Efficacy and tolerance of short cycles of palliative radiotherapy in advanced anal sac adenocarcinomas in dogs

Abstract

Successful management of advanced anal sac gland adenocarcinomas (AGASACA) often requires a multimodal approach. In recent years the role of palliative radiotherapy (pRT) has become increasingly more important, mainly in dogs with severe local clinical signs. This study describes efficacy and tolerance of short cycles of pRT. Retrospective single-institution study including dogs with AGASACA treated with at least one short cycle of pRT (IMRT, 4Gy BID for two consecutive days). All dogs were staged with CT and followed to death. Outcome measures were PFS and OST from the first pRT. Potential prognostic factors were evaluated. Twelve dogs were included (one stage II, four stage III, seven stage IV). Three dogs had one cycle, seven had two cycles, and two had three cycles of pRT. All dogs experienced a clinical benefit after pRT. One dog also had surgery, four systemic treatment, and five both. Eleven dogs died during the follow-up (four for local progression, six for systemic progression, one unknown). The median PFS was 198 days (95% CI 98—298) and the OST was 250 days (95% CI 124—376). Mild, acute, short-term GI toxicity occurred in seven dogs (five G1; two G2). None of the dogs experienced clinically relevant late toxicity. Short cycles of radiotherapy can be effectively used as a palliative treatment for advanced AGASACA with minimal toxicity. The small number of patients did not allow to assess the benefits of combining RT with other treatment modalities or the identification of meaningful prognostic factors.

Introduction

Apocrine gland anal sac adenocarcinoma (AGASACA) is a locally invasive tumor with moderate metastatic potential. The most common sites of metastasis at diagnosis are the medial and/or internal iliac, and/or sacral lymph nodes, with a metastatic rate from 26.0% to 96.0% [1]. Distant metastases can occur in the lungs, liver, spleen, bones, and other less common sites [1]. Mortality is typically secondary to locoregional disease progression [1]. The biological behavior of these tumors is complex and unpredictable [2].

Management of advanced AGASACA often requires multimodal treatment combining surgery, radiotherapy and systemic medical therapy. Surgical removal of the primary tumor and of the metastatic sublumbar lymph node is considered the most effective treatment for patients with AGASACA; this approach provides immediate clinical benefit and has been associated with an increased median survival time (MST) [1, 3]. However, large lymph node size, predominant cystic component of the lymph node, entrapment of major vessels, and vertebral invasion can greatly complicate surgical outcome [4]. Recent evidence suggests that after surgical removal of the primary tumor and lymph node metastases, adjuvant treatments are recommended [1, 3]. The use of chemotherapy in dogs with AGASACA, including drugs such as carboplatin, cisplatin, and mitoxantrone, has historically been recommended after surgery, but its efficacy remains uncertain. Studies have shown inconsistent results, with some showing no significant improvement in overall survival and others suggesting potential benefits that did not reach statistical significance. Further well-powered clinical trials are needed to clarify the role of chemotherapy in this setting [1]. The use of toceranib phosphate, receptor tyrosine kinase inhibitor, showed a clinical benefit in 69.0% to 88.0% of dogs with AGASACA, with a reduction in associated paraneoplastic hypercalcemia in some cases [5, 6].

Radiotherapy has shown remarkable efficacy on the measurable disease with an overall response rate ranging from 38.0% to 75.0%, proving to be an excellent technique both as adjuvant therapy following surgery and as monotherapy for inoperable tumors [7, 8].

In previous studies, the use of fractional doses greater than 3 Gy administered with non-conformal planning was associated with a high rate of long-term life-threatening complications in more than 50.0% of the cases [2, 9, 10]. This finding encouraged the use of conformal techniques (3D CRT, IMRT) with hypofractionated, moderately hypofractionated and hyperfractionated protocols, which showed very good tolerability in terms of long-term side effects [11,12,13,14,15]. In recent years, attention has increased regarding the possible use of Stereotactic Body Radiotherapy (SBRT) in the therapy of advanced AGASACA, showing acceptable treatment tolerability [16, 17].

Cyclical hypofractionated palliative radiotherapy regimen (QUAD shot) is a commonly used protocol in human medicine for advanced stage cancer, with a symptomatic improvement in over 80.0% of cases and minimal acute toxicity [18,19,20]. In veterinary medicine the QUAD shot protocol has been first described in cats with sinonasal carcinomas, with an improvement of the clinical signs reported in 85.0% of the patients, without severe acute and late toxicities [21]. In a recent retrospective study of 108 dogs with macroscopic solid tumors treated with at least one course of QUAD shot, including two cases of advanced stage AGASACA, the clinical tumor response was 92.0%, with only 3.0% of patients having acute toxicities above grade 2 and no late side effects above grade 2 [22].

To our knowledge, no study in the literature has described the efficacy and tolerance of short courses of pRT in dogs with advanced AGASACA. The objective of this study was to evaluate the efficacy and the development of early and late toxicities following one or more short cycles of pRT in dogs with advanced-stage or non-surgically resectable AGASACA. In human medicine, the QUAD shot protocol is typically administered in a maximum of 3 cycles repeated every 3 to 4 weeks. In this study, we adapted the QUAD shot approach for dogs, with cycles repeated at 6- to 8-week intervals. This extended interval was chosen to potentially mitigate acute toxicity.

Methods

Patient selection

This retrospective single-institution study included dogs diagnosed with advanced-stage (≥ Stage III) or non-surgically resectable AGASACA who underwent at least one cycle of pRT. Masses were considered unresectable based on criteria including invasion of major blood vessels, adherence to the rectal wall, or if resection posed a high risk of significant morbidity due to size or location. Patient characteristics and clinical signs at the time of diagnosis were recorded; the clinical signs at presentation were categorized as local or systemic. Diagnosis of AGASACA was confirmed through cytology or histology. All patients underwent clinical examination, hematological and biochemical blood analysis. Ionized calcium was measured in all patients with elevated total calcium and hypercalcemic patients (ICa2 +  > 1.41 mmoL/L) were documented. Urinalysis, including urinary protein-to-creatinine ratio (UPC) assessment in all patients with suspected proteinuria, was conducted reporting cases of proteinuria (UPC ≥ 1). Computed tomography (CT) served as the diagnostic modality for the staging prior to the first pRT and for the following restaging. Lymph node involvement was evaluated based on CT findings or ultrasound-guided fine needle aspiration (FNA). Details regarding surgical procedures, including the type of surgery performed and its timing relative to radiotherapy treatment, were collected. The stage of the disease was classified according to Polton et al. [23]. Only dogs that met the above inclusion criteria were included in the study.

Adjuvant treatments

Several patients received adjuvant systemic therapy, either before or after pRT. Depending on presentation findings, clinical stage and compliance of the owner toceranib or carboplatin was administered. For each patient, the specific drug and duration of treatment were recorded. All patients underwent regular clinical and laboratory monitoring to document the side effects of the adjuvant treatment. Therapeutic modalities and toxicities were categorized by or under the supervision of an ECVIM-CA (Onc) certified oncologist. Toxicities were assessed using the criteria outlined by the Veterinary Cooperative Oncology Group-Common Terminology Criteria for Adverse Events (VCOG-CTCAE v2) following investigational therapy in dogs and cats [24].

The efficacy of systemic therapies administered before the first pRT was evaluated using caliper measurement, thoracic radiographs, ultrasonography, or CT, according to the RECIST criteria [25].

CT acquisition

CT images for restaging or RT planning were acquired in general anesthesia using a GE BrightSpeed Edge 8-slice CT scanner with settings of 315 mA and 150 kV. Both thoracic and abdominal images were acquired to ensure comprehensive staging of all the patients. Following native CT acquisition, computed tomographic data were obtained immediately after intravenous injection of contrast medium (GE Healthcare, AccupaqueTM 2 ml/kg, 300 mg Iodine/ml). All dogs were positioned in sternal recumbency, with a 1 cm tissue-equivalent bolus applied in the perineal region. A syringe full of gel was placed intrarectally to avoid irradiating the whole anus/rectum. A slice thickness of 1.25 mm in helical mode was used to collect the images.

Radiation therapy treatment planning and delivery

Anatomic contouring was performed by an ECVIM-CA (Rad Onc) radiation oncologist using imported CT images and Eclipse External Beam Planning software from Varian Medical Systems (version 15.1 Copyright © Varian Medical Systems, Inc.). The following organs at risk (OARs) were contoured: rectum, colon, urinary bladder, ureters and spinal cord. The identifiable primary tumor and metastatic lymph nodes volumes visible in CT after contrast medium injection were contoured as the gross tumor volume (GTV). The sublumbar lymph nodes (medial iliac, internal iliac, and sacral) were always included in the radiotherapy plan, even when their size appeared normal. The clinical target volume (CTV) was obtained by an expansion of the GTV margin by 5 mm. The planning target volume (PTV) was created by a further isotropic expansion of the CTV margin by an additional 3–7 mm. The intensity-modulated RT (IMRT) treatment plan was developed by a medical physicist (SASRO, SGSMP), who determined the number of fields and angles. The protocol consisted of a total dose of 4 × 4 Gy (16 Gy) administered twice daily (BID) in two consecutive days. The maximum number of pRT cycles performed was three. The treatment was delivered under general anesthesia using the Varian Medical Systems Clinac 2100 C/D linear accelerator. Before every single treatment, the positioning was verified using portal imaging. The total number of pRT cycles administered and the intervals between them were documented.

Follow-up after RT

Every patient was discharged after the treatment with a variable supportive therapy according to the animal's clinical signs. We asked the owners to give us regular updates about the clinical signs in the first two weeks following the end of the treatment. A clinical recheck was always performed after three weeks from the end of the RT. To minimize the risk of confounding radiotherapy-related toxicities with those resulting from systemic treatments, systemic therapies were discontinued prior to initiation of radiotherapy and resumed three weeks after pRT, but only in patients who showed no signs of acute side effects. The toxicities were assessed following the ACVR and ECVDI consensus statement [26], under the supervision of a radiotherapist (Dip. ECVIM-CA Rad Onc). Chronic toxicities were defined as those occurring three months after the end of treatment [26]. Six to eight weeks after the first radiation therapy treatment, a second cycle of palliative radiotherapy was recommended. The following clinical and laboratory investigations depended on the individual case, most commonly with an interval of six weeks. All clinical and laboratory tests performed after the first cycle of pRT were recorded.

Treatment objective response was determined according to RECIST criteria [25] by comparing the initial CT scan used for radiotherapy planning with subsequent CT scans for further radiotherapy planning or restaging. Up to five target lesions were evaluated, with a maximum of two per organ. Overall response rate (ORR) was calculated by comparing the sum of the long-axis diameters of non-lymph-node target lesions, including the primary tumor and distant metastases, and the short-axis diameter of lymph nodes between the two CT scans. In cases where lymph node metastases involved entire lymph centers, the largest short-axis diameter of the affected lymph center was measured. Complete response (CR) was characterized by complete regression of disease. Partial response (PR) was defined as a reduction of at least 30% in the sum of the diameters of target lesions, with no appearance of new lesions. Progressive disease (PD) was defined as an increase of more than 20% in the sum of the diameters of target lesions, or the appearance of new lesions. Stable disease (SD) was defined as the absence of criteria for CR, PR, or PD [25].

In addition to assessing overall response, local response to treatment was quantified using the same RECIST criteria [25] but focused only on the target lesions included in the radiotherapy plan, such as the primary tumor and sublumbar lymph node metastases.

Statistical analysis

Data were collected using Excel (Microsoft 365). A commercial software (SPSS version 23) was used for statistical analysis. Clinical data and follow-up information were obtained either from medical records or through telephone conversations with the referring veterinarian or the owner until the 30.06.2024. Patients who were lost to follow-up were censored at the time of their last visit or phone call. In cases where it was not possible to definitively exclude a correlation between the neoplastic disease and the cause of death, the neoplastic disease was considered the cause of death. Data were assessed for normality using the Shapiro–Wilk test. For normally distributed data, values are presented as mean ± SD, whereas non-normally distributed data are presented as median with range. When indicated, both measures were reported to provide a more comprehensive view of the data.

Progression-free survival (PFS) was defined as the time expressed in days from first RT until local or systemic disease progression or death. Progression was defined on the basis of an objective CT response, worsening clinical signs, or a 20% or greater increase in palpable primary or nodal disease. Overall Survival Time (OST) was defined as the time expressed in days from first RT until death from any cause. In patients in whom the last update was made by telephone, the specific date of progression or death, if not clearly communicated/remembered by the owner, was approximated to the first day of the month.

The following characteristics were analyzed as potential prognostic factors: primary tumor size, lymph nodes involvement, number of involved lymph nodes, size of involved lymph nodes, presence of distant metastasis, performing of surgery and administration of systemic treatments. Survival analysis was performed using the Kaplan–Meier (KM) method with groups compared using the log rank test. Statistical analyses with a p-value < 0.05 were considered statistically significant.

Results

Patient population, clinical presentation, staging, surgery and systemic therapy

Twelve dogs met the inclusion criteria of the study. All the patient characteristics, the location, the size of the primary tumor and metastases, clinical signs at the presentation and the performance of surgery are shown in Table 1. The mean age was 10.5 ± 1.6. The mean weight was 21.9 ± 8.9 kg. Three dogs (25.0%) were presented with a macroscopic mass without local signs, 9 out of 12 patients (75.0%) had clinical signs. Seven of the dogs (77.8%) were presented with local signs and 2 (22.2%) with systemic signs of disease due to the hypercalcemia. Diagnosis of AGASACA was made in 50.0% of the dogs through histology and in 50.0% with cytology. The median interval between diagnosis and staging performed in our clinic was 11 days, ranging from 0 to 718 days. From the dogs enrolled in the study, 1 (8.3%) was classified as stage II, 1 (8.3%) as stage IIIa, 3 (25.0%) as stage IIIb and 7 (58.3%) as stage IV. The longest diameter of the primary tumor ranged from 5.0 to 51 mm with a mean of 28.4 ± 17.5 mm and a median of 31.0 mm. In the 11 dogs with lymph node metastasis, 4 (36.4%) had a single lymph node involved, 1 (9.1%) two lymph nodes and 6 (54.5%) three or more lymph nodes. The longest diameter of the metastatic lymph nodes ranged from 19.0 to 80.0 mm with a median of 40.5 mm. All the stage IV patients 7/12 (58.3%) had lung metastasis.

Table 1 Patient characteristics, location, size of the primary tumor and lymph node, clinical signs and surgery

Five of seven patients (41.7%) presented with stage IV disease, underwent surgery before the pRT; from those patients 4 underwent sacculectomy and 1 sacculectomy and lymphadenectomy. Only one dog (patient No. 2) underwent sacculectomy after the first cycle of pRT. Adjuvant systemic therapy was started in 8 patients (66.7%) at different stages of the follow-up. The therapy used, duration of treatment, toxicity and state of remission before the first pRT are reported in Table 2.

Table 2 Systemic therapy, duration of treatment and toxicities

All the dogs in the study underwent at least one cycle of pRT. Each patient received 4 × 4 Gy (16 Gy) administered twice daily (BID) in two consecutive days. Nine patients (75.0%) were treated with 2 cycles of pRT, with a median interval between the first and the second pRT cycle was 62 days (44—87). Only 2 patients (16.7%), dogs no. 8 and 9, received the third cycle of pRT, with an interval between the second and the third cycle of respectively 307 and 62 days.

Response

The owners of the dogs with local signs (No. 3, 4, 6, 9, 10, 11, 12) reported an improvement of the clinical signs at the first recheck, three weeks after the end of the pRT, in all cases. The two hypercalcemic dogs (No. 5, 8) had both resolution of the hypercalcemia; No. 5 after starting toceranib and No. 8 after the first pRT; the hypercalcemia never recurred. The objective CT response to the first pRT was evaluable only in 10 patients (83.3%) and is reported in Table 3. The median time between the CT for RT planning and the follow-up CT scan was 61 (50—71) days. Considering only the target lesions included in the radiation field, 4 dogs (40.0%) experienced PR and 6 (60.0%) SD, while including target lesions outside the radiation field, two dogs (20.0%) experienced PR, 5 (50.0%) SD, and 3 (30.0%) PD. In none of the cases the disease was locally progressive.

Table 3 State of remission after the first pRT

The objective CT response after the second pRT was evaluable only in 3 patients after an interval of 63, 190 and 320 days. Considering only the target lesions included in the radiation field, 2 dogs were in PR and 1 had in SD, while including target lesions outside the radiation field, 1 dog was in PR, 1 had SD and 1 had PD. In none of the cases the disease was locally progressive.

Survival analysis

All dogs were followed until death or censored on the 30.06.2024. The median follow-up was 284 days (151—802 days). Of the 12 dogs included in the study, 11 were dead and only one was still alive. Of the 11 death four died of local disease, six died of distant and one unknown.

The median PFS was 198 days (95% CI 98—298); the median PFS for dogs without distant metastasis was 372 (95% CI 81—663) and for dogs with distant metastasis was 68 (95% CI 59—184).

The OST was 250 (95% CI 124—376). There are no significant prognostic factors.

Patient no. 2 was the only still alive and free of disease at the time of the data examination, with a follow up time from the first pRT of 663 days. Patient No. 2 presented with stage II AGASACA, but was deemed not surgically resectable because the tumor was adherent to the rectal wall. After pRT, the tumor became amenable to surgery and underwent closed sacculectomy without complications.

Toxicities

The toxicities described during the treatment are described in the Table 4. In the entire follow-up period no clinically relevant late side effects were described. The only adverse event described in our patients was an acute lower gastrointestinal (GI) toxicity. After the first pRT cycle 7 patients (58.3%) showed acute lower GI toxicity of grade 1 (71.4%) or 2 (28.6%). After the second pRT cycle, 5 patients (55.5%) showed acute lower GI toxicities; in 3 (60%) of this group we described a grade 1 and in 2 (40%) cases a grade 2 acute lower GI toxicity. In the 2/12 patients that underwent the third cycle pRT no acute toxicities were recorded. Antibiotic therapy with metronidazole (15–25 mg/kg BID orally) was initiated only for grade 2 acute adverse events. None of the patients in the study required periods of hospitalization.

Table 4 Acute toxicities

Discussion

In Europe, the limited number of radiotherapy centers and the high costs of the treatment make this treatment modality inaccessible to many pet owners. Additionally, the extended duration of fractionated radiotherapy treatments, often exceeding a week, discourages many owners to pursue this kind of treatment. The primary goal of a palliative course of radiotherapy is to improve local clinical signs and, consequently, to improve the quality of life without risking severe acute side effects. Although palliative treatment does not aim to increase MST, severe local clinical signs are the main leading reason to euthanasia in dogs with anal sac adenocarcinoma [27].

The aim of our retrospective study was to evaluate the efficacy and tolerability of short cycles of pRT in dogs with advanced stage or non-surgically resectable AGASACA.

About 40.0% of the patients in our study were presented after failing systemic therapeutic modalities (Toceranib or Carboplatin) and 7 patients exhibited severe local clinical signs, which resolved with the first pRT. This clinical response is comparable with the 92.0% response reported by Sylvester et al. [22] in 108 dogs with macroscopic solid tumors treated with QUAD shot and with previous studies on palliative-intent radiotherapy for AGASACA reporting a response from 63.0% to 80.0% [7, 11]. We also evaluated the CT-based objective response according to the RECIST criteria [25], comparing the first CT for RT planning with a follow-up CT. This evaluation was possible in 83.3% of dogs who completed the first cycle of pRT and in 33.3% of dogs that completed the second cycle of pRT. The overall response rate (ORR) was 20.0% after the first cycle and 33.3% after the second cycle of pRT. The ORR in the previous studies on palliative-intent radiotherapy for AGASACA was of 38.0% [7], 55.6% [11] and 78.0% [8]. Despite that, differences in diagnostic methods and volumetric assessment criteria among the previous studies hinder a reliable comparison of ORR [7, 8, 11]. In our study, CT served as the sole diagnostic modality for assessing the objective response. Its greater diagnostic sensitivity in detecting small pulmonary nodules provides greater confidence in determining PD in dogs with distant metastasis [28]. All cases with documented progressive disease in our study were due to systemic progression, while if we were evaluating only the response of the target lesions in the radiation field, we would obtain a ORR of 40.0% after the first pRT and of 66.7% after the second pRT. In-field progression was not documented in any of the dogs. Moreover, the lack of a standardization in the follow-up CT, might have led to an underestimation of the volumetric response to the treatment. We also would like to emphasize that although the sublumbar metastases did not show a marked volumetric decrease to qualify as a PR, many patients developed a significant cystic component, which, in our opinion, was responsible for the improvement of the obstructive clinical signs associated with these masses.

Our patients showed a PFS and OST of respectively 198 (95% 68—501) and 250 (95% CI 124—376) days. These data are lower than those described by McQuown et al. [7], McDonald et al. [8], Meier et al. [11] and more recently by Valenzuela et al. [29]. The outcome in patients affected by AGASACA is closely dependent on the specific clinical stage [23], making it very difficult to compare studies that are heterogeneous in terms of the stage of the patients included. Among our patients, more than 50.0% were presented in stage IV disease, which is a higher percentage than reported in the other studies.

None of the potential prognostic factors analyzed in our study was statistically significant, probably because of the low number of patients included. After the first and second pRT, respectively 58.0% and 55.0% of patients experienced acute lower GI toxicities. In none of the cases did the toxicities exceed Grade 2. No clinically relevant late side effects were observed. The degree of acute toxicity described in previous palliative protocols ranges from 17.0% to 72.0%, with most cases involving Grade 1 and 2 toxicities [7, 8, 11, 29]. This large variability can be attributed to differences in treatment schedules, treatment planning (manual vs. conformal), hardware technologies, number of patients, clinical stage and study design. For this reason, a direct comparison of toxicities reported for palliative hypofractionated RT protocols in canine AGASACA is challenging. In our patients, although acute toxicities exceeded 50.0% of patients, in all instances they were mild and resolved without hospitalization and within the third week after the end of the treatment. The concurrent administration of systemic therapies, and the multiple anesthesia performed in 2 consecutive days could also be in part responsible for some of the acute gastrointestinal side effects. However, it is important to note that all systemic therapies were discontinued for three weeks after completing radiotherapy, in order to reduce the theoretical risk of confusing side effects of the two treatments.

This study presents several limitations. First of all, the small sample size of patients is a significant limitation. The study also lack a standardized design; it was not always possible to standardize the modalities of adjuvant therapies and the number and the interval between the pRT cycles. The retrospective and non-standardized nature of the follow-up has also made it difficult to objectively assess the progression of clinical signs in the weeks and months following the pRT cycle. The short follow-up period in some patients may have led to an underestimation of chronic toxicities. It was also not possible to assess a CT based objective response in all patients. The administration of multiple systemic therapies may have influenced the treatment response and the identification of acute toxicities. Although all animals were followed until the end of the follow-up period or death, euthanasia was often performed by the local veterinarian, inevitably lacking detailed information on the clinical condition.

In our experience short courses of pRT may be considered in cases of dogs with advanced stage or non-surgically resectable disease, severe local clinical signs, patients not suitable for multiple anesthesia, or owner with logistical or financial issues. While the cost of a single short pRT cycle is more affordable than more fractionated protocols, repeating three cycles with their respective CT plans can still result in a considerable cost. The use of short cycles of pRT, although the full protocol (3 cycles) was administrated only in a minority of cases (N = 2), showed significant palliative benefits while maintaining manageable early toxicities, making it a viable option for patients with advanced AGASACA who meet the above criteria.

Conclusion

Short cycles of pRT can be effectively used as palliative treatment for advanced AGASACA with an acceptable grade of early toxicities and no clinically relevant late toxicities. The small number of patients did not allow to assess the benefits of combining RT with other treatment modalities or the identification of meaningful prognostic factors.