|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 1999 American Society for Clinical Oncology Preradiotherapy Computed Tomography as a Predictor of Local Control in Supraglottic CarcinomaFrom the Departments of Radiology, Radiation Oncology, and Biostatistics, University of Florida College of Medicine, Gainesville, FL; Department of Radiology, Netherlands Cancer Institute, Amsterdam, the Netherlands; and Department of Radiology, University Hospitals Leuven, Leuven, Belgium. Address reprint requests to Anthony A. Mancuso, MD, Department of Radiology, Box 100374, University of Florida College of Medicine, Gainesville, FL 32610-0374; email mancuso{at}xray.ufl.edu
PURPOSE: To determine the utility of pretreatment computed tomography (CT) for predicting primary site control in patients with supraglottic squamous cell carcinoma (SCC) treated with definitive radiotherapy (RT). MATERIALS AND METHODS: Pretreatment CT studies in 63 patients were reviewed. Minimum length of follow-up was 2 years. Local recurrence and treatment complications resulting in permanent loss of laryngeal function were documented. Tumor volume was calculated using a computer digitizer, and pre-epiglottic space (PES) spread was estimated. The data were analyzed using a combination of Fisher's exact test, logistic regression modeling, and multivariate analyses. Five-year local control rates were calculated using the product-limit method.
RESULTS: Local control rates were inversely and roughly linearly related to tumor volume, although there seemed to be a threshold volume at which primary site prognosis diminished. Local control was 89% in tumors less than 6 cm3 and 52% when volumes were CONCLUSION: Pretreatment CT measurements of tumor volume permits stratification of patients with supraglottic SCC treated with RT alone (which allows preservation of laryngeal function) into groups in which local control is more likely and less likely. Pre-epiglottic space spread is not a contraindication to using RT as the primary treatment for supraglottic SCC.
THE TREATMENT OF supraglottic laryngeal carcinomas varies among institutions. Radiotherapy (RT) has proven to be an effective treatment for T1, T2, and early T3 supraglottic carcinomas, with local control rates similar to those achieved with conservation surgery.1-10 Radiotherapy, like conservation laryngectomy, allows preservation of laryngeal function. Moreover, patients who refuse surgery or who are otherwise poor candidates for supraglottic laryngectomy because of underlying medical conditions or the anatomic extent of disease may be treated effectively with RT. Because of these advantages and the existence of centers capable of modern treatment planning and delivery, a growing number of patients are being treated with RT alone. Selection for treatment with definitive RT has historically been based primarily on results of the clinical examination and the patient's willingness to return for frequent follow-up visits.5,8,10 However, surgical salvage treatment undertaken as a result of local failure after RT has a greater complication rate than does a similar procedure performed as the initial step of treatment. Our group has found that up to one third of patients undergoing salvage laryngectomy after RT failures may experience a significant complication.8 It would be helpful to select those patients with lesions most likely to be locally controlled with RT so that the number of patients eventually undergoing salvage laryngectomy, with its potential added risks, could be reduced.
Pretreatment computed tomography (CT) with volumetric analysis of the primary tumor has been shown to be a potentially effective predictor of local control in a variety of laryngeal tumors treated with RT alone.11-13 Our study updated the previous study at our institution in which patients with supraglottic carcinomas were identified who had a higher likelihood of local control based on pretreatment tumor volume.12 In that study, probability of local control was 83% for tumors with volumes The purpose of this study was to determine whether pretreatment volumetric CT analysis and the amount of pre-epiglottic space (PES) involvement by tumor can be used to predict the likelihood of local control and preservation of laryngeal function in patients with a supraglottic carcinoma treated with RT.
Sixty-three patients with SCC of the supraglottic larynx who were treated with definitive RT between 1982 and 1991 underwent pretreatment, contrast-enhanced CT and were followed for a minimum of 2 years after the completion of therapy. Tumor staging was based on guidelines of the American Joint Committee on Cancer.14 Patients with definite laryngeal cartilage invasion or extralaryngeal spread on CT were excluded from this study; these patients were never treated with definitive RT but received palliative RT or surgery with or without postoperative RT. The two T4 carcinomas included in this analysis were considered stage T4 because of tenderness of the larynx and/or fullness in the neck suggestive of framework invasion or direct spread of tumor into the adjacent soft tissues of the neck. Computed tomography did not confirm this clinical impression of framework invasion. In this study, one patient had T1, 23 had T2, 37 had T3, and two had T4 carcinomas. All tumors were treated with 6,500 to 7,440 cGy through once- or twice-daily continuous-course irradiation. In some patients, this was followed by a planned neck dissection. The treatment techniques have been described previously.8 Each patient underwent pretreatment, contrast-enhanced CT of the supraglottic larynx, with 3- to 5-mm-thick contiguous sections. After 1986, all studies from this institution were performed with contiguous 3-mm sections, field of view of 16 to 18 cm, and a 512 x 512-mm matrix. All pretreatment studies were retrospectively reviewed, and a consensus was reached by two head-and-neck radiologists who were unaware of patient outcomes. Volumes of all primary tumors were measured by outlining the lesion on each image where it was visible. No attempt was made to differentiate the tumor from related edema. Each study was evaluated and when necessary corrected for errors related to patient movement, respiratory misregistration, or technologist error in selection of slice location. The resulting stack of outlines was then transferred into a treatment-planning digitizer, and the final tumor volumes (in cubic centimeters) were calculated after correcting for a magnification factor and the slice thickness. Pre-epiglottic space spread was estimated and expressed as 0% to 25%, 25% to 75%, and greater than 75% involvement of the full volume of the PES. Minimum length of follow-up for recurrent disease or major treatment complications was 2 years. Only those patients who were continuously disease-free for 2 years after the completion of RT were included. Patients who died from intercurrent disease before 2 years had passed were excluded from the analysis even though they were disease-free at the primary site. Major treatment complications included a laryngeal necrosis and/or edema that resulted in complete, permanent loss of laryngeal function; ultimately, these patients were permanently tracheostomy-dependent or underwent laryngectomy. Mild laryngeal pain or hoarseness in a patient with an otherwise normally functioning larynx was classified as a minor complication and not considered in the outcome analysis. One patient with a massive (65 cm3) tumor was excluded from analysis, to avoid skewing of the volume data. This patient's disease was controlled at the primary site, and the patient remained disease-free but was permanently tracheostomy-dependent. Statistical analysis with regard to local control rates and preservation of laryngeal function was performed using a combination of Fisher's exact test and logistic regression modeling. The rates of local control (63 patients) and local control with preservation of laryngeal function (60 patients) were also calculated, using the product-limit method.15,16 Significance levels between the curves were calculated using the log-rank test.16,17
Multivariate analyses were performed using the forward stepwise log-rank test of the association of covariates.18 The following were evaluated in multivariate analyses: primary tumor volume (less than 6 cm3 v
Overall, disease in 47 (75%) of the 63 patients was controlled with RT alone, whereas disease recurred in 16 patients (25%) at the primary site. The overall results of treatment with respect to T stage are listed in Table 1. Three patients whose disease was controlled at the primary site underwent total laryngectomy for laryngeal necrosis, and no tumor was found in the specimen obtained (Table 2). Thus, disease in 44 (70%) of 63 patients who were treated with definitive RT was controlled at the primary site, with preservation of laryngeal function, whereas 19 patients (30%) lost laryngeal function because of either local failure (n = 16) or a treatment-related complication (n = 3).
The distribution of tumor volumes with respect to local control and preservation of laryngeal function is shown in Fig 1 and Table 2. The linear trend relating tumor volume to both the rate of local control and preservation of laryngeal function is shown in Fig 2. Increasing median tumor volumes were associated with a progressively lower likelihood of control at the primary site. For larger-volume tumors, an increase in median tumor volume was associated with an even lower likelihood of preservation of laryngeal function.
Disease in three patients with tumor volumes
Local control rates at 5 years were as follows: overall, 69%; T2 tumors less than 6 cm3, 76%; T3 tumors less than 6 cm3, 86%; and T3 tumors
The effect of PES involvement by tumor on local control rates was also evaluated. Statistical analysis demonstrated no significant difference between the amount of PES involvement by the primary tumor and local control rates (P = .11) (Table 3). However, the degree of PES involvement seemed to affect the likelihood of preserving laryngeal function. No patient with less than 25% PES involvement experienced a significant treatment complication; therefore, the overall chance of local control with preservation of laryngeal function remained at 80%. Three patients with greater than 25% PES involvement, whose disease was locally controlled, subsequently developed laryngeal necrosis and underwent total laryngectomy. This difference in organ preservation based on PES less than 25% versus
Multivariate analyses of local control after RT revealed that tumor volume significantly influenced this end point (P = .0005). Differences were not significant for the remaining variables: stage of regional lymph node involvement (P = .1706), PES invasion (P = .5731), sex (P = .5897), vocal cord mobility (P = .9046), and T stage (P = .9863). Multivariate analyses of local control with preservation of laryngeal function revealed similar findings, with only primary tumor volume being significant (P = .0001). Differences were not significant for stage of regional lymph node involvement (P = .2073), sex (P = .4420), vocal cord mobility (P = .6888), T stage (P = .8888), and PES invasion (P = .9222).
This study updated the earlier, smaller study at our institution concerning the value of CT in selecting patients for definitive RT of a supraglottic carcinoma.12 In that study, measurement of tumor volume by CT was found to permit stratification of patients with a supraglottic SCC into groups very likely (83%) and less likely (46%) to have disease controlled at the primary site with definitive RT; the threshold volume was 6 cm3. Another group11 showed that tumor volume in a laryngeal carcinoma is important in predicting disease-free interval in T2 to T4 laryngeal (all sites) carcinomas treated with radical RT. In that study, mean tumor volume for lesions controlled by RT was 8.9 cm3. The suggested correlation between local control with RT alone and tumor volume as seen on CT was corroborated, but not quantitated, by our group in a previous report.19 In 1971, Fletcher20 reported that volume of disease, rather than an arbitrary stage, was the overriding factor determining the radiocurability of uterine cervix SCC. Tumor volume has a similar effect on the curability of a supraglottic SCC. The radiobiologic reasons for this are discussed elsewhere.10,21 Of the 63 patients in our study, there were three with tumor volumes of approximately 2 cm3 and one other with a volume of 5.4 cm3 in whom RT failed at the primary site. The smaller of these tumors were probably biologically aggressive and/or inherently insensitive to RT, factors that may be explained by cellular kinetics. These lesions might also have been affected by host/tumor/environmental factors such as suboptimal immune status. Many groups are working on ways to incorporate tumor and host biologic markers into the clinical-diagnostic decision-making process. These objective criteria, along with objective volume measurements, should be welcome criteria in the multifactorial analysis of treatment options for the patient with supraglottic cancer. Eventually, the methods of end results reporting will also need to account for such objective criteria.
A valid criticism of the earlier study by our group was that a single additional treatment failure in the tumors less than 6 cm3 would destroy the statistical significance of that threshold volume.21 This was overcome in the current study, especially when organ preservation is used as an end point. Thirty-four of 38 patients with tumors less than 6 cm3 had organ preservation, whereas 15 of 25 with tumors An additional criticism raised in that editorial21 concerned the knowledge and technique required to measure tumor volume. The specific issue was the inclusion of surface secretions in the measured volume as a source of potential error. The volume of such secretions would have no significant impact on the measurements. Moreover, no attempt is made to differentiate the tumor from the related edema. The elimination of this potentially subjective differentiation makes the teaching of measurement techniques to surgical and radiation oncologists, as well as diagnostic radiology trainees, relatively simple. Physicians from all of these disciplines in our program have mastered the techniques in a short period. A knowledge of normal anatomy as well as some understanding of the natural history (spread patterns) of supraglottic cancer is required. This knowledge is a requisite for any person who is involved in the care of patients with a supraglottic cancer. Tumor-volume measurement may not be suitable for physicians who encounter only an occasional case of laryngeal carcinoma. However, these volumes can be measured by any expert, such as a radiologist, radiation oncologist, or surgical oncologist, who regularly participates in the care of patients with head and neck cancer. Needed are good-quality images and an appropriate workstation or digitizer. The volume data reported here are currently used at our institution for several purposes: (1) to ensure a more accurate, informed consent process when the relative value of surgery and RT for local control and organ preservation is being discussed with the patient; (2) to help decide whether a patient who is medically, psychologically, and socially suitable for partial laryngectomy is more likely to be cured by that procedure than by RT or total laryngectomy; (3) to select patients with high-volume tumors for a trial of chemotherapy followed by triage to RT (for organ preservation in good responders) or total laryngectomy (in poor responders); and (4) to identify patients with high-volume tumors who are treated with definitive RT and are at relatively high risk for failure at the primary site (these patients may benefit from posttreatment imaging and/or radionuclide surveillance for recurrence). These uses will be discussed in light of the current study findings. Another group9,10 attained the same local control rate (an actuarial rate of 90% at 2 years) with hyperfractionated RT for selected supraglottic carcinoma without using measured volumes. Tumor bulk, edema, PES invasion, and vocal cord mobility are used to differentiate lesions more likely to be controlled with hyperfractionated RT from those that are less likely to be controlled. This group used CT to evaluate these tumors but made no apparent attempt to quantitate the CT findings. The CT data were likely used to estimate bulk disease in the PES and paraglottic space and to identify the infrequent cases of occult framework invasion and extralaryngeal spread seen in supraglottic carcinoma. Twenty-three (61%) of the 38 patients treated with RT in this group's experience were staged T2, suggesting a bias toward relatively low-volume lesions without extensive PES spread; this profile probably matches that of the majority of our 38 patients with tumors less than 6 cm3. A more quantitative approach to estimating tumor volume rather than bulk estimation might produce more uniformity in decision making across institutions; however, the approach of Lee and colleagues9,10 seemed to work for this highly experienced group. Although it cannot be proven, some of the patients triaged to total laryngectomy on the basis of their subjective selection criteria perhaps could have had organ preservation with RT if volume criteria similar to those suggested here were used. As a result of the collection of the data presented in the current report, patients at our institution who have higher-volume tumors and are suitable for partial laryngectomy are now informed of the potentially increased likelihood of organ preservation with that surgical procedure. This theoretical improvement in surgical organ preservation by such triage is yet to be proven. It is likely that many patients with higher-volume tumors will still require total laryngectomy for local control because they are anatomically or medically unsuitable for partial laryngectomy. A survival benefit may be difficult to prove because there are so few ultimate failures at the primary site in a well-screened group of patients with supraglottic cancer treated with hyperfractionated RT or partial laryngectomy. Survival is linked more to control of regional metastases (neck disease) than to primary site control in a supraglottic carcinoma.
Patients with higher-volume tumors are also being triaged to chemotherapy as a predictor of organ preservation with RT. Currently, such triage is not based on a specific threshold volume. The trend in our institution is to suggest this approach for the larger tumors (> 12 to 14 cm3) that would also necessitate total laryngectomy for local control. Follow-up CT in this group of patients has the advantage of allowing a quantitative assessment of response to chemotherapy. Volume reduction of
Finally, patients with tumor volumes Features—other than the heretofore largely subjective estimation of bulk (large or deeply infiltrating tumors)—associated with local failure after radical irradiation of a supraglottic carcinoma include T stage and invasion of the PES and tongue base.1-4,7 On the basis of T stage alone, the historical cure rates for supraglottic carcinomas at our institution have been as follows: T1, 100%; T2, 85% to 90%; and T3, 70%. When rates of local control, as traditionally reported on the basis of T stage, are compared with the results of this investigation, it becomes clear that the T3 carcinoma group is the most heterogeneous with respect to likelihood of local control with definitive RT (Table 1). This lack of predictability in the T3 carcinoma group is related to the limitations of even the best laryngoscopic examination for evaluating the deep extent of disease, compared with modern sectional imaging techniques such as CT and magnetic resonance imaging (MRI). The negative prognostic implications of extensive PES spread for both surgical and RT attempts at organ preservation have been emphasized in the literature. Few surgical or radiation oncologists have discussed their specific methods for evaluating PES invasion, and none has presented a quantitative or semiquantitative pretherapy analysis based on CT findings. Computed tomography or MRI is the only means of precisely evaluating this space before pathologic staging; this has been known for approximately 20 years.27
In our study, PES involvement was estimated in 25% increments. For the final analysis, the group was split into those patients with limited (< 25%) and those with more extensive ( High-volume PES invasion may serve to identify a subset of patients at high risk for framework invasion. In our practice, invasion of the PES increases staging from T2 to T3; disease in patients with obvious cartilage destruction or spread to the extralaryngeal soft tissues via the thyrohyoid membrane as seen on CT is staged T4. At least one group28 suggests that disease in all patients with PES spread should be staged T4, partially because of the risk of framework invasion. This probably leads to overstaging if CT evidence of limited PES spread is used as a criterion. However, given that the larynx was preserved in only four (36%) of 11 patients with greater than 75% PES invasion,28 this upstaging may be justified only in patients with such advanced PES involvement. In summary, it is relatively simple, when good-quality CT images are available, to determine tumor volume in supraglottic carcinomas. In selected groups of patients with supraglottic carcinoma, mainly those tumors characterized as large T2 or T3 by traditional staging methods, tumor volume may be a prime consideration in the choice of the method for preserving the larynx and tumor volume does not have a significant effect on the risk of ultimate treatment failure. CT (or MRI) evaluation of the PES may also contribute to better treatment decision making in the same group of patients. Some effort should be made to begin to integrate volume determinations into the system of end results reporting if the findings of this and previous similar studies are corroborated by other investigators.
1. Fletcher GH, Jesse RH, Lindberg RD, et al: The place of radiotherapy in the management of the squamous cell carcinoma of the supraglottic larynx. AJR Am J Roentgenol 108:19-26, 1970[Abstract] 2. Fletcher GH, Hamberger AD: Causes of failure in irradiation of squamous-cell carcinoma of the supraglottic larynx. Radiology 111:697-700, 1974[Medline] 3. Goepfert H, Jesse RH, Fletcher GH, et al: Optimal treatment for the technically resectable squamous cell carcinoma of the supraglottic larynx. Laryngoscope 85:14-32, 1975[Medline] 4. Harwood AR, Beale FA, Cummings BJ, et al: Management of early supraglottic laryngeal carcinoma by irradiation with surgery in reserve. Arch Otolaryngol 109:583-585, 1983[Medline] 5. Mendenhall WM, Million RR, Cassisi NJ: Squamous cell carcinoma of the supraglottic larynx treated with radical irradiation: Analysis of treatment parameters and results. Int J Radiat Oncol Biol Phys 10:2223-2230, 1984[Medline] 6. Million RR, Cassisi NJ, Mancuso AA: Larynx, in Million RR, Cassisi NJ (eds): Management of Head and Neck Cancer: A Multidisciplinary Approach (ed 2). Philadelphia, PA, Lippincott, 1994, pp 431-503 7. Robbins KT, Davidson W, Peters LJ, et al: Conservation surgery for T2 and T3 carcinomas of the supraglottic larynx. Arch Otolaryngol Head Neck Surg 114:421-426, 1988 8. Weems DH, Mendenhall WM, Parsons JT, et al: Squamous cell carcinoma of the supraglottic larynx treated with surgery and/or radiation therapy. Int J Radiat Oncol Biol Phys 13:1483-1487, 1987[Medline] 9. Lee NK, Goepfert H, Wendt CD: Supraglottic laryngectomy for intermediate-stage cancer: U.T.M.D. Anderson Cancer Center experience with combined therapy. Laryngoscope 100:831-836, 1990[Medline] 10. Goepfert H, Lee N, Wendt CD, et al: Treatment of squamous cell cancer of the supraglottic larynx, Silver CE (ed): Laryngeal CancerNew York, NY, Thieme Medical, 126-182, 1991 11. Gilbert RW, Birt D, Shulman H, et al: Correlation of tumor volume with local control in laryngeal carcinoma treated by radiotherapy. Ann Otol Rhinol Laryngol 96:514-518, 1987[Medline] 12. Freeman DE, Mancuso AA, Parsons JT, et al: Irradiation alone for supraglottic larynx carcinoma: Can CT findings predict treatment results? Int J Radiat Oncol Biol Phys 19:485-490, 1990[Medline] 13. Lee WR, Mancuso AA, Saleh EM, et al: Can pretreatment computed tomography findings predict local control in T3 squamous cell carcinoma of the glottic larynx treated with radiotherapy alone? Int J Radiat Oncol Biol Phys 25:683-687, 1993[Medline] 14. American Joint Committee on Cancer: Manual for Staging of Cancer (ed 2). Philadelphia, PA, Lippincott, 1983, pp 37-42 15. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958 16. SAS Technical Report P-179: Additional SAS/STAT Procedures, Release 6.03. Cary, NC, SAS Institute, 1988, pp 49-89 17. Lawless JE: Statistical Models and Methods for Lifetime Data. New York, NY, Wiley, 1982, pp 420-422 18. Kalbfleisch JD, Prentice RL: The Statistical Analysis of Failure Time Data. New York, NY, Wiley, 1980 19. Isaacs JH, Mancuso AA, Mendenhall WM: CT scanning as an aid to selection of therapy in T2-T4 laryngeal cancer. Head Neck Surg 99:455-464, 1988 20. Fletcher GH: Carcinoma of the uterine cervix. AJR Am J Roentgenol 111:225-242, 1971 21. Peters LJ, Wendt CD: Tumor volume as a predictor of radiocurability: A drop in the bucket? Int J Radiat Oncol Biol Phys 19:49l-438, 1990 (editorial)
22.
Lefebvre JL, Chevalier D, Luboinski B, et al: Larynx preservation in pyriform sinus cancer: Preliminary results of a European Organization for Research and Treatment of Cancer phase III trial: EORTC Head and Neck Cancer Cooperative Group. J Natl Cancer Inst 88:890-899, 1996
23.
Mukherji SK, Mancuso AA, Kotzur IM, et al: Radiologic appearance of the irradiated larynx: Part I. Expected changes. Radiology 193:141-148, 1994
24.
Mukherji SK, Mancuso AA, Kotzur IM, et al: Radiologic appearance of the irradiated larynx: Part II. Primary site response. Radiology 193:149-154, 1994 25. Greven KM, Williams DM III Keyes JW, et al: Distinguishing tumor recurrence from irradiation sequelae with positron emission tomography in patients treated for larynx cancer. Int J Radiat Oncol Biol Phys 29:841-845, 1994[Medline] 26. Mancuso AA, Drane WE, Mukherji SK: The promise FDG in diagnosis and surveillance of head and neck cancer. Cancer 74:1193-1195, 1994 (editorial) [Medline] 27. Mancuso AA, Calcaterra TC, Hanafee WN: Computed tomography of the larynx. Radiol Clin North Am 16:195-208, 1978[Medline] 28. Gregor RT: The preepiglottic space revisited: Is it significant? Am J Otolaryngol 11:161-164, 1990[Medline] Submitted October 28, 1997; accepted October 26, 1998. This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 1999 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
6 cm3 (P = .0012). The likelihood of maintaining laryngeal function also varied with tumor volume: 89% for tumors less than 6 cm3 and 40% for tumors 
6 cm3 and 46% for lesions with volumes greater than 6 cm3. Similar results were obtained by another group.
