1

Introduction

Adenoid cystic carcinoma (ACC) is a relatively uncommon malignancy that accounts for approximately 22% of all salivary gland cancers . The peak incidence is in the sixth and seventh decades of life with a wide range; there is a slight female preponderance . ACC occurs most often in the minor salivary glands of the head and neck . The most common minor salivary gland site is the posterolateral hard palate . Spiro found the following primary site distribution in a series of 281 patients treated primarily with surgery alone at the Memorial Sloan-Kettering Cancer Center (New York) between 1939 and 1973: parotid, 54 patients (19%); submandibular gland, 45 patients (16%); and minor salivary gland, 182 patients (65%) . Primary site may influence prognosis in that some minor salivary glands ACCs may be less amenable to a complete resection than those arising in the major salivary glands and have a lower probability of cure with radiotherapy alone.

The natural history of ACC is characterized by an indolent growth rate , a relatively low probability of regional lymph node metastases , and a high likelihood of hematogenous dissemination . The most common site of distant metastases is the lung . Miglianico et al at the Institute Gustave Roussy (Villejuif, France) reported a 38% 5-year survival rate after the diagnosis of distant metastases; two patients lived longer than 10 years . Because late recurrences occur with some frequency, long follow-up is necessary to assess the effectiveness of treatment.

ACC has a high propensity for perineural invasion (PNI) . PNI is grouped into two categories: incidental and clinical. Incidental PNI includes asymptomatic patients with evidence of microscopic PNI detected only by histopathology. Clinical PNI includes patients with evidence of a cranial neuropathy on physical examination and/or radiographic evidence of gross tumor involvement along the tract of the nerve. Magnetic resonance imaging (MRI) is used to detect and define the extent of PNI. Of note, it is rare to observe radiographic evidence of PNI in an asymptomatic patient. PNI is sometimes associated with “skip” lesions along the nerve, which significantly increases the risk of a recurrence after resection even if negative margins are obtained.

ACC may be histologically graded based on the solid tumor component: grade 1, no solid component; grade 2, less than 30% solid component; and grade 3, more than 30% solid component . Franzén et al reported on 51 patients treated at the National Hospital (Oslo, Norway) and observed that prognosis was adversely influenced by grade 3 histology ( P = .0001), DNA aneuploidy, and S-phase of 6% or greater ( P = .0044). Luna et al investigated a series of 26 patients with submandibular gland ACC treated at the MD Anderson Cancer Center (Houston, Tex) and found that aneuploid tumors were associated with a higher likelihood of solid cytoarchitecture, lymph node metastases, and advanced stage. Norberg-Spaak et al at the Linköping University Hospital (Linköping, Sweden) analyzed MIB1 positivity as an indicator of proliferation in a series of 31 patients and found that the mean percent of MIB1-positive cells significantly correlated with histologic grade: grade 1, 4.8%; grade 2, 8.0%; and grade 3, 20.2% ( P = .053). The likelihood of recurrence was significantly lower for patients with low MIB1 positivity. Spiro described a 33% 20-year survival rate for a series of 275 patients treated at the Memorial Sloan-Kettering Cancer Center (New York, NY); a trend towards improved survival for patients with low- and intermediate-grade tumors disappeared after 10 years of follow-up .

The purpose of this study is to evaluate the efficacy of radiotherapy (RT) alone or combined with surgery for patients with previously untreated ACCs of the head and neck.

2

Materials and methods

One hundred twenty patients with previously untreated ACCs of the head and neck received RT alone or combined with surgery at the University of Florida (Gainesville, Fla) between August 1966 and March 2008.There were 60 men and 60 women. The median age was 56 years (range, 13–82 years). One hundred patients were white, 16 were black, and 4 were of other ethnicities. No patient had distant metastases at presentation and all were treated with curative intent. Patients were excluded if they had ACCs arising in the skin or the external auditory canal. Overall, the median length of follow-up was 8.6 years (range, 0.3–39.1 years). For living patients, the median follow-up was 11.6 years (range, 3.2–39.1 years). Seven patients were lost to follow-up at a median of 11.8 years (range, 3.2–25.2 years). All patients lost to follow-up were disease-free at last follow-up and are currently alive according to the Social Security Death Index. For the purposes of data analysis, they were censored at last follow-up. The number of patients alive at various time intervals after treatment was as follows: 5 years, RT alone (25 patients) and surgery plus RT (51 patients); 10 years, RT alone (16 patients) and surgery plus RT (29 patients); and 15 years, RT alone (12 patients) and surgery plus RT (16 patients), respectively.

Patients were staged according to the 2002 American Joint Committee on Cancer (AJCC) staging system. Our series included AJCC T0 (1 patient), T1 (26 patients), T2 (25 patients), T3 (14 patients), T4 (54 patients), N0 (113 patients), N1 (2 patients), N2a (1 patient), N2b (2 patients), and N2c (2 patients). Because of the relatively few patients in this analysis, T4A and T4B lesions were grouped together (T4), as were overall stage IVA and IVB cancers (IV). Although there is no overall AJCC staging system for lacrimal gland cancers, 6 of 7 patients had T4 tumors and were classified as having stage IV disease. Thirty-five patients (29%) presented with clinical PNI, while 41 patients (34%) had incidental PNI. The oral cavity was the most common primary site (38 patients, 32%). The characteristics of the patient population are depicted in Table 1 .

The treatment philosophy over the study period was to resect the primary lesion if it appeared to be completely resectable, if the patient was thought to be medically operable, and if the functional and cosmetic result were judged to be acceptable. Postoperative RT was recommended for essentially all patients. Patients treated with RT alone were those with incompletely resectable cancers and those few with early-stage tumors located in sites, such as the soft palate, where primary RT was recommended because of the functional deficit associated with resection. In the early years of the study, a small subset of patients underwent a course of preoperative RT followed by a reevaluation and surgery if the tumor response to RT was thought to be suboptimal. Overall, 44 patients (37%) received RT alone, 73 patients (61%) received postoperative RT, and 3 patients (2%) received preoperative RT. Of the 76 patients receiving surgery, 51 patients (67%) had positive microscopic margins, 23 patients (30%) had negative margins, and 2 patients (3%) had gross residual disease.

The clinically negative neck was electively dissected at the time of surgery at the discretion of the attending physician; 11 patients received an elective neck dissection. Likewise, elective neck RT was used at the discretion of the attending radiation oncologist in patients with tumors located in primary sites thought to be relatively rich in capillary lymphatics. Sixty-four of 101 patients (63%) with an undissected clinically node-negative neck received elective neck RT; the remainder were observed.

Seven patients presented with clinically positive neck nodes; 6 of 7 underwent resection of the primary tumor and a neck dissection that revealed positive node(s). All 6 patients received postoperative RT to the neck in addition to the primary site. One patient was treated with RT alone to a base of tongue primary lesion and the clinically positive neck; a planned postradiotherapy neck revealed residual disease in the neck that was successfully resected.

All patients were treated with megavoltage photons and/or electron beams; no patient received neutrons. Sixty-five patients (54%) were treated with once-daily fractionation, 49 patients (41%) received twice-daily fractionation using a hyperfractionated schedule, and 6 patients (5%) received a concomitant-boost altered-fractionation schedule . Hyperfractionation was more likely to be used during the latter part of the study period for patients treated with RT alone to improve the likelihood of local control and/or reduce the risk of late complications, such as optic neuropathy. One hundred eleven patients (93%) received continuous-course RT and 9 patients (7%) received planned split-course RT. The latter technique was employed at the University of Florida between 1970 and 1974 and has since been abandoned. The initial treatment portals routinely included the path of potential PNI between the primary site and the skull base. The median external-beam doses for the 3 RT groups were as follows: RT alone, 74.2 Gy (range, 60–79.2 Gy); postoperative RT, 69.6 Gy (range, 10.5–75.6 Gy); and preoperative RT, 50 Gy (range, 45–61.3 Gy).

Two patients received interstitial brachytherapy as a part of their RT treatment. One patient received a radium needle implant and 10.5 Gy external-beam RT after a wide local excision of the primary tumor. Another patient with an oral tongue primary cancer received an iridium hairpin implant after external-beam RT.

Three patients (3%) received adjuvant chemotherapy. One patient with T4N0 ACC of the tonsillar fossa received RT and 4 cycles of intra-arterial cisplatin (RADPLAT) . A second patient underwent surgery for a T4N0 ACC of the maxillary sinus and had diffusely positive margins; he received weekly paclitaxel during a course of hyperfractionated postoperative RT. A third patient underwent an initial surgery for a T2N0 ACC of the lacrimal gland followed by intraarterial cisplatin and IV adriamycin. The patient subsequently declined further surgery and elected to receive definitive RT.

SAS and JMP software were used to perform all statistical computations (SAS Institute, Cary, NC). The rates of local control, neck control, local-regional control, distant metastases-free survival, cause-specific survival, and overall survival were calculated using the product-limit method . The log-rank test statistic tested for significant differences endpoints strata of potential explanatory variables for each of these endpoints. Multivariate analyses were performed using proportional hazards regression; backward selection provided the most parsimonious model of selected prognostic factors for each endpoint . Multivariate prognostic factors included the following: primary site (major vs. minor salivary gland), T stage (T0–T2 vs. T3–T4), clinical N stage (N0 vs. N-positive), overall stage (I–III vs. IV), nerve invasion (none or incidental vs. clinical), treatment group (RT alone to the primary site vs. RT and surgery), and elective neck RT (none vs. partial or total elective neck RT). For the multivariate analysis of neck control, 18 patients were excluded (7 patients with clinically positive neck nodes and 11 patients with clinically negative neck nodes who underwent an elective neck dissection). Histologic grade was not included because the philosophy of the attending pathologists was to not grade ACCs during the time period of the study.

Severe complications were defined as those necessitating hospitalization, surgical intervention, and/or resulting in death .

2

Materials and methods

One hundred twenty patients with previously untreated ACCs of the head and neck received RT alone or combined with surgery at the University of Florida (Gainesville, Fla) between August 1966 and March 2008.There were 60 men and 60 women. The median age was 56 years (range, 13–82 years). One hundred patients were white, 16 were black, and 4 were of other ethnicities. No patient had distant metastases at presentation and all were treated with curative intent. Patients were excluded if they had ACCs arising in the skin or the external auditory canal. Overall, the median length of follow-up was 8.6 years (range, 0.3–39.1 years). For living patients, the median follow-up was 11.6 years (range, 3.2–39.1 years). Seven patients were lost to follow-up at a median of 11.8 years (range, 3.2–25.2 years). All patients lost to follow-up were disease-free at last follow-up and are currently alive according to the Social Security Death Index. For the purposes of data analysis, they were censored at last follow-up. The number of patients alive at various time intervals after treatment was as follows: 5 years, RT alone (25 patients) and surgery plus RT (51 patients); 10 years, RT alone (16 patients) and surgery plus RT (29 patients); and 15 years, RT alone (12 patients) and surgery plus RT (16 patients), respectively.

Patients were staged according to the 2002 American Joint Committee on Cancer (AJCC) staging system. Our series included AJCC T0 (1 patient), T1 (26 patients), T2 (25 patients), T3 (14 patients), T4 (54 patients), N0 (113 patients), N1 (2 patients), N2a (1 patient), N2b (2 patients), and N2c (2 patients). Because of the relatively few patients in this analysis, T4A and T4B lesions were grouped together (T4), as were overall stage IVA and IVB cancers (IV). Although there is no overall AJCC staging system for lacrimal gland cancers, 6 of 7 patients had T4 tumors and were classified as having stage IV disease. Thirty-five patients (29%) presented with clinical PNI, while 41 patients (34%) had incidental PNI. The oral cavity was the most common primary site (38 patients, 32%). The characteristics of the patient population are depicted in Table 1 .

Table 1

Patient population (N = 120)

Variable	No. of patients
Primary site
Parotid gland	17 (14%)
Submandibular gland	7 (6%)
Oral cavity	38 (31%)
Oropharynx	13 (11%)
Hypopharynx	1 (1%)
Larynx	2 (2%)
Nasopharynx	5 (4%)
Nasal cavity	10 (8%)
Paranasal sinus	14 (12%)
Lacrimal gland	7 (6%)
Other	6 (5%)
T stage
T0 ⁎	1 (1%)
T1	26 (22%)
T2	25 (21%)
T3	14 (12%)
T4	54 (44%)
N stage
N0 †	113 (93%)
N1	2 (2%)
N2A	1 (1%)
N2B	2 (2%)
N2C	2 (2%)
Overall stage
I	25 (21%)
II	24 (20%)
III	15 (13%)
IV	56 (46%)
Nerve invasion
None or incidental	85 (71%)
Clinical	35 (29%)
Surgery
None	44 (37%)
Primary site	59 (48%)
Primary site and neck dissection	15 (13%)
Neck dissection	2 (2%)
Margin status
Negative	23 (19%)
Microscopically positive	51 (42%)
Gross residual	2 (2%)
No surgery	44 (37%)
Radiotherapy to primary site
Alone	44 (37%)
Preoperative	3 (3%)
Postoperative	73 (60%)
Fractionation
Once daily	65 (54%)
Hyperfractionation	55 (46%)
Split course
No	111 (93%)
Yes	9 (7%)

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Delayed lymph node metastases after elective neck dissection in patients with oral and oropharyngeal cancer and pN0 neck
2. Traumatic tympanic membrane perforations: a study of etiology and factors affecting outcome
3. Comparing capsule exposure using extracapsular dissection with partial superficial parotidectomy for pleomorphic adenoma
4. Cochlear implantation outcomes in patients with far advanced otosclerosis

Stay updated, free articles. Join our Telegram channel

Join