herbst RS, Leon X, et al: Overview of theefficacy of cetuximab in recurrent and/or metastatic squa-mous cell carcinoma of the head and neck in patients whopreviously failed platinum-based therapies. Cancer 12:2710,

currence, and cumulative survival, P � .778.


mor re

2008

staging and grading as prognostic factors in maxillary squamous cell carcinoma

Documents