KNOWLEDGE SUMMARY

Keywords: CAT; CHEMOTHERAPY; FELINE; MAMMARY CARCINOMA; MASTECTOMY; SURVIVAL TIME

In cats with mammary carcinomas does the addition of chemotherapy compared with surgery alone result in increased survival time?

Gabriela Gonzalez-Ormerod, DVM PGDip (VCP) MRCVS^1*

¹ Langford Veterinary Services, United Kingdom
* Corresponding author email: gg1161@alumni.bristol.ac.uk

Vol 9, Issue 4 (2024)
Submitted: 03 August 2023; published: 18 Nov 2024; next review: 12 Apr 2026
DOI: https://doi.org/10.18849/ve.v9i4.688

PICO question

In cats with mammary carcinomas undergoing surgical removal, does the addition of adjuvant chemotherapy compared with surgical removal alone result in increased survival time?

Clinical bottom line

Category of research

Treatment.

Number and type of study designs reviewed

Five retrospective cohort studies were critically reviewed.

Strength of evidence

Weak.

Outcomes reported

In cats with mammary carcinomas undergoing surgical removal, the addition of adjuvant chemotherapy compared with surgical removal alone was significantly associated with an increase in disease-specific survival time in one of the studies. This statistical significance was not found in the other four papers.

Conclusion

All five studies reviewed presented weak evidence for the clinical question due to their retrospective nature and weak study design. It is, therefore, concluded that there is not enough evidence to suggest that cats with mammary carcinomas that have undergone surgical treatment will have a longer survival time if treated with adjuvant chemotherapy. More prospective, placebo-controlled, double-blinded studies are needed to understand the clinical significance of adjuvant chemotherapy in cats with mammary carcinomas.

How to apply this evidence in practice

The application of evidence into practice should take into account multiple factors, not limited to: individual clinical expertise, patient’s circumstances and owners’ values, country, location or clinic where you work, the individual case in front of you, the availability of therapies and resources.

Knowledge Summaries are a resource to help reinforce or inform decision making. They do not override the responsibility or judgement of the practitioner to do what is best for the animal in their care.

Clinical scenario

A 10-year-old female neutered Domestic Shorthair cat “Candy” is brought to your clinic because her owners noticed a mass on her abdomen. The owners report that she is otherwise healthy and is eating, drinking, urinating and defecating normally. On physical examination you palpate a well-defined 4 cm mildly ulcerated mass associated with Candy’s right inguinal mammary gland.  The rest of her mammary glands are normal on palpation. You take fine-needle aspirates of the mass and perform full bloods (haematology and biochemistry), urine analysis, abdominal ultrasonography and thoracic radiographs. Your workup confirms that Candy has a mammary adenocarcinoma and there is cytological evidence of spread to her inguinal lymph nodes without any evidence of distant metastasis. Candy has a stage III mammary carcinoma according to the modified World Health Organization (WHO) staging system. Her owners consent to a staged bilateral mastectomy but you have had a difficult conversation with them as stage III–IV mammary tumours are associated with a poorer prognosis and survival times despite surgery. Candy’s owners want to do absolutely everything to help prolong her life expectancy and have asked if adjuvant chemotherapy is an option. Is there any evidence to suggest that surgical treatment of mammary carcinomas plus adjuvant chemotherapy results in increased survival times?

The evidence

The literature search revealed five retrospective cohort studies relevant to the PICO question (De Campos et al. (2014); Gemignani et al. (2018); Ito et al. (1996); McNeill et al. (2009); Petrucci et al. (2021)). All five studies presented weak evidence for the clinical question largely due to the fact that their retrospective nature meant that there was variability in patient’s disease stage, a lack of standardisation of surgical and chemotherapeutic protocols as well as the absence of placebo groups. There were also noticeable limitations when retrospectively determining disease-specific causes of death and survival times. Furthermore, there is limited applicability of the results in first opinion practices given that most of the data was obtained from referral teaching hospitals and/or specialised private practices outside of the UK.

Summary of the evidence

De Campos et al. (2014)

 

Gemignani et al. (2018)

 

Ito et al. (1996)

 

McNeill et al. (2009)

 

Petrucci et al. (2021)

 

Appraisal, application and reflection

Mammary neoplasms are the third most frequent tumour in cats and mainly affect female cats between the ages of 10–12 years (De Campos et al., 2014). Approximately 80–90% of feline mammary neoplasms are malignant carcinomas and are characterised by local invasion of surrounding tissues and a high rate of metastasis to local lymph nodes, lungs, and other sites (McNeill et al., 2009). Feline mammary carcinomas (FMCs) are classified into four stages (I–IV) based on the modified World Health Organization (WHO) clinical staging system. Higher stages have a poorer prognosis and shorter survival times (Petrucci et al., 2021; Gemignani et al., 2018, and Ito et al., 1996). FMCs are classified based on primary tumour size, lymph node involvement and presence of distant metastasis (McNeill et al., 2009). A histological grade (I–III) can also be used to further classify these neoplasms (De Campos et al., 2014). The current treatment of choice is radical bilateral or unilateral mastectomy. However, despite aggressive surgical treatment median survival times (ST) after surgery in stage III–IV FMCs is less than one year (Lo et al., 2019). Despite the poor prognosis of high-grade FMCs with radical surgery alone, there are only a small number of studies that explore the clinical benefits of using adjuvant chemotherapy.

The literature search for the PICO question revealed five relevant articles, all of which were observational and retrospective. Gemignani et al. (2018) is a multicentre study that includes data from referral university hospitals and private surgically specialised hospitals. The data for De Campos et al. (2014) and Ito et al. (1996) studies were obtained from one referral teaching hospital in Brazil and Japan, respectively. Petrucci et al. (2021) and McNeill et al. (2009) were the only studies that included data from referral teaching hospitals and private veterinary hospitals. Crucially, the results obtained may not be applicable to first opinion clinical practice given that most of the studies were performed in referral or specialist level hospitals. Surgical technique and surgeon experience could affect surgical outcome and survival times. In this instance, chemotherapy may play a bigger role with less experienced surgeons in first opinion practices where complete surgical excision may not be achieved. Interestingly, however, in McNeill et al. (2009) study 59% (43/73) of surgeries were performed in general practice and there were no statistically significant differences found between treatment groups in regard to the type of hospital where surgery was performed. It is also worth noting that none of the studies were performed in the United Kingdom and therefore the results may not be applicable to a population of British cats. The prevalence of different cat breeds within a feline population could have an effect on the outcomes studied. Hayes et al. (1981) found that Siamese cats had twice the risk of developing mammary carcinomas compared to other breeds. This breed was overrepresented in Ito et al. (1996) cohort group and could have had an effect on this study’s overall survival times.

All five papers appraised use the modified WHO staging system in order to stage their cohort groups. However, given the retrospective nature of all of the papers appraised staging was not standardised. In Petrucci et al. (2021) study all cats had pre-surgical thoracic radiographs and an abdominal ultrasound or a full body Computed Tomography (CT). In McNeill et al. (2009) paper 92% (67/73) of cats had pre-operative thoracic radiographs but only 32% (23/73) had an abdominal ultrasound. Moreover, if local lymph nodes were palpably normal or not removed during surgery these were deemed to be normal and therefore the cat was automatically classified as having a stage I FMC. Gemignani et al. (2018), De Campos et al. (2014), and Ito et al. (1996) make reference to the WHO staging system. However, there is no further information on how many cats had pre-surgical imaging or whether local lymph nodes were sampled for cytology. The lack of a standardised pre-operative staging system could result in local and distant metastases being missed and the incorrect classification of cats into lower FMC stages. This potentially underestimates the effect of adjuvant chemotherapy on low stage FMCs and could inadvertently include cats with stage IV FMCs in these studies (De Campos et al. (2014), Gemignani et al. (2018), and Ito et al. (1996)). In both instances this would result in a reduction in STs in all cohort groups. Conversely, Matos et al. (2012) further argue that some nodules that are not sampled and assumed local or distant metastasis could actually be benign or even different neoplasms. Therefore, any future prospective studies on FMCs need to have a standardised pre-surgical staging protocol (including sampling of local lymph nodes and suspected distant metastasis). Unfortunately, a cohort group with these specifications is difficult to obtain in a retrospective and multicentre study.

Cats with stage I, II, and III FMCs were included in four of the papers appraised (Gemignani et al., 2018; Ito et al., 1996; McNeill et al., 2009; Petrucci et al., 2021), though most of the cats included in McNeill et al. (2009) were stage I–II. De Campos et al. (2014) is the only paper that exclusively considers cats with stage III FMCs. None of the papers include cats with known distant metastasis (stage IV). In those papers that include cats in different stages it is difficult to ascertain whether STs are influenced by severity of disease. Both Petrucci et al. (2021) and Gemignani et al. (2018) concluded that higher stages (stages III–IV) have lower STs compared to stages I–II. In Petrucci et al. (2021) multivariate analysis cats in stage III had 2.4 times higher risk of recurrence and 3.1 times higher risk of death compared to cats in stage I (adjusting for other variables). Therefore, stage is an important variable in determining STs and should be taken into account in any statistical analysis of FMCs. This is especially relevant in those studies where a multivariate analysis is not performed and STs are considered independently of variables such as lymph node metastasis at the time of surgery, development of regional or distant metastasis or type of mastectomy (McNeill et al., 2009 and Ito et al., 1996).

Any future prospective study should look at the effect of adjuvant chemotherapy on STs in populations of cats with the same tumour stage. Crucially, this would give us a better understanding as to whether the use of adjuvant chemotherapy is most useful in cats with advanced or low-stage FMCs. McNeill et al. (2009) suggest that cats with stage I or II FMCs already have a good prognosis with radical surgery alone so using adjuvant chemotherapy may be less justified, especially given the potential side-affects as well as the economic and emotional commitment required from owners associated to this additional therapy. McNeill et al. (2009) therefore argue that adjuvant chemotherapy plays a bigger role in prolonging ST in stage III FMCs. Alternatively, Borrego et al. (2009) argue that staging has its limitations and one cannot underestimate microscopic metastasis that may go undiagnosed during initial staging. In their study they found that some cats with stage I-II FMCs that had complete margins post-operatively later developed local and/or distant metastasis. This suggests that a multimodal treatment protocol including radical surgery and adjuvant chemotherapy should be implemented in all cats with FMCs, regardless of stage, in order to achieve maximum STs. However, Borrego et al. (2009) study is also retrospective and therefore lacks standardisation of pre-surgical staging. The prognostic significance of micrometastasis in FMCs, therefore, remains uncertain.

Type of surgery has also been shown to be significantly associated to STs in FMCs, with radical mastectomies resulting in prolonged disease-free intervals compared to lumpectomies (Gemignani et al., 2018, Petrucci et al., 2021). Petrucci et al. (2021) and Gemignani et al. (2018) only include cats that have had radical unilateral or bilateral mastectomies. De Campos et al. (2014), McNeill et al. (2009), and Ito et al. (1996), however, include cats that have had local as well as radical unilateral or bilateral mastectomies. In those papers that include cats that had both local and radical surgery STs could vary as a consequence of the type of surgery performed. Interestingly, McNeill et al. (2009) found that in those cats treated with unilateral mastectomies STs were significantly longer in the surgery plus adjuvant chemotherapy group. This statistical significance was not found when type of surgery was not taken into account. This is a surprising result given that Gemignani et al. (2018) argue that bilateral radical mastectomies are protective compared to unilateral mastectomies. However, one should interpret McNeill et al. (2009) results cautiously given that the population of cats that had unilateral mastectomies (n = 18/73) was very small and likely underpowered. Moreover, due to the fact that a multivariate statistical analysis was not performed it is difficult to evaluate the effect that other variables (such as surgeon experience, lymph node metastasis, or histologic grade) had on STs in this group. None of the papers that include lumpectomies in their surgical protocol perform a multivariate analysis, so it is impossible to ascertain whether type of surgery had an effect on STs in these studies (De Campos et al. (2014), Ito et al. (1996) and McNeill et al. (2009)).

The chemotherapy protocols in all studies were also clinician dependent and therefore very variable. For example, in Petrucci et al. (2021) paper not all cats in group 2 received the same number of chemotherapy cycles, while the cats in McNeill et al. (2009) study received variable doses and number of cycles as part of their chemotherapy protocol. Furthermore, the chemotherapy agents, doses and/or number of cycles vary between papers and therefore comparing results between studies is impossible. The chemotherapeutic agents used range from doxorubicin, cyclophosphamide, carboplatin, and vincristine but the predominant chemotherapeutic agent in most papers is doxorubicin (Gemignani et al., 2018; McNeill et al., 2009 and Petrucci et al., 2021). However, the dose and number of doxorubicin cycles used vary between papers and within the same studies. In McNeill et al. study (2009) 81% (29/36) of cats in the group received more than or equal to 3 cycles of treatment, while 67% (24/36) received 4 or more cycles. Novosad et al. (2006) and Borrego et al. (2009) argue that the number of doxorubicin cycles used is significantly associated with disease free intervals (DFI). They argue that cats that received fewer than 4 cycles of doxorubicin had a statistically lower DFI than those cats that received 4 or 5 cycles. Therefore, the STs in McNeill et al. (2009) could potentially be longer if all cats had received an adequate number of doxorubicin cycles. This may be complicated in practice given that owners and/or veterinarians may elect to discontinue chemotherapy treatment if considerable side-effects are noted or due to monetary constraints.

In addition, there is also a strong bias in all the studies appraised, given that the decision to perform adjuvant chemotherapy is left to owners and clinicians. One could argue that the owners that elect more aggressive treatment (i.e. chemotherapy) are more likely to perform more frequent monitoring (and detect progression of disease and other comorbidities sooner) and less likely to elect cessation of treatment which could result in prolongation of STs. Conversely, clinicians may elect to use chemotherapy in those cases with more advanced FMC stages where surgical treatment does not achieve a long enough ST on its own. Therefore, selecting for cases that have an intrinsically poorer ST and consequently reducing the STs in the adjuvant chemotherapy groups. It is important that future prospective studies are double-blinded and randomised in order to minimise owner and clinician bias.

All the papers appraised use median survival time as a studied outcome. Gemignani et al. (2018), Ito et al. (1996), McNeill et al. (2009) and Petrucci et al. (2021) further include DFI (also described as disease free survival, remission length and progression free interval) as a studied outcome. In a population of cats with cancer DFI is a more useful outcome to study. Using survival times as a measurable outcome has multiple limitations. Firstly, STs carry a strong owner bias as it is dependent on owners electing to stop treatment or euthanise based on perception of reduced quality of life, side-effects of treatment or monetary constraints. Ito et al. (1996), McNeill et al. (2009) and Petrucci et al. (2021) study overall survival times with death by any cause. This is not an adequate measurable outcome for the clinical question at hand given that many of the deaths documented in their studies could be due to progression of other comorbidities. This is especially relevant in the cohort groups for all of the studies appraised given that most cats were geriatric at the time of diagnosis. In De Campos et al. (2014) and Gemignani et al. (2018) disease specific survival times are measured. However, given the retrospective nature of these studies it is also difficult to unequivocally determine the cause of death of the cats. Most of the studies rely on clinical notes and/or owner phone calls in order to determine cause of death. None of the studies include an owner questionnaire template and therefore one cannot assess if the questions asked were biased. In any future perspective studies on FMCs, sampling tissue that causes suspicion for local or distant metastases and performing post-mortems with histopathology on all cats may be crucial to determining disease progression and cause of death.

Furthermore, measuring DFI does not come without limitations as it is very much influenced by the ability of disease progression to be noticed by owners and/or clinicians. Some owners may be more aware than others of the clinical signs (i.e. coughing, exercise intolerance) that should prompt a veterinary visit, while others may be more comfortable in periodically palpating their cats’ mammary glands in order to monitor for progression. Owner compliance, available funds, and owner availability may strongly influence how regularly a patient is seen by a veterinary surgeon. Moreover, disease progression is not confirmed using cytology or histopathology in all of the cats in the studies appraised. In Petrucci et al. (2021) only 40.3% (29/72) of local or distant metastasis are confirmed with cytology, 30.5% (22/72) had confirmed metastasis by histological evaluation post-mortem and 29.2% (21/72) of metastasis were assumed using imaging alone. Finally, in Petrucci et al. (2021)s, cats receiving adjuvant chemotherapy were seen every 3 weeks as opposed to cats that only had surgery who were seen at least every 3 months. More regular veterinary visits in the chemotherapy groups could increase detection of disease progression. This could decrease DFIs and potentially also reduce STs given that some owners may elect to euthanise sooner if there is evidence of disease progression despite use of chemotherapy.

The literature search for the PICO question showed that there are very few studies on the use of adjuvant chemotherapy in FMCs and none of them offer enough strength of evidence to conclude whether it makes a difference to survival times. Only Gemignani et al. (2018) found that the use of chemotherapy had a statistically significant effect on prolonging disease specific ST and therefore was considered protective in the multivariate analysis. However, similar to the other studies appraised, it does not have enough strength of evidence to suggest that cats with mammary carcinomas that have undergone surgical treatment will have a longer survival time if treated with adjuvant chemotherapy. Larger prospective, double-blinded and randomised studies are needed to properly evaluate the clinical relevance of adjuvant chemotherapy in feline mammary carcinomas.

Methodology

Search strategy

 

Exclusion / inclusion criteria

 

Search outcome

 

ORCiD

Gabriela Gonzalez-Ormerod: https://orcid.org/0000-0001-5043-1069

Conflict of interest

The author declares no conflicts of interest.

References

1. Borrego, J.F., Cartagena, J.C. & Engel, J. (2009). Treatment of feline mammary tumours using chemotherapy, surgery and a COX-2 inhibitor drug (meloxicam): a retrospective study of 23 cases (2002–2007). Veterinary and  Comparative Oncology. 7(4) 213–221. DOI: https://doi.org/10.1111/j.1476-5829.2009.00194.x
2. De Campos, C.B., Nunes, F.C., Lavalle, G.E. & Cassali, G.D. (2014). Use of Surgery and Carboplatin in Feline Malignant Mammary Gland Neoplasms with Advanced Clinical Staging. In Vivo. 28(5), 863–866.
3. Gemignani, F., Mayhew, P.D., Giuffrida, M.A., Palaigos, J., Runge, J.J., Holt, D.E., Robertson, N.A., Seguin, B., Walker, M., Singh, A., Liptak, J.M., Romanelli, G., Martano, M., Boston, S.E., Lux, C., Busetto, R., Culp, W.T.N., Skorupski, K.A. & Burton, J.H. (2018). Association of surgical approach with complication rate, progression-free survival time, and disease-specific survival time in cats with mammary adenocarcinoma: 107 cases (1991–2014). Journal of the American Veterinary Medical Association. 252(11), 1393–1402. DOI: https://doi.org/10.2460/javma.252.11.1393
4. Hayes, H.M., Milne, K.L. & Mandell, C.P. (1981). Epidemiological features of feline mammary carcinoma. Veterinary Record. 108(22), 476–479. DOI: https://doi.org/10.1136/vr.108.22.476
5. Ito, T., Kadosawa, T., Mochizuki, M., Matsunaga, S., Nishimura, R. & Sasaki, N. (1996). Prognosis of Malignant Mammary Tumor in 53 Cats. Journal of Veterinary Medical Science. 58(8), 723–726. DOI: https://doi.org/10.1292/jvms.58.723
6. Lo, T-Y., Feng, Y-C., Yang, Y-P., Liao, J-W. (2019) With surgical removal or adjuvant chemotherapy increase the two-year survival and associated clinicopathologic factors in cats with mammary carcinoma. Taiwan Veterinary Journal. 45(03), 1–10. DOI: https://doi.org/10.1142/S1682648519500021
7. Matos, A.J.F., Baptista, C.S., Gärtner, M.F. & Rutteman, G.R. (2012). Prognostic studies of canine and feline mammary tumours: The need for standardized procedures. The Veterinary Journal. 193(1), 24–31. DOI: https://doi.org/10.1016/j.tvjl.2011.12.019
8. McNeill, C.J., Sorenmo, K.U., Shofer, F.S., Gibeon, L., Durham, A.C., Barber, L.G., Baez, J.L. & Overley, B. (2009). Evaluation of Adjuvant Doxorubicin-based Chemotherapy for the Treatment of Feline Mammary Carcinoma. Journal of Veterinary Internal Medicine. 23(1), 123–129. DOI: https://doi.org/10.1111/j.1939-1676.2008.0244.x
9. Novosad, C.A., Bergman, P.J., O'Brien, M.G., McKnight, J.A., Charney, S.C., Selting, K.A., Graham, J.C., Correa, S.S., Rosenberg, M.P. & Gieger, T.L. (2006). Retrospective evaluation of adjunctive doxorubicin for the treatment of feline mammary gland adenocarcinoma: 67 cases. Journal of the American Animal Hospital Association. 42(2), 110–120. DOI: https://doi.org/10.5326/0420110
10. Petrucci, G.N., Henriques, J., Lobo, L., Vilhena, H., Figueira, A.C., Canadas-Sousa, A., Dias-Pereira, P., Prada, J., Pires, I. & Queiroga, F.L.  (2021). Adjuvant doxorubicin vs metronomic cyclophosphamide and meloxicam vs surgery alone for cats with mammary carcinomas: A retrospective study of 137 cases. Veterinary and Comparative Oncology. 19(4), 714–723. DOI: https://doi.org/10.1111/vco.12660