PROSPECTIVE EVALUATION OF A 5 × 4 Gy PRESCRIPTION FOR PALLIATION OF CANINE NASAL TUMORS

Abstract

We evaluated the efficacy of palliative radiation therapy using 5 × 4 Gy given daily in 18 dogs with nasal tumors. Dogs with malignant nasal tumors were evaluated for response rate, response duration, and survival. Seventy-eight percent of the dogs achieved complete resolution of clinical signs, and 16.5% had partial resolution of their signs. Overall median response duration for all dogs was 178 days after one course of radiation therapy. Six dogs received a second course of therapy when their disease progressed using the same daily 5 × 4 Gy scheme, and all six responded for a median time of 129.5 days for an overall median survival time in these six dogs of 309 days. Based on these results, a radiation prescription of 5 × 4 Gy appears to be useful palliatively in dogs with a malignant nasal tumor.

Introduction

Due to early onset of bone invasion, surgical resection of nasal tumors is not curative and reported survival ranges from 3 to 9 months.[1, 2] Improved survival of 47.7 months was found when an accelerated radiation protocol was followed by exenteration of the nasal cavity but chronic rhinitis and/or osteomyelitis-osteonecrosis were more common in dogs receiving surgery.[3]

Megavoltage radiation therapy has been used historically as the primary treatment for canine nasal tumors. Curative-intent radiation therapy, including intensity-modulated radiation therapy, has yielded median survival of 6.7–23 months.[3-9] Variation in protocols exist but generally curative-intent radiation therapy for nasal tumors has involved using 10–20 fractions given either daily or on a Monday–Wednesday–Friday schedule.[10, 11] Although definitive radiation therapy improves survival, some limitations exist. The potential for acute radiation side effects, the time commitment required, and the high cost of curative-intent radiation therapy are all factors that may preclude some owners from pursuing definitive treatment. Curative intent intensity-modulated radiation therapy can result in a similar survival time while reducing severity of acute and late toxicity to surrounding tissue.[9] Additionally, stereotactic radio-surgery administered in 1–3 consecutive daily fractions, reduces the time commitment. However, the limited availability of intensity-modulated radiation therapy and stereotactic radiosurgery is a problem. As a result of all of these issues, many canine nasal tumor patients are suitable candidates for palliative radiation therapy.

Palliative radiation therapy aims to achieve relief of pain and other clinical signs associated with nasal tumors while minimizing acute radiation side effects and extended treatment times.[12] Palliative radiation therapy can decrease the incidence of epistaxis associated with nasal tumors, which improves quality of life.[13] In 56 dogs treated with four weekly fractions of 9 Gy, 41 (73%) had complete resolution of epistaxis while 15 (27%) had partial resolution.[13]

There is no standard palliative prescription for canine nasal tumors. Use of coarse fractionation with large fractional doses is common. When using either a 4–5 × 6 Gy biweekly or a 3–4 × 8 Gy weekly, or 10 × 3 Gy weekly prescription in 38 dogs, there was a 100% response rate with a median duration of response of 10 months and median overall survival of 10.1 months.[14] In 48 other dogs treated with various fractionation schemes, the majority being 3 × 8 Gy given on day 0, 7, 21, the overall response rate was 91.6% with 66% complete resolution of clinical signs and a median response duration of 120 days.[15] Finally, in two other studies of 12 and 56 dogs using a weekly 4 × 9 Gy protocol, median survival was 441 days and 212 days with 95% response rates.[13, 16]

Concurrent use of cisplatin and carboplatin has been evaluated for improving response in palliative intent radiation therapy.[17] The use of carboplatin and/or cisplatin in 95 dogs treated with palliative radiation therapy led to an improved response rate and duration of response compared to dogs not receiving chemotherapy.[17] Timing of chemotherapy administration did not influence response rate or duration of response.[17]

Our goal was to evaluate prospectively the efficacy of 5 fractions × 4 Gy given daily to dogs with nasal tumors. Efficacy was assessed using clinical response rate, time to response, duration of response, and median survival.

Materials and Methods

Eighteen dogs with a malignant nasal tumor were treated between February 2008 and September 2009. There were 13 neutered females, one intact female, and four neutered males. There were five mixed breeds, three golden retrievers, two Staffordshire terriers, and one each of beagle, Labrador retriever, Pomeranian, Bassett hound, Weimaraner, Chow, Jack Russell terrier, and American Eskimo. Median age was 9.5 years (range 4–15 years). Enrollment criteria included the presence of a nasal tumor confirmed by cytology or histopathology, a nasal mass on skull images, at least one observable clinical sign, negative fungal titers, no previous chemotherapy, surgical resection, or radiation therapy, and no concurrent potentially life-threatening illness likely to result in death or euthanasia within 4 months independent of the nasal tumor. The 4 months cut off was chosen because untreated nasal tumors have a reported median survival of 95 days.

Seven dogs had a nasal sarcoma (39%). Three dogs had chondrosarcomas, one dog had an osteosarcoma, one dog had a fibrosarcoma, and two dogs had unclassified sarcomas based on cytologic diagnoses. Eleven dogs (61%) had a nasal carcinoma. Five dogs with nasal carcinomas had histopathologic diagnosis (four dogs with adenocarcinomas and one dog with a squamous cell carcinoma), while six dogs with nasal carcinomas had only cytological diagnosis.

Information obtained included signalment, presence, and duration of clinical signs prior to starting radiation therapy, biochemical profile, routine hematologic data, urinalysis, presence or absence of metastasis, concurrent medications, adverse effects from radiation, and response to treatment. Parameters evaluated for response to treatment were resolution of epistaxis, anorexia, nasal congestion, and subjective pain (Table 1). This information was obtained by having owners complete questionnaires on day 7, 14, 21, 28, and 35 following radiation therapy, then monthly until recurrence of symptoms, death from any cause, or loss to follow-up.

Table 1. Nasal Tumor Response Parameters

Subjective pain level
	1	2	3	4	5
	No pain				Severe pain
Appetite
	1	2	3	4	5
	Normal appetite				Not eating
Nasal congestion
	1	2	3	4	5
	No congestion				Severe congestion
Nose bleed
	1	2
	Yes	No

All dogs were treated with five daily doses of 4 Gy using 6 MV photons on a Monday through Friday schedule. Dogs that started therapy on a Tuesday were treated through Friday, had a weekend break, finishing the protocol the following Monday. Treatment plans were designed based on computed tomography (CT) (n = 10) or magnetic resonance imaging (MRI) (n = 2) images. When CT or MRI was not performed due to owner preference, skull radiographs (n = 6) were used for treatment planning. For this manual planning, the field extended from the most rostral aspect of nasal planum to 1 cm caudal to the frontal sinuses, thus both sides of nasal cavity and frontal sinuses were treated equally. Of dogs having CT or MRI, three had destruction of the cribriform plate and extension of the tumor into the cranial cavity. For dogs planned based on CT or MRI, the Planning Target Volume (PTV) was the tumor with a 1-cm margin in all directions.

Six dogs received carboplatin at 150 mg/m² approximately 1 h prior to their first treatment. All dogs were treated with parallel-opposed beams. All dogs were anesthetized using intravenous propofol (Abbott, Chicago, IL) induction and maintained with isoflurane. Dogs that had a clinical response after the first course of radiation therapy were eligible for a second course provided a minimum of 2 months disease-free interval was achieved. The second course was performed using the same daily 5 × 4 Gy radiation scheme.

Survival was calculated as time from the day of diagnosis to time of death. Dogs lost to follow up or who were still alive at the time of statistical analysis were censored. The Kaplan–Meier product-limit method was used to estimate survival. Log (survival) was used to calculate the 95% confidence intervals, and the lower confidence interval was calculated using a modified effective method,[18] which, due to relatively heavy censoring resulting from loss to follow-up, is expected to give more reliable bands near the tails of the survival curve. No direct hypothesis tests were performed, and instead, comparisons were made to 95% confidence intervals reported from untreated patients in a similar study design to assess whether this treatment scheme improved survival compared to untreated patients.

A thirty-five day cut off for determining response was selected to allow a 4-week duration for acute radiation side effects to resolve. Response to treatment was categorized as complete clinical response (CCR), partial clinical response (PCR), or stable disease (SD). Response was classified as CCR if complete resolution of epistaxis, anorexia, nasal congestion, and subjective pain were achieved on day 35 as per questionnaire and physical examination. Partial clinical response (PCR) was defined as apparent improvement in one or all of the parameters but not complete resolution. Stable disease (SD) was defined as no apparent improvement or worsening in one or all of the parameters 35 days postradiation. Progressive disease was defined as recurrence of epistaxis, pain, antibiotic-resistant nasal discharge or congestion, and development or progression of facial deformity.

Time to response (TTR) and progression-free interval (PFI) were measured for patients with CCR and PCR. Time to response was defined as the number of days from start of radiation therapy to improvement or resolution of clinical parameters described above. The progression-free interval was defined as time between achieving CCR and PCR to recurrence. Recurrence was based on recurrence of clinical signs with consistent physical examination findings. CT was not performed postradiation therapy due to cost. Median TTR and PFI were calculated for CCRs, PCRs, and CCRs and PCRs.

Results

Of the 18 dogs, 12 had been receiving nonsteroidal antiinflammatory drugs prior to treatment and four were administered piroxicam at the start of the study. All 16 dogs continued receiving nonsteroidal antiinflammatory drugs, while two dogs did not.

Median survival was 309 days (Fig. 1). Fourteen dogs (78%) achieved a CCR, while three (16.5%) had a PCR, for an overall response rate of 94.5%. One dog (5.5%) had SD. Median time to response for the responders was 20 days. Median PFI for the responders was 178 days (Fig. 2). Of the 14 dogs with CR, six received a second identical course of radiation therapy when disease progression occurred. Of the dogs that received a second course, all achieved a CCR, with an additional median PFI of 129.5 days (Fig. 2) and a total PFI of 307.5 days.

Figure 1. Survival curve. Median survival 309 days (95% CI: 200 to infinity). Vertical dashes: lost to follow up. Dotted lines: confidence intervals.

Figure 2. Progression-free interval box plot (PFI). Median overall PFI after 1st treatment 178 days. Median PFI after 2nd treatment 129.5 days.

The PFI for the six dogs receiving carboplatin prior to irradiation (182 days, range 143–250 days) was similar to those without carboplatin (174.5 days, range 61–510 days).

The mean PFI for the six dogs having a manual treatment plan was similar to patients with computer-based planning (219.3 vs. 223.4 days).

At the time of enrollment, no dog had lymph node or distant metastases. One dog developed ipsilateral lymph node metastases 1 month after the second course of radiation. This dog was started received chemotherapy and had SD for 4 months, then experienced nonresponsive epistaxis and nasal congestion.

Nine dogs developed acute radiation side effects, all of which were grade 1 and 2 mucositis, which on average resolved in 16 days. Three dogs developed potentially late-term side effects. One leukotrichia, one keratoconjunctivitis sicca (KCS), and one leukotrichia and KCS 2 months after the second-radiation course.

Discussion

There were multiple reasons the 5 × 4.0 Gy radiation scheme was used, including reduced cost, convenient schedule, reduced acute reaction, and low biologically effective dose on late responding tissue, resulting in a decreased potential for late-term radiation side effects thus allowing one repeat course of therapy.

The median survival of 309 days (10.3 months) is comparable to the median survival of 6.7–23 months resulting from curative-intent[3-9] and the median survival of 4–10.1 months resulting from palliative radiation protocols.[14, 15]

In the current study, response to treatment was evaluated using four clinical parameters: presence of epistaxis, severity of pain, nasal congestion, and anorexia. These four parameters were chosen because based on our experience they influence the quality of life. Additionally, epistaxis was chosen because it allowed objective evaluation of the clinical response to treatment. Eighty-nine percent of patients experienced alleviation of their clinical symptoms after the first course of therapy with a median PFI of 178 days. This response rate is comparable to other studies that evaluated responses to palliative radiotherapy in dogs with nasal tumors.

Nine of 18 dogs developed acute radiation side effects. All of these were grade 1 and 2 mucositis that were manageable and did not overtly affect the quality of life. These acute radiation side effects resolved in approximately 16 days. Three dogs (16%) developed late-term radiation side effects. Two dogs developed permanent KCS with no loss of vision and responded well to topical ophthalmic medications. One of the two dogs with KCS also developed leukotrichia at the radiation site. One dog developed leukotrichia only. No other late-term side effects developed. Overall, the acute side effects were well tolerated.

One reason we chose the 5 × 4.0 Gy radiation scheme was due to its relatively low biologically effective dose (BED). A common palliative radiation therapy prescription in veterinary medicine has been 3 × 8.0 Gy, which has a BED for late responding tissue of 88 Gy₃. The BED for late responding tissue from our 5 × 4.0 Gy prescription is 46.7 Gy₃, which implies a reduced potential for late radiation effects. Only one of the six dogs that received two courses of treatment developed a late effect (leukotrichia and KCS).

In people, concurrent use of chemotherapy and curative intent radiotherapy increases local control and survival for a variety of tumors.[19-21] Concurrent use of carboplatin and/or cisplatin for palliative radiation therapy was evaluated in 95 dogs and there was a significant increase in response rate and duration of response compared to dogs receiving radiation therapy alone.[17] We found no effect from the addition of chemotherapy but our patient numbers are too small to interpret the significance this finding.

Sixteen dogs (88.9%) received a nonsteroidal antiinflammatory drug throughout the study; drugs used were piroxicam (n = 9), carprofen (n = 4), and meloxicam (n = 3). The antiinflammatory effects of these drugs may have contributed to alleviation of nasal congestion and pain. Twelve of these dogs received a drug that had Cox-1 and Cox-2 inhibiting effects but based on recent work by others this is not likely to have influenced outcome.[22]

In summary, our results support a 5 × 4 Gy prescription being effective for palliation of signs associated with malignant nasal tumors in dogs.