Combined Hypofractionated Radiotherapy and Chemotherapy Versus Hypofractionated Radiotherapy Alone for Cats with Localized Sinonasal Lymphoma
ABSTRACT
Despite considering hypofractionated radiotherapy (HRT) a useful treatment option for feline localized sinonasal lymphoma (stage I), the benefits of additional chemotherapy remain controversial. This retrospective cohort study evaluated the efficacy of the early initiation of chemotherapy in combination with HRT (HRTC) to prolong the progression-free survival (PFS) and overall survival (OS) in cats with localized sinonasal lymphoma compared with HRT alone. While 24 eligible cats received HRT alone (HRT group), 18 received HRTC (HRTC group). The total median administered dose was 35 Gy, with one fraction per week, for a median of five fractions. In the HRTC group, the chemotherapy protocol was cyclophosphamide, doxorubicin, vincristine, and prednisolone (CHOP)-based and cyclophosphamide, vincristine, and prednisolone (COP)-based in 14 (78%) and 4 cats (22%), respectively. Cats in the HRTC group had significantly longer PFS (677 versus 104 days; P = .04) and OS (983 versus 263 days; P = .04) than those in the HRT group. Considering the poor outcome in the HRT group despite the cats having received rescue chemotherapy for progressive disease, the early initiation of additional chemotherapy along with HRT may be recommended for feline localized sinonasal lymphoma.
Introduction
Lymphoma is one of the most common feline sinonasal tumors, accounting for 30–65% of such tumors.1–3 Feline sinonasal lymphoma is often diagnosed at stage I (tumor localized to the nasal cavity and/or nasopharynx), and these tumors are highly sensitive to radiation therapy as the primary mode of treatment.4–6 Both hypo-fractionated and conventional fractionated radiotherapy protocols have been described, with median survival times following radiotherapy, with or without chemotherapy, ranging from 155 to 955 days.7–11 Disease progression following radiotherapy is observed as local recurrence within the radiation field and outside it in the regional lymph nodes and/or distant sites.7,8,10 A recent study, which used an advanced planning method using conformal dose distribution, reported that the median survival time of cats with localized sinonasal lymphoma treated with radiotherapy alone was 922 days.10 However, one-third of the cats in this study developed systemic progression soon after the radiotherapy, thereby suggesting that some cases considered as stage I disease might have had microscopic disease outside the nasal cavity at the time of diagnosis.
Chemotherapy is the primary treatment for most feline lymphomas because they typically develop systemic progression and are highly chemosensitive. However, the median survival time of cats with sinonasal lymphoma treated with chemotherapy alone is short, ranging from 98 to 154 days.2,8,11,12 Although combined radiotherapy and chemotherapy can theoretically be useful in controlling progression in cats with localized sinonasal lymphoma, the benefits of using additional chemotherapy in cats treated with radiotherapy remain controversial. In a previous study, the median survival times of cases treated with radiotherapy alone or combined radiotherapy and chemotherapy were reportedly 456 and 176 days, respectively, which were not significantly different.8 However, cats in the combined radiotherapy and chemotherapy group received a large variety of chemotherapeutic agents, and those who received only prednisone were also included in the chemotherapy group. Furthermore, various radiotherapy machines were used (orthovoltage, megavoltage unit, and Cobalt-60), and a range of radiation doses (10–57 Gy) was delivered. In another study of 19 cats with sinonasal lymphoma treated with combined conventional fractionated radiotherapy and chemotherapy, the cats achieved a long overall survival (OS) time (955 days).7 Although cats who had received prednisone alone or a single injection of L-asparaginase were excluded from the aforementioned study, various radiotherapy machines were used, including orthovoltage, megavoltage unit, and Cobalt-60. Moreover, the outcomes of the cats in this study were not compared with those of the cats treated with radiotherapy alone.
Compared with conventional fractionated radiotherapy, hypofractionated radiotherapy (HRT) has lower financial costs and is more convenient with fewer anesthesia procedures, whereas a hypofractionated protocol tends to increase the risk of late radiation effects due to the larger dose per fraction. A previous study reported that the OS of cats with localized sinonasal lymphoma treated with HRT was 432 days, and the incidence of late effects was low (5.1%).9 In this study, the use of chemotherapy was not a significant prognostic factor; however, cats who had received rescue chemotherapy after the development of relapse were included in the chemotherapy group in this study, which could have biased the OS in this group. In a more recent retrospective study, the authors of that study compared the survival time between the cats treated with HRT and with combined HRT and chemotherapy (HRTC).11 This study reported the OS of cats in the HRT and HRTC group was 1013 and 160 days, respectively. However, cases that received chemotherapy as rescue treatment for relapse after the HRT were also categorized into the HRTC group. Furthermore, this study included cases that were diagnosed as greater than stage I at the time of initiation of the HRT. Upon the inclusion of more cases with recurrent or advanced disease in the HRTC group than those in the HRT group, the case selection method would exert a significant impact on the survival analysis. In fact, when cats were regrouped based on the treatment modality (HRT or HRTC) that they received until the relapse, the PFS of the HRT and HRTC groups was 84 and 342 days, respectively; the PFS of the HRT group was extremely short relative to the OS of that group (1013 days). Therefore, the efficacy of using additional chemotherapy has not yet been clarified when the cats were treated with HRT.
In the authors’ experience, cats who developed relapse after treatment with HRT alone had a poor prognosis even if they received chemotherapy after relapse. In the present study, we hypothesized that the initiation of chemotherapy during or soon after the HRT would improve the outcome of cats with localized sinonasal lymphoma compared with the cats who were treated with HRT alone. The objective of this retrospective cohort study was to evaluate the efficacy of the early initiation of chemotherapy in combination with HRT, compared with HRT alone, for prolonging PFS and OS in cats with localized sinonasal lymphoma.
Materials and Methods
Case Selection
The Animal Medical Center electronic database of Gifu University was reviewed from February 2011 to January 2021 to identify cats diagnosed with localized lymphoma in the nasal cavity and/or nasopharynx by cytology or histopathology. We included cats with localized sinonasal lymphoma (i.e., stage I disease) who were treated with HRT using 4 MV a linear acceleratora, with or without chemotherapy. Tumors with intracranial invasion were also classified as stage I disease, because they were considered a local extension of primary tumors.
Cats were divided into two treatment groups: HRT group and HRTC group. Cats placed in the HRTC group were those who started chemotherapy, either with a CHOP- or COP-based protocol, during or within 1 wk of completing radiation therapy. Cats who did not receive chemotherapy, or corticosteroids only, were placed in the HRT group. Cats were excluded from the HRT group if they received L-asparaginase or any other chemotherapy before HRT or were lost to follow up within 3 mo of completing HRT.
Medical Record Review and Follow-Up Data
Data collected from the medical database included signalment, body weight, clinical signs at the time of presentation, prior administration of corticosteroids, presence or absence of anemia, results of feline leukemia virus and feline immunodeficiency virus tests, imaging findings, radiation protocol, radiation dose statistics, the dates of HRT initiation and end, chemotherapy protocol, the dates of chemotherapy initiation and end, the dates and location of relapse, rescue therapy for relapse, and the date and cause of death.
All cats received whole-body computed tomography (CT) and abdominal ultrasonography before HRT. The absence of lymphoma in the mandibular lymph nodes was determined by aspiration cytology or CT findings with no enlargement and lack of asymmetry in the mandibular lymph nodes. The liver, spleen, and kidneys were considered to not be involved if they had normal echogenicity on ultrasound. CT images were used to evaluate potential involvement of the cribriform plate or brain.
The referring veterinarians or owners were contacted via telephone, email, and fax during data collection for additional follow-up information. Confirmation of recurrence of the lymphoma was made based on clinical examinations, imaging findings, and cytology/histopathology when possible.
Treatment and Response
HRT was delivered once weekly to all cats. The individualized radiotherapy plans were generated with a three-dimensional CT-based computer-generated treatment planning systemb, using a customized bite block and/or a vacuum-mattress immobilization devicec in the treatment position (sternal recumbency). The linear accelerator was equipped with a multileaf collimator (forty-one 10-mm leaves) for the delivery of planned three-dimensional conformal radiation therapy. Treatments were delivered in two to nine unequally weighted fields in multiple oblique beam arrangements. Megavoltage portal imaging was performed before the treatment to verify the appropriate positioning before each treatment. Gross tumor volume (GTV) was defined by the contrast-enhancing area on the CT images. In 24 cats, the clinical target volume was contoured to include regions at a risk of microscopic disease extension based on personal judgement by the experienced veterinarians (S.G., R.I., and T.M.). The planning target volume (PTV) was defined by a margin of 0–5 mm (median: 2 mm) around the clinical target volume. The PTV was defined by adding a margin of 0–10 mm (median: 4 mm) to the GTV in the remaining 18 cats. Organ at risk contouring practices were not standardized. The radiation dose prescription point was set at the isocenter rather than covering 95% of the PTV. The source-axis distance was 100 cm, and source-to-surface distances were based on the cat size and were automatically determined by the treatment planning system. No cat received prophylactic irradiation of the regional lymph nodes. Records of the radiotherapy protocols included total radiation dose, number of fractions, and dose per fraction. The doses absorbed by 2%, 50%, and 98% of the GTV or PTV (D2%, D50%, and D98%, respectively) were used for plan evaluation in both treatment groups. D2% and D98% represent the near maximum and near minimum doses in the GTV or PTV. The relative absorbed dose was calculated as follows; (100 × D2% or D98% dose) ÷ prescribed dose. We also calculated the homogeneity index, which is defined as (D2% – D98%) ÷ D50%, to quantify the dose homogeneity.
All cats in the HRCT group received either a CHOP-based or a COP-based chemotherapy regimen. Although there were some variations (e.g., the omission of L-asparaginase and/or methotrexate), standard treatment schedules were used as previously reported.13–17
Tumor response assessments were based on changes in the tumor size as measured by CT at the end of the radiotherapy schedule and were classified as complete response (disappearance of all target lesions), partial response (≥30% decrease in sum of longest diameter of target lesions compared to baseline), progressive disease (≥20% increase in sum of longest diameter of target lesions compared to baseline or appearance of 1 or more new lesions), or stable disease (<30% decrease or <20% increase in sum of longest diameter of target lesions) according to the Veterinary Cooperative Oncology Group RECIST guidelines,18 when possible. The objective response rate, defined as the proportion of the cats with a complete or partial response, and the complete response rate were confirmed by CT. Complete and partial clinical improvement were defined as complete and partial improvement of the clinical symptoms, respectively. When clinical symptoms did not improve or worsened this was defined as disease progression.
Statistical Analyses
Continuous variables, including age, body weight, prescribed total radiation dose, GTV, PTV, and dose statistics of the GTV and PTV (homogeneity index and the relative absorbed dose of D2% and D98%), were compared between the groups using the Mann-Whitney U test. Categorical variables, including sex, complete response rate, objective response rate, complete clinical signs improvement, prior treatment with corticosteroids, presence or absence of anemia, cribriform plate destruction, and intracranial invasion, were compared between the groups using the χ2 or Fisher’s exact test.
PFS was the primary endpoint, and OS was the secondary endpoint in the survival analyses. The PFS was calculated as the time from treatment initiation to the date disease progression was identified or death from any cause occurred. Cats were censored if they had not developed progressive disease by the time of data analysis or if they were lost to follow-up before the development of progressive disease. The OS was calculated as the time from the treatment initiation to that until death due to any cause. Cats that were alive or lost to follow-up by the time of data analysis were censored. The Kaplan-Meier method was used to estimate and display the distribution of PFS and OS, and the log-rank test was used to compare the survival distribution between the HRT and HRTC groups. The probability of developing a lymphoma relapse in regional lymph nodes and/or distant sites was compared between the HRT and HRTC groups by the odds ratio using Fisher’s exact test. All statistical analyses were performed using the EZR software, which is a graphical user interface for R and is a modified version of R-Commander.19 Differences were considered significant if P < .05.
Results
Case Characteristics
Of the 58 cats with localized sinonasal lymphoma who received HRT during the study period, 42 (72%) were included in the present study. We excluded 16 cats for the following reasons: pretreatment with chemotherapy (n = 6), lost to follow-up during 3 mo after HRT initiation (n = 5), a single injection of chemotherapy (n = 3), and the initiation of chemotherapy more than 1 wk after completion of the HRT (n = 2). Of the 42 cats, 24 and 18 were enrolled in the HRT and HRTC groups, respectively.
The included breeds comprised domestic shorthair (n = 38), American shorthair (n = 1), Maine coon (n = 1), Russian blue (n = 1) and Somali (n = 1). The median age of the population was 11 yr (range: 2 – 17 yr). Tumors were diagnosed by histopathology and cytology in 35 and 7 cats, respectively. Tumor immunotypes were available for 6 cats by immunohistochemistry and 3 cats by polymerase chain reaction for antigen receptor rearrangements, and all of the tumors were of B-cell origin in these cats. The most prevalent clinical sign was nasal discharge (n = 27), followed by facial deformity (n = 23), sneezing (n = 19), increased respiratory noise (n = 13), loss of appetite (n = 11), and signs of neurological disease (n = 3). The nasal discharge was characterized as epistaxis in 19 cats. Other case demographics are presented in Table 1.
Aspiration cytology of the mandibular lymph node showed no evidence of lymphoma in 9 cats; lymph nodes were confirmed to be normal via CT in an additional 33 cats. Of the 13 cats with intracranial invasion, nasopharyngeal tumors invaded the base of the brain without cribriform plate destruction in 2 cats. There was no difference between the HRT and HRTC groups in any of the following variables: age, sex, body weight, results of feline immunodeficiency virus and feline leukemia virus, presence or absence of anemia, cribriform plate destruction, and intracranial invasion.
Treatment
All cats received the varied HRT regimen and completed the intended course. They were administered a total dose of 28–48 Gy (median: 35 Gy) of radiation in four to eight fractions (median: five fractions). Dose statistics in both treatment arms are shown in Table 2. There was no difference between the HRT and HRTC groups in the prescribed total dose, the number of fractions, GTV, and homogeneity index and the relative absorbed dose of D2% and D98% for GTV and PTV. PTV in the HRT group was significantly larger than that in the HRTC group (P = .02).
Fourteen cats (78%) in the HRTC group were treated with the CHOP-based protocol, whereas 4 (22%) were treated with the COP-based protocol. Chemotherapy was initiated during the radiotherapy schedule in 9 cats, whereas the remaining 9 cats received it soon after the radiotherapy schedule had been completed. The median time between the first radiotherapy course and the initiation of the chemotherapy was 22 days. Of the 18 cats in this group, 7 completed the intended chemotherapy protocol, and in 10, it was discontinued prematurely because of disease progression (n = 7), owner’s request (n = 2), or adverse effects (n = 1). The remaining 1 cat died from chemotherapy-related toxicity. The median total chemotherapy duration of the cats who completed the intended protocol and those who discontinued the protocol were 248 and 99 days, respectively.
Follow-Up and Survival Assessment
Tumor responses were evaluated by CT performed on the day of the final irradiation in 32 of 42 cats. Between cats receiving HRT and those receiving HRTC, no significant differences were found in objective response rate (HRT 88% versus HRTC 94%; P = .99), complete response rate (HRT 31% versus HRTC 25%; P = .99), or complete improvement of clinical signs (P = .81). Three cats in the HRT group (13%) and seven in the HRTC group (39%) were still alive at the time of data analysis. In the HRT group, 16 cats (67%) died of progressive disease, 3 (13%) from uncertain causes, and 1 (4%) of a disease other than lymphoma; 1 cat (4%) was lost to follow-up. In the HRTC group, five cats (28%) died of progressive disease, two (11%) from uncertain causes, two (11%) of diseases other than lymphoma, and one (6%) from severe adverse effects of the chemotherapy; one cat (6%) was lost to follow-up. We observed recurrence outside the radiation field in the regional lymph nodes, at distant sites, and in both regional lymph nodes and distant sites of 2 cats (HRT: n = 0, HRTC: n = 2), 14 cats (HRT: n = 11, HRTC: n = 3), and 2 cats (HRT: n = 2, HRTC: n = 0), respectively. The probability of developing these relapses was 59% in the HRT group and 29% in the HRTC group; furthermore, 42% in the HRT group and 22% in the HRTC group developed within 6 mo. The difference of the probability of developing these relapses between the HRT and the HRTC group was not statistically significant (odds ratio, 3.35; P = .11) (Table 3). The distant relapse sites in these cats included the kidneys (n = 13), lungs (n = 3), liver (n = 1), spleen (n = 1), skin (n = 1), and stomach (n = 1).
The median PFS and OS in the HRT group were 104 days (range, 31–1212 days) and 263 days (range, 50–1857 days), respectively, which were significantly lower than 677 days (range, 28–1964 days; P = .04) and 983 days (range, 38–1964 days; P = .04), respectively, in the HRTC group (Figure 1). The 1- and 2-yr survival rates for cases in the HRTC group were 66% and 58%, respectively, compared with 45% and 28%, respectively, in the HRT group.
Citation: Journal of the American Animal Hospital Association 58, 5; 10.5326/JAAHA-MS-7267
Nine of the 17 cats (53%) who developed progressive disease in the HRT group received rescue therapy that included chemotherapy alone (n = 5), radiotherapy and chemotherapy (n = 3), and radiotherapy alone (n = 1). Seven of eight cats (88%) with progressive disease in the HRTC group received rescue therapy that included either chemotherapy alone (n = 5) or radiotherapy and chemotherapy (n = 2). The median survival time following a relapse in the cats treated with rescue therapy was 103 days (range: 13–405 days) and 169 days (range: 17–475 days) in the HRT and HRTC groups, respectively.
Discussion
Although several studies have assessed different survival parameters (OS, disease-free survival, PFS, and time to progression) in relation to localized sinonasal lymphoma in cats treated with radiotherapy, the benefits of using chemotherapy plus radiotherapy remain controversial.7–11 PFS was the primary endpoint in this study and was defined as the time elapsed between the treatment initiation and tumor progression or death from any cause. In retrospective studies, it is often challenging to determine whether the cats died without lymphoma progression, especially when the follow-up information was collected from referring veterinarians and owners as long as postmortem examinations to confirm the absence of lymphoma were not performed. In previous studies, however, cats who supposedly died from causes unrelated to the lymphoma were excluded from PFS and progression-free interval analyses; therefore, these missing cases might have biased the analyses.7,9,11 In general, PFS, which measures the time from treatment initiation to disease progression or death from any cause, is considered a more appropriate endpoint than OS to evaluate the effectiveness of adjuvant chemotherapy because PFS is not affected by rescue therapies. The median PFS in the HRTC group of this study was significantly longer than that in the HRT group, which suggested that the early initiation of chemotherapy combined with HRT might provide survival benefit, although multivariate analysis was not performed because of the sample size. A previous study reported OS for cats treated with HRT alone was 1013 days, which was much longer than OS for HRT group (263 days) in the present study.11 However, their results are not comparable with ours because the studied populations were different. For example, the previous study categorized cases that relapsed after HRT and received rescue chemotherapy into the HRTC group and not the HRT group. However, on categorizing cases in relation to treatment modalities received before relapse in the previous study, the PFS was short in the HRT group (84 days) relative to that in the HRTC group (342 days), although the time to initiation of chemotherapy was not described. This was similar to our findings; the PFS of the HRT and HRTC groups was 104 and 677 days, respectively.
Within 6 mo following the end of the radiotherapy, 42% and 22% of the cats in the HRT and HRTC groups, respectively, experienced disease progression outside the radiation field (i.e., regional lymph nodes and/or distant sites), which suggested that our population might have had cases of microscopic disease outside the nasal cavity at the time of staging. Prior administration of corticosteroids treatment might be considered to have had an adverse effect on the staging, because it might mask the disease. Moreover, aspiration cytology of the mandibular and medial retropharyngeal lymph nodes, kidneys, spleen, and liver were not routinely performed, which may have caused cats to be understaged. Although there was no difference in the proportion of cats treated with prior administration of corticosteroids between both groups in the present study, future studies are needed in a population of cases with complete staging prior to receiving corticosteroids. Distant relapse was observed in 16 cats, 13 of whom (81%) had renal involvement, as has been recently reported.10 In contrast, older studies have reported that renal involvement of feline sinonasal lymphoma was rare.7,8 Although the cause for this difference in frequency across studies is unclear, a careful kidney evaluation should be performed at staging and follow-up. Cats receiving rescue therapy in the HRT group following lymphoma relapse had a poor prognosis with a median survival time of 103 days, which was similar to that reported previously.10
An optimal chemotherapy protocol for feline sinonasal lymphoma has not been established, and COP- or CHOP-based protocols are commonly used.12,20,21 In one study on feline localized sinonasal lymphoma, 41 out of 44 (93%) cats were treated with chemotherapy alone using mainly a COP-based protocol.12 The cats reportedly experienced some clinical improvement with a median OS of 140 days. This outcome suggests that it is difficult to control these tumors over a long term by chemotherapy alone, even though they are highly chemosensitive. In our study, cats in the HRTC group had a good prognosis, and most were treated by a CHOP-based protocol (14/18; 78%). However, the benefit of doxorubicin inclusion in this multiagent chemotherapy protocol remains controversial, and a recent study reported that radiotherapy was a risk factor for increased serum creatinine concentrations in cats treated with doxorubicin.22,23
Our study had several limitations associated with its retrospective design. First, we did not standardize the period between the end of radiotherapy and follow-up examinations. Second, the treatment protocols and the time to initiation of chemotherapy differed between cats. This is because they were decided upon by the treating clinician and depended on the owner’s request. Furthermore, the dose intensity of chemotherapy might have influenced the survival analysis. We did not attempt to evaluate the dose intensity with survival time. This can be attributed to the unavailability of data for several cases, which were commonly treated by the referring veterinarian. The above-mentioned limitations possibly introduced bias in the survival analysis. Thus, a future prospective study and multivariate analysis with a larger sample size would better confirm the survival benefits of chemotherapy.
Conclusion
Our results revealed that the early initiation of chemotherapy was a factor associated with PFS in cats with localized sinonasal lymphoma treated with HRT in a univariate analysis. Cats treated with HRT alone had a poorer outcome than those treated with HRTC, despite having received rescue chemotherapy for progressive disease. Based on our study results, the early initiation of additional chemotherapy may be recommended to improve the outcome for feline localized sinonasal lymphoma treated with HRT.
Kaplan-Meier survival curves comparing (A) PFS and (B) OS in cats with localized sinonasal lymphoma receiving HRT or HRTC. The median PFS for cats in the HRT and HRTC groups was 104 and 677 days, respectively. The median OS for cats in the HRT and HRTC groups was 263 and 983 days, respectively. Tick marks indicate censored cats. HRT, hypofractionated radiotherapy; HRTC, combined hypofractionated radiotherapy and chemotherapy; OS, overall survival; PFS, progression-free survival.
Contributor Notes
