Cervical Cancer Brachytherapy in Canada: A Focus on Interstitial Brachytherapy Utilization

Volume 96  Number 2S  Supplement 2016 Author Disclosure: S. Sekii: None. N. Murakami: None. S. Nakamura: None. T. Kashihara: None. K. Kobayashi: None. K. Harada: None. M. Kitaguchi: None. K. Takahashi: None. K. Inaba: None. H. Igaki: None. Y. Ito: None. J. Itami: None.

2756 Postoperative Radiation Therapy for Cervical Cancer: Three-Dimensional Conformal Radiation Therapy (3DCRT) Versus Intensity Modulated Radiation Therapy (IMRT) M. Kurokawa,1 Y. Iwai,1 G. Togasaki,1 A. Kanazawa,1 M. Watanabe Nemoto,2 R. Harada,1 H. Kobayashi,1 H. Usui,3 A. Mitsuhashi,3 and T. Uno2; 1Chiba University Hospital, Chiba, Japan, 2Diagnostic Radiology and Radiation Oncology, Graduate School of Medicine, Chiba University, Chiba, Japan, 3Department of Reproductive Medicine, Graduate School of Medicine, Chiba University, Chiba, Japan Purpose/Objective(s): To compare retrospectively the clinical outcomes of cervical cancer patients treated with postoperative radiation therapy using 3DCRT vs. IMRT. Materials/Methods: Between 1999 and 2015, 93 cervical cancer patients who were initially treated with radical hysterectomy and pelvic lymphadenectomy, and had at least one of pathologic risk factors for recurrence such as pelvic lymph node metastasis, the tumor size 4cm, and parametrium invasion, received postoperative radiation therapy. Eighty-three patients received concurrent chemoradiation therapy (CCRT) with daily low-dose (6.5-8 mg/m2) cisplatin and 10 patients received radiation therapy alone. Patients were separated into 2 groups regarding treatment era: 75 patients treated with 3DCRT between 1999 and 2010, and 18 patients with IMRT thereafter. The median age was 47 years. Pathological T-stages were T1a2 in 1 patient, T1b1 in 30, T1b2 in 21, T2a1 in 7, T2a2 in 12 and T2b in 22. Forty patients had lymph node metastasis. Forty-five patients presented with highrisk features (pelvic lymph node metastases and/or parametrium invasion). The median total dose of whole pelvic radiation therapy was 50 Gy. Results: The 2-year and 5-year relapse-free survival (RFS) rates were 82.6% and 81.3%, and the 2-year and 5-year overall survival (OS) rates were 94.2% and 86.1%. The factors such as histology, the numbers of lymph nodes, and presence of high-risk features did not significantly influence the patient’s prognosis. With regard to treatment modality, the 2-year RFS rate and OS rate were 81.1% and 94.6% for 3DCRT, and 93.8% and 90% for IMRT, respectively. RFS favored IMRT, but the difference did not reach significance. In high-risk patients, 2-year RFS and OS rates were 79.7% and 93.1%, respectively. The relapse occurred in 17 patients (22.7%) treated with 3DCRT, and 2 patients (11.1%) with IMRT. Among patients treated with 3DCRT, patterns of recurrence were lung metastases in 7 patients, vaginal stump in 4, extra-field lymph node recurrence in 3, pelvic lymph node recurrence in 1, skin metastasis in 1, and bone metastasis in 1, whereas both two relapses following IMRT occurred in the vaginal stump. Acute toxicities of grade 3 and 4 leukopenia were 24% and 2.7% in 3DCRT, and 27.8% and 5.6% in IMRT, respectively (n.s.). Acute grade 3 gastrointestinal (GI) and chronic  grade 3 GI toxicities were 21.3% and 6.7% in 3DCRT, and 22.2% and 11.1% in IMRT, respectively (n.s). Conclusion: The postoperative radiation therapy for cervical cancer had been switched from 3DCRT to IMRT. To date, therapeutic effects as well as toxicities of IMRT were very similar to those of 3DCRT. Two patients treated with IMRT experienced the relapse in the vaginal stump. Author Disclosure: M. Kurokawa: None. Y. Iwai: None. G. Togasaki: None. A. Kanazawa: None. M. Watanabe Nemoto: None. R. Harada: None. H. Kobayashi: None. H. Usui: None. A. Mitsuhashi: None. T. Uno: None.

2757 Intensity Modulated Radiation Therapy for Vulvar Cancer: Predictors of Local Control and Survival Y.J. Rao, J.K. Schwarz, T.A. DeWees, D. Mullen, M. Powell, D.G. Mutch, and P.W. Grigsby; Washington University School of Medicine, St. Louis, MO

Poster Viewing E309 Purpose/Objective(s): Vulvar cancer is an uncommon disease and limited guidance is available from prospective trials. We reviewed our institutional experience treating vulvar cancer with intensity modulated radiation therapy (IMRT) within the last decade. Materials/Methods: We retrospectively reviewed patients with vulvar cancer receiving IMRT at our institution between 2005 and 2015. Patients with distant metastatic disease or recurrent disease were excluded. Age, FIGO stage, tumor size, histology, grade, radiation, surgery, number and size of involved LN, ECE, LVSI, surgical margin, depth of invasion, brachytherapy, and concurrent chemotherapy were assessed. Endpoints of interest were local recurrence and overall survival. SAS was used for Cox univariate and multivariate analysis. Results: 44 patients were identified with mean follow-up of 23 months. Six patients were FIGO stage I, 7 patients were stage II, 23 were stage III, and 8 were stage IV. The included FIGO stage IV patients had AJCC T3 disease or positive pelvic LN but no distant metastases. Thirty-five (80%) patients received PET staging and 11 (25%) received MRI staging. Fourteen (32%) patients received definitive radiation, 24 (55%) received adjuvant radiation, and 6 (14%) received neoadjuvant radiation. Twentythree (52%) patients received bilateral inguinal lymph node dissection and 2 (4.3%) received sentinel LN sampling. All patients received once daily external beam radiation by IMRT to doses ranging 50.4-70 Gy. 5 (11%) patients received RT to the vulva only, while the remainder received RT to the vulva, pelvic, and inguinal nodes. 8 (18%) patients received high dose rate brachytherapy with doses ranging from 10-20 Gy. 14 (32%) patients received concurrent chemotherapy; most commonly weekly cisplatin. 14 (32%) patients developed recurrence: 8 at the vulva (18%), 2 at the inguinal LNs (4.5%) and 4 at distant metastases (9.1%). On univariate and multivariate analysis, tumor size (HR 1.05, P Z 0.01) and surgical excision of the vulva (HR 0.13, P<0.01) were the significant factors predicting local control. No significant difference in local control was seen in radiation dose, adjuvant vs neo-adjuvant RT, brachytherapy, or concurrent chemotherapy. Eight (18%) patients ultimately died of vulvar cancer in the follow-up period. Tumor size (HR 1.03, P Z 0.03) was the only factor predictive of overall survival on multivariate analysis. Three (6.8%) patients developed lymphangitis as a consequence of treatment. Conclusion: Radiation, either alone or with surgery, plays an important role in achieving local-regional control of vulvar cancer. In this cohort who all received IMRT, tumor size and surgical treatment of the vulva are the major contributors to local control. We achieved a low rate of inguinal node recurrence. The use of concurrent chemotherapy was not correlated with local control or survival. Author Disclosure: Y.J. Rao: None. J.K. Schwarz: None. T.A. DeWees: None. D. Mullen: None. M. Powell: None. D.G. Mutch: None. P.W. Grigsby: None.

2758 Cervical Cancer Brachytherapy in Canada: A Focus on Interstitial Brachytherapy Utilization A. Taggar,1,2 T. Phan,1 L. Traptow,1 R.N. Banerjee,2 and C.M. Doll1; 1Tom Baker Cancer Centre, Calgary, AB, Canada, 2University of Calgary, Calgary, AB, Canada Purpose/Objective(s): Brachytherapy (BT) techniques for cervical cancer (CC) in Canada have changed over the last decade, with evolution to HDR and image-guided BT. However, there are currently no national data on the use of interstitial BT (IBT) in the management of patients with CC. The purpose of this study was to document IBT utilization in Canadian centers, as well as update details of CC BT practices. The hypothesis is that the majority of Canadian cancer centers have adopted IBT for cervical cancer. Materials/Methods: All Canadian centers with gynecologic BT services (n Z 32) were identified, and one gynecology radiation oncologist per center was sent a 33-item e-mail questionnaire regarding the center’s practice for CC BT in 2015. Responses are reported and compared with practice patterns identified in a 2012 Canadian survey. Results: The response rate was 81% (26 of 32 centers). The majority (92%) used high dose rate (HDR) BT, identical to the 2012 survey. 93%

E310

International Journal of Radiation Oncology  Biology  Physics

(24 of 26) of centers had transitioned to 3D MRI/CT based planning by 2015, vs. 80% in 2012. 65% of centers incorporated MRI for treatment planning in 2015 compared to 37% in 2012; the majority (13/16) using a combination of MRI and CT. 50% (13 of 26 centers) had the capacity to perform IBT; in those that did not, 50% referred patients to other centers who performed the procedure. Of centers performing IBT, the majority (9/ 13) used free-hand template-based techniques. For IBT, a median of 5 (range 2-20) needles/catheters was used and an average of 4 (range 1-5) fractions were delivered. Catheters were placed using a variety of imageguidance modalities: pre-op imaging (54%), intra-op MRI (8%), intra-op CT (23%), and intra-op U/S (46%). The majority (77%) of centers performing IBT used inverse planning techniques. The most common dose/ fractionation schedules were 6 Gy x 5 fractions (40%), 8 Gy x 3 fractions (19%) and 7 Gy x 4 fractions (15%). Conclusion: In Canada, treatment of cervical cancer continues to evolve. Interstitial BT has been adopted by half of the responding centers. As more centers move to MRI-based image-guided treatment planning, IBT will become an even more integral part of cervical cancer treatment. Author Disclosure: A. Taggar: None. T. Phan: None. L. Traptow: None. R.N. Banerjee: None. C.M. Doll: None.

of HR HPV and PET-CT may be useful to identify those at highest risk of local failure and in need of closer follow-up. Author Disclosure: K. Orwat: None. A. Mendoza: None. J. Young Pierce: None. S.L. Cooper: None.

2759 Predictive Value of High-Risk Human Papillomavirus and Positron Emission Tomography/Computed Tomography for Local Control After Chemoradiation for Cervical Cancer K. Orwat,1 A. Mendoza,1 J. Young Pierce,1 and S.L. Cooper2; 1Medical University of South Carolina, Charleston, SC, 2Medical University of South Carolina, Charleston, SC Purpose/Objective(s): Current guidelines for post-treatment surveillance after chemoradiation (CRT) for cervical cancer do not incorporate high risk human papillomavirus (HR HPV) testing or the use of routine imaging. HR HPV has been shown to have superior sensitivity than cytology in identifying high-grade cervical lesions, and persistent HR HPV infection following treatment for high grade squamous intraepithelial lesions is associated with a high risk for recurrent disease. Our study analyzes the relationship between post-treatment HR HPV and PET-CT in predicting local recurrence (LR) in locally advanced cervical cancer patients who received CRT. Materials/Methods: The records of consecutive patients with locally advanced cervical cancer (FIGO stage IB1-IVB) who were treated with definitive CRT at a single institution from April 2009 to July 2015 were reviewed for post-treatment cervical cytology, HR HPV status, and PET-CT positivity. HR HPVand PET-CT were generally performed three months from completion of CRT but were included if between three and twelve months from CRT. The sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) of post treatment cervical cytology, HR HPV status, and PET-CT were determined with respect to LR. Univariate analysis and logistic regression were used to determine the effects of age, stage, race, histology, smoking status, cytology, HR HPV and PET-CT on LR. Results: We identified 53 patients with median age of 50 and median follow-up of 14.7 months. Eighty-one percent were squamous cell carcinoma (SCC), 13% adenocarcinoma, and 6% were neuroendocrine. All patients had post-treatment cytology, 35 patients had HR HPV status, 29 patients had PET-CT, and 23 patients had both HR HPV and PET-CT. There were five LRs. 25 patients (47%) had positive cervical cytology. 7 of 35 tested patients (20%) were positive for HR HPV. 13 of 29 tested patients (45%) had residual PET positivity. Cervical cytology, HR HPV and PETCT had a sensitivity of 60%/40%/100%, specificity of 42.6%/85.4%/56%, accuracy of 44.2%/81.1%/59.3%, PPV of 10%,22.2%/15.4%, and NPV of 90.9%/93.2%/100%. Two of four patients with residual PET positivity and HR HPV positive had LR. HR HPV (P Z 0.75) and smoking status (P Z 0.061) had borderline significance on univariate analysis. Combining PET positivity and HR HPV was significantly related to LR on univariate analysis (P Z 0.001). No factors were statistically significant in relation to local recurrence on logistic regression. Conclusion: The benefit of monitoring of HR HPV status following CRT for locally advanced cervical cancer remains unknown. The combination

2760 Patterns of Care of Intensity Modulated Radiation Therapy Usage in Postoperative Management of Uterine Cancer V. Osborn,1,2 D. Schwartz,1 M. Shao,2 A. Lee,3 A.T. Wong,2 and D. Schreiber2; 1SUNY Downstate Medical Center, Brooklyn, NY, 2 Veterans Affairs NY Harbor Healthcare System, Brooklyn, NY, 3Kings County Hospital Center, Brooklyn, NY Purpose/Objective(s): To analyze the patterns of care regarding intensity modulated radiation therapy (IMRT) usage in the postoperative management of uterine cancer. Materials/Methods: The National Cancer Database was queried to identify women with endometrial adenocarcinoma who underwent total abdominal hysterectomy and bilateral salpingooopherectomy followed by postoperative external beam radiation between 2004 and 2012. In order to be included, women had to have received either 1.8 Gy or 2 Gy per day to a dose of 4400-5600 cGy. Descriptive statistics were used to analyze IMRT usage by year and facility type, and were compared via the Chi Square test. Multivariate logistic regression was used to identify covariables associated with IMRT usage while controlling for age, race, year of diagnosis, facility type and insurance status. Results: There were 7,839 women included in this study. The overall IMRT utilization was 18.7% and increased from 1.9% in 2004 to 32.4% in 2012 (P<0.001). Usage varied by facility type, with academic facilities delivering IMRT 20.3% of the time, compared to 19.0% of treatments at comprehensive cancer centers, and 15.5% at community cancer centers (P Z 0.045). On multivariate logistic regression, more recent year of diagnosis was strongly associated with increased IMRT usage. Compared to the year 2004, the year 2005 had an adjusted odds ratio for IMRT usage of 2.73, 95% CI 1.57-4.72, P<0.001. This increased each year through 2012, where the adjusted odds ratio for IMRT was 24.90, 95% CI 15.24-40.67, P<0.001. Those with private insurance (OR 1.61, 95% CI 1.15-2.26, P Z 0.01) were more likely to receive IMRT, as well as those treated at an academic cancer center (OR 1.28, 95% CI 1.03-1.59, P Z 0.03). Conclusion: In this hospital-based registry, IMRT usage for postoperative radiation in uterine cancer has rapidly increased in utilization between 2004 and 2012. Academic cancer centers appear to have adapted IMRT faster than non-academic centers. Author Disclosure: V. Osborn: None. D. Schwartz: None. M. Shao: None. A. Lee: None. A.T. Wong: None. D. Schreiber: None.

2761 Clinical Outcome After Adjuvant Vaginal Brachytherapy for Stage I-II Papillary Serous Clear Cell Endometrial Cancer P. Youn,1 C.L. Angel,1 B. DuBeshter,1 S.G. Thomas,1 R.G. Moore,1 D.P. Singh,2 and K.C. Bylund1; 1University of Rochester Medical Center, Rochester, NY, 2Wilmot Cancer Institute, University of Rochester, Rochester, NY Purpose/Objective(s): To report outcomes following adjuvant high-dose rate vaginal brachytherapy (VBT) or external beam radiation therapy (EBRT) with or without chemotherapy(C) for Stage I-II papillary serous (PS) or clear cell (CC) endometrial cancer. Materials/Methods: Upon approval by institutional review board for this retrospective review, we identified 88 women with stage I-II PS (n Z 74), CC (n Z 8), mixed PS/CC (n Z 7) endometrial cancer who had surgical staging between 2007 and 2014. Recurrence free survival (RFS) and overall survival (OS) was estimated using Kaplan-Meier methods. Cox proportional hazard models were used for comparison. Results: Median patient age was 66 years (range 39-92). Surgical staging constituted hysterectomy, bilateral salpingo-oophorectomy, pelvic (94%) and para-aortic (66%) lymphadenectomy, peritoneal washing (57%), and