Marcio Fagundes

Radiation therapy for chordomas of the base of skull and cervical spine: patterns of failure and outcome after relapse

PURPOSE To determine the patterns of failure and outcome following relapse of chordomas of the base of skull and cervical spine. METHODS AND MATERIALS Between November 1975 and October 1993, 204 patients were treated for chordoma of the base of skull or cervical spine, of which 63 have developed relapse. These 63 patients constitute the main focus of this study. Forty-five patients presented with base of skull and 18 with cervical spine tumors. All patients received combined proton and photon beam radiation. The median prescribed dose was 70.1 cobalt-Gray equivalent (CGE) (range 66.6-77.4). There were 25 males and 38 females, with a median age of 41 years (range 7-66). Median follow-up was 54 months (range 8-158). RESULTS Sixty-three of the 204 patients treated (31%) had treatment failure. Among the 63 patients who relapsed, 60 (95%) experienced local recurrence, and in 49 patients (78%), this was the only site of failure. Two of 63 patients (3%) developed regional lymph node relapse and 3 of 63 (5%) developed surgical pathway recurrence (1 left neck, 1 palate and 1 nasal cavity). Thirteen of 204 patients relapsed in distant sites, accounting for 20% (13 of 63) of all patients with recurrence in this series. The most common metastatic sites were lungs and bones presenting in 7 of 13 and 6 of 13 patients, respectively. Only 2 of 13 patients failed with isolated distant metastasis. The actuarial 3- and 5-year survival rates after local relapse (60 patients) were 44 and 5%, respectively. Following distant failure (13 patients), the 3- and 5-year survival rates were 25 and 12%, respectively. After any relapse (63 patients) the corresponding survival rates were 43 and 7%. Following local relapse, 49 of 60 patients underwent salvage therapy consisting of subtotal resection in most patients (46 of 49). The remaining 11 of 60 patients received supportive care only. Salvage therapy resulted in stable or improved status without subsequent disease progression in 26 of 49 (53%), and progressive disease in 16 of 49 patients (33%). The actuarial 2- and 5-year overall survival rates following relapse for the 49 patients who underwent salvage treatment were 63 and 6%, which favorably compared to the 2-year survival rate of 21% for those who received supportive care only (p = 0.001). CONCLUSION Local relapse is the predominant type of treatment failure for chordomas of the base of skull and cervical spine. Salvage treatment may relieve symptoms; however, most patients will ultimately succumb to their disease. Poor long-term survival rates following relapse emphasize the importance of a combined treatment approach with experienced surgeons and radiation oncologists at the time of primary treatment. For most patients, only permanent local tumor control will offer a chance of cure.

Proton Therapy for Local-regionally Advanced Breast Cancer Maximizes Cardiac Sparing

Abstract Purpose: To evaluate the potential of proton therapy in sparing cardiac/coronary structures when compared with 3-dimensional conformal radiation therapy (3DCRT), helical tomotherapy (HT), and intensity-modulated radiation therapy using volumetric modulated arc therapy (VMAT). Materials and Methods: Comparative treatment planning was performed using computed tomography scans of 10 patients with left-sided stage III breast cancer after mastectomy, targeting the chest wall, axilla levels I to III, and the supraclavicular and internal mammary nodes (IMN) to 50.4 Gy (radiobiologic equivalent [RBE]) in 28 fractions. Organs at risk were heart, lungs, contralateral breast, unspecified healthy tissues, and coronary arteries. Plans were also compared that included IMNs for protons, but not for photons. Results: Mean heart dose of 1.2 Gy (RBE) was lowest with protons when compared with 6.8, 10.2, and 8.2 Gy for 3DCRT, HT, and VMAT, respectively (P < .05). The mean left anterior descending artery (LAD) dose ...

Validation of rectal sparing throughout the course of proton therapy treatment in prostate cancer patients treated with SpaceOAR

Abstract The purpose of this study was to investigate the consistency of rectal sparing using multiple periodic quality assurance computerized tomography imaging scans (QACT) obtained during the course of proton therapy for patients with prostate cancer treated with a hydrogel spacer. Forty‐one low‐ and intermediate‐risk prostate cancer patients treated with image‐guided proton therapy with rectal spacer hydrogel were analyzed. To assess the reproducibility of rectal sparing with the hydrogel spacer, three to four QACTs were performed for each patient on day 1 and during weeks 1, 3, and 5 of treatment. The treatment plan was calculated on the QACT and the rectum V90%, V75%, V65%, V50%, and V40% were evaluated. For the retrospective analysis, we evaluated each QACT and compared it to the corresponding treatment planning CT (TPCT), to determine the average change in rectum DVH points. We were also interested in how many patients exceeded an upper rectum V90% threshold on a QACT. Finally, we were interested in a correlation between rectum volume and V90%. On each QACT, if the rectum V90% exceeded the upper threshold of 6%, the attending physician was notified and the patient was typically prescribed additional stool softeners or laxatives and reminded of dietary compliance. In all cases of the rectum V90% exceeding the threshold, the patient had increased gas and/or stool, compared to the TPCT. On average, the rectum V90% calculated on the QACT was 0.81% higher than that calculated on the TPCT. The average increase in V75%, V65%, V50%, and V40% on the QACT was 1.38%, 1.59%, 1.87%, and 2.17%, respectively. The rectum V90% was within ± 1% of the treatment planning dose in 71.2% of the QACTs, and within ± 5% in 93.2% of the QACTs. The 6% threshold for rectum V90% was exceeded in 7 out of 144 QACTs (4.8%), identified in 5 of the 41 patients. We evaluated the average rectum V90% across all QACTs for each of these patients, and it was found that the rectum V90% never exceeded 6%. 53% of the QACTs had a rectum volume within 5 cm3 of the TPCT volume, 68% were within 10 cm3. We found that patients who exceeded the threshold on one or more QACTs had a lower TPCT rectal volume than the overall average. By extrapolating patient anatomy from three to four QACT scans, we have shown that the use of hydrogel in conjunction with our patient diet program and use of stool softeners is effective in achieving consistent rectal sparing in patients undergoing proton therapy.

A comparison between hydrogel spacer and endorectal balloon: An analysis of intrafraction prostate motion during proton therapy

&NA; The purpose of this study was to evaluate intrafraction prostate motion in patients treated with proton therapy and an endorectal balloon or a hydrogel spacer using orthogonal x‐rays acquired before and after each treatment field. This study evaluated 10 patients (662 fields throughout treatment) treated daily with an endorectal balloon (ERB) and 16 patients (840 fields throughout treatment) treated with a hydrogel spacer (GEL) without an ERB. Patient shifts were recorded before and after each treatment field, correlated with a treatment time, using x‐ray imaging and implanted fiducial alignment. For each shift, recorded in X, Y, and Z, a 3D vector was calculated to determine the positional change. There was a statistically significant difference in the mean vector shift between ERB (0.06 cm) and GEL (0.09 cm), (P < 0.001). The mean includes a large number of zero shifts, but the smallest non‐zero shift recorded was 0.2 cm. The largest shifts were, on average, in the Z direction (anterior to posterior). The average Z shift was +0.02 cm for both ERB and GEL. There was no statistical difference between ERB and GEL for shifts greater than 0.3 cm (P = 0.13) or greater than 0.5 cm (P = 0.36). For treatment times between 5 and 9 min, a majority of shifts were less than 0.2 cm, 85.9% for ERB and 73.2% for GEL. There was a significant positive correlation between the vector shifts and field time for both ERB (r = 0.2, P < 0.001) and GEL (r = 0.07, P < 0.04). We have shown that prostate motion is clinically comparable between an ERB and a hydrogel spacer, and the time dependencies are similar. A large majority of shifts for both ERB and hydrogel are well within a typical robust planning margin. For GEL patients, we chose to maintain slightly larger planning margins than for ERB due to already improved rectal sparing with GEL.

In Regard to Habl et al

tomographyecomputed tomography). The results are shown in Table 1. The data confirm that recurrences in a nontargeted level IB or RP are very rare. As Gunn and Garden point out, nontargeted level IB or RPs will to some extent receive dose, and we do not disagree that even fairly low levels of radiation may influence the risk of recurrence. However, the fact that patients with more advanced cancersmost likelywill receivemore unintentional dose to these regions because of the volume of disease makes analyses of this very difficult. A likely outcome of a retrospective analysis would be a correlation between the dose received (unintentionally) and a higher risk of recurrence. These issues are relevant for analyses of recurrence localization of all tumor types treated with modern, highly conformal radiation therapy. Moreover, unintentional irradiation of part of a lymph node region is not an argument for inclusion of the whole level. We aimed for a pragmatic approach to guide targeting, and other studies on elective nodal regions have used a similar methodology (3). Regarding hypopharyngeal carcinomas, Gunn and Garden read the data selectively. Although a low number of hypopharyngeal carcinoma patients become long-term survivors (often because of causes not directly related to their malignant disease), the cohort includes 120 hypopharyngeal cancers, which contribute time at risk of recurrence, and not 24 as stated in the comment. Recurrences in level IB and the retropharyngeal regions are rare, and the rationale of including these in elective radiation therapy is difficult to investigate in a setting in which these levels are included routinely. Thus, despite the concerns raised by Gunn and Garden, we believe that our dataset offers valuable information and supports a selective approach for the inclusion of level IB and the RPs.