Patterns of local recurrence and risk of skin recurrence in soft tissue sarcomas after surgical resection

PURPOSE/OBJECTIVE(S)

Soft tissue sarcomas (STSs) are rare, heterogeneous tumors primarily managed surgically with radiotherapy (RT) added to improve local control. Although there is a theoretical risk of skin seeding at the time of surgery, current ASTRO Clinical Practice Guidelines recommend against routine use of bolus during pre-operative or post-operative radiotherapy for STS due to increased risk of skin toxicity without clear benefit. However, in the modern era with contemporary treatment planning techniques, the risk of skin recurrence has not been assessed. We performed a detailed analysis of the patterns of local recurrence (LR) in patients with STS treated with surgical resection with or without RT.

MATERIALS/METHODS

We performed a retrospective analysis of 206 adult patients with extremity or trunk STSs evaluated in the Department of Radiation Oncology at our institution over 14 years (2007-2021) and identified all patients with LR. The depth of the recurrence and location relative to the radiation field in patients treated with radiotherapy was evaluated. Progression free survival (PFS) and overall survival (OS) were analyzed using the Kaplan-Meier method, and cumulative incidence of distant metastasis (CIDM) was calculated with competing risk analysis from the date of LR.

RESULTS

Of the 206 patients evaluated, 20 experienced a LR after surgical resection with or without RT. Among patients with a LR, 5 patients (25.0%) were treated with surgery alone and 15 patients (75.0%) received surgery and RT. In patients treated with RT, 46.7% had pre-operative RT, 53.3% had post operative RT, and bolus was used in 5 patients (33.3%). Surgical margins were close (<1mm) in 4 patients (20.0%) and positive in 10 patients (50.0%). Tumor grade was intermediate/high in 15 patients (75%) and the median tumor size was 9 cm (range 5-12). LR occurred in the muscle in 8 patients (40%), subcutaneous tissue in 11 patients (55.0%), and skin in 1 patient (5.0%). The patient with a skin recurrence was treated with surgery alone and the tumor involved the skin at presentation. In the patients treated with RT, LR occurred within radiation field in 10 patients (66.7%). Median follow up time after local recurrence was 12.9 months. At 3 years after LR, PFS was 43.9%, OS was 81.7%, and CIDM was 16.7%.

CONCLUSION

Skin recurrences were exceedingly rare after surgical resection of STSs, and only occurred in a tumor that involved the skin at initial presentation. Due to the increased dose to the normal skin and subcutaneous tissues when bolus is used, which increases risk of toxicity, these findings support current recommendations against routine use of bolus in STSs not involving the skin at presentation.

Total Lymphoid Irradiation in the Treatment of Heart Transplant Rejection: Single Institution Long-Term Results

PURPOSE/OBJECTIVE(S)

Acute cardiac allograft rejection is a major cause of morbidity and mortality post-transplant. Total lymphoid irradiation (TLI) has been shown to cause long-lasting alterations in lymphocyte sub-sets, and thereby induces a state of partial response that is useful to prevent organ transplant rejection. In our previous report, treatment with TLI resulted in decreased episodes of acute rejection and steroid dosage requirements for up to 9.1 years post cardiac transplantation. In this study, we report an update on the short- and long-term effects of TLI in the treatment of cardiac transplant rejection.

MATERIALS/METHODS

We performed a retrospective analysis of patients with intractable cardiac allograft transplantation evaluated in the department of radiation oncology at our institution over 23 years (1998-2021). Cardiac allograft rejection was defined as multiple or persistent episodes of rejection despite aggressive immunosuppressive therapy. The prescribed dose was 8 Gy in 10 fractions to be delivered over 5 weeks. All episodes of pathologic acute cardiac rejection, duration of immunosuppression, and episodes of acute infection requiring treatment before and after TLI were documented. Overall survival (OS) was analyzed using Kaplan-Meier method. The median time to first cardiac allograft rejection and the median time to first acute rejection was monitored.

RESULTS

Of the 29 patients included, 21 (72.4%) had ISHLT grade 1R, 5 (17.2%), grade 2R, and 2(6.9%) grade 3R rejection. Rejection grade was unknown for 1 patient. The mean number of acute rejections after transplant was 14.4 episodes per patient (range 1-44) with median time to first rejection of 0.63 months. The median radiation dose was 8 Gy (range 7.2-8 Gy) delivered over a mean of 31 days. Treatment was stopped early in 2 (6.5%) patients (after 7.2 Gy and 7.9 Gy) for severe bone marrow suppression and drop in CD4 counts, respectively. The number of rejections dropped from 14.2 to 3.97 per patient over a median time of 29.0 months vs 28.1 months pre and post TLI (p = 0.01). The frequency of ISHLT acute grade 3R rejections decreased from 0.58 to 0.06 episodes per patient pre and post TLI (p = 0.02). Immunosuppressant requirements decreased from 2.83 before TLI to 2.69 after TLI (p = 0.71). Infection rates per patient were 0.71 pre TLI and remained at 0.63 post TLI. Treatment was well tolerated with 4 (13.8%) grade 2 or higher radiation toxicity. Median follow up was 76.3 months. Median and 3-year OS were 53.4 months and 60.9%, respectively.

CONCLUSION

TLI is a safe and effective treatment for control of intractable cardiac rejection. The episodes of acute rejection and immunosuppressant requirements are decreased, but with long-term follow up, these patients continue to experience rejection. However, TLI lowers the frequency of ISHLT moderate to severe rejection episodes for up to 6.4 years. There was no increased risk of severe infections and there were only minimal short- and long-term grade 2 or higher radiation toxicities.

Racial Disparities in Brachytherapy Treatment among Women with Cervical and Endometrial Cancers: A United States Cohort Study between 2004 and 2017

PURPOSE/OBJECTIVE(S)

Brachytherapy (BT) improves clinical outcomes among women with cervical and endometrial cancers. Recent trends found decreasing BT rates in the United States; however, racial/ethnic differences have not been investigated completely. Here we aim to analyze 1) variations in BT practice patterns, 2) racial differences of BT, and 3) predictors of BT by race among women with cervical and endometrial cancers in the United States.

MATERIALS/METHODS

A retrospective cohort of women with endometrial and cervical cancers in the US was evaluated using the National Cancer Database, between 2004 and 2017. Primary endpoint was treatment with BT. Women ≥18 years of age were included for FIGO Stage IA-IVA, non-surgically treated cervical cancers and high intermediate risk (PORTEC-2 and GOG-99 definition) or FIGO Stage II-IVA endometrial cancers. Racial groups were in accordance with federal guidelines including Asian, American Indian and Alaska Native (AIAN), Black, Native Hawaiian and other Pacific Islander (NHPI), and White. BT practice patterns and trends over time were evaluated by race. To identify predictors of BT by race, logistic regression calculated adjusted odds ratios (aOR) and 95% confidence intervals (95% CI) after assessing for multicollinearity.

RESULTS

A total of 13,857 women with cervical cancer and 140,942 women with endometrial cancer were included. Median follow-up was 69 months. Median (IQR) age was 53 (43-63) years for cervical cancer and 64 (57-72) years for endometrial cancer. Most cervical cancers were Stage III (44%), squamous cell carcinoma (88%), and lymphovascular space invasion (LVSI) positive (49%). Most endometrial cancers were Stage IB (45%), endometrioid histology (82%), and LVSI-positive (50%). Between 2004 and 2017, endometrial cancer BT increased (34% to 57%). BT rates for cervical cancer ranged from 65% NHPI, 63% Asian, 62% White, 59% AIAN, and 57% Black. BT rates for endometrial cancer ranged from 26% Black, 25% White, 25% Asian, 21% AIAN, and 20% NHPI. By race, only Black women with cervical cancer were significantly less likely to receive BT (aOR = 0.8; 95% CI = 0.7-0.9), compared to White women. Only NHPI women with endometrial cancer were less likely to receive BT (aOR = 0.7; 95% CI = 0.5-1.0), compared to White women. Community cancer center treatment was the only factor associated with a decreased odds of BT for both Black women with cervical cancer (aOR = 0.7; 95% CI = 0.6-0.9) and NHPI women with endometrial cancer (aOR = 0.3; 95% CI = 0.1-0.8), compared to academic centers.

CONCLUSION

In this study of women diagnosed with endometrial and cervical cancers with stages known to benefit from BT, NHPI women with endometrial cancer and Black women with cervical cancer were less likely to receive brachytherapy compared to White women. Community engagement is needed to increase BT access for patients treated at community cancer centers to ensure equitable gynecologic cancer care.

Patterns of Failure for Poor Response to Neoadjuvant Chemotherapy in Gastric Cancer

PURPOSE/OBJECTIVE(S)

Peri-operative chemotherapy has emerged as the standard of care in the treatment of locally advanced gastric cancer. Improved treatment strategies are needed for patients with a poor response to neoadjuvant chemotherapy and postoperative chemoradiation has been proposed as a potential method of treatment intensification. We aimed to describe the patterns of failure for patients with no or partial response (NR, PR) to pre-operative chemotherapy.

MATERIALS/METHODS

We retrospectively reviewed charts of patients with locally advanced gastric cancer treated at our institution from 2008 to 2022 with pre-operative chemotherapy followed by surgery with a D2 resection. We excluded patients who received pre-operative or post-operative radiation therapy, or patients who had a complete response (CR) following neoadjuvant chemotherapy. Cumulative incidence of locoregional failure (LRF) and distant metastases (DM) were calculated from the time of diagnosis with competing risk analysis with death as a competing risk and censored at the last follow up. Overall survival (OS) was analyzed using Kaplan-Meier.

RESULTS

A total of 57 patients were identified and 60% are male. Median follow-up time was 31.3 months (range 6.9-181.5 months), and the median age at diagnosis was 68 years old (range 30-86). The most used pre-operative chemotherapy regimen was FLOT (38.6%), followed by FOLFOX (29.8%), and ECF/ECX/EOX (19.3%), and a majority of patients (57.9%) received adjuvant chemotherapy. The most common histology was adenocarcinoma (85.9%), and 61.4% of patients had poorly differentiated disease. Thirty-three (57.9%), and 9 patients (15.8%) were characterized as PR and NR to neoadjuvant therapy by pathology, respectively. The distribution of pathologic stages following neoadjuvant chemotherapy were as follows: 17 patients (29.8%) with Stage IA-B, 12 patients (21.1%) with Stage IIA-B, 15 patients (26.3%) with Stage IIIA, 9 patients (15.8%) with Stage IIIB, and 3 patients (5.3%) with Stage IIIC disease. Two patients had positive lymph nodes with a T0 primary tumor. Median OS was 51.4 months (95% confidence interval [CI] 39.8-68.4) for the entire cohort, and 92.1 months (95% CI 34.6-73.0) versus 42.8 months (95% CI 23.1-88.0) for patients with a PR, and NR, respectively (p = 0.14). The 2-year cumulative incidence of LRF and DM was 7.4% (95% CI 0.4-14.3) and 36.3% (95% CI 23.6-49.1), respectively. Of the five patients who experienced LRF, three of the patients also eventually or concurrently developed DM. 60% of these patients would have had their first site of recurrence covered by standard post-operative radiation treatment volumes.

CONCLUSION

Patients with locally advanced gastric cancer who do have less than a CR to chemotherapy have poor outcomes due to high rates of DM. Adjuvant locoregional therapy such as radiation is unlikely to affect survival while DM is the predominant pattern of progression. Further studies are needed to identify adjuvant therapies to improve distant control.