Performance of severity parameters to detect chemotherapy-induced pain and distress in mice

Material and methods

Welfare assessment

The health status of each animal was monitored daily at the same time and prior to the performance of other interventions by two experienced unblinded female researchers. Alterations in body weight, general state, spontaneous behaviour and treatment specific parameters were documented. For each parameter, descriptions of different health states are allocated to a point grading system, where 0 points describe no alteration of the physiological state and ≥ 20 points highest severity and humane endpoint. The points in each category are summed up to a total score serving as a basis for further interventions defined in the score sheet (Table 1). It was defined as if scores of at least 15 points persist for more than 48 hours, the animals are removed from the experiment and euthanized.

OPEN IN VIEWER

Table 1.

Score sheet design.

Parameter	points
Body weight
Unaffected	0
Variation < 5%	1
Weight reduction 5–10%	5
Weight reduction 11–20%	10
Weight reduction > 20%	20
General state
Fur smooth and shiny, stoma clean, eyes clear and shiny	0
Fur faulty (reduced/ excessive body hygiene)	1
Fur lusterless, disheveled, scruffy stoma, eyes turbid, increased muscle tone	5
Dirty fur, sticky or moist stoma, abnormal posture, eyes turbid, increased muscle tone	10
Cramps, paralysis (trunk muscle, limbs), breath sounds, cold temperature	20
Spontaneous behaviour
Normal behaviour (e.g. sleep, reaction to blow and touch, curiosity, social contacts)	0
Slight variation to normal behaviour (e.g. reduced curiosity, reduced explorative behavior)	1
unusual behaviour (e.g. limited motor function, hyper kinetic, reduced mobility)	5
Self-isolating, lethargy, distinct hyper kinetic, stereotypy of behaviour, coordination disorder	10
Pain noises by grabbing, self-amputation (automutilation)	20
Treatment specific
Neurotoxicity (stagger, loss of sensitivity)	10
Anemia (pale/ cyanotic skin)	10
Superficial bleeding/ hematoma	10
Enlarged spleen/ lymph nodes	20
Soft stool	5
Persistent diarrhoea	20
Heavy infection (high body temperature, purulent secretory discharge)	20
Acute kidney failure (oedema, uraemia, lethargy, hyperventilation)	20
Degree of strain (DS)
DS0 = no strain
DS1 = minor strain: careful observation necessary	0
DS2 = moderate strain: consult veterinary, signs of pain: analgesia (800 mg/kg	1–9
BW metamizol/day in drinking water), careful observation (4 × day), notice to veterinary/PI)	10–20
DS3 = severe strain: consult animal protection commissary, veterinary support necessary,  termination of experiment, euthanasia	>20

Results

Cisplatin treatment

Cisplatin-treated mice decreased in body weight resulting in a score of 2.2 ± 2.5 (day 1) and 2.8 ± 3.6 (day 2). On day 3 body weight was decreased to approximately 10% of baseline and scored with 9.0 ± 2.2 points. Alterations of general health state could only be detected on day 3 when the fur showed a slightly scrubby appearance, and mice were scored with 1 point. All other parameters did not show alterations. The overall degree of strain defined by the total score could be allocated as a mild strain throughout the experiment (Figure 1(a)).

OPEN IN VIEWER

Figure 1.

Severity assessment after cisplatin treatment. (a) Cisplatin treatment resulted in a mild severity in mice. A reduction in body weight was the main parameter that could be scored. On the last day slight changes in the fur appearance became visible. • timepoint of chemotherapy injection. (b) Body weight changes after injection of cisplatin. Data presented as mean ±standard deviation, *p < 0.05, n = number of animals.

Blood analyses revealed a significant decrease in white blood cell count on day 3 (dF = 3, p = 0.005), whereas platelet number was significantly increased (dF = 3, p = 0.0163). Cisplatin had no effect on the number of red blood cells (Table 2). Serum analyses showed an elevation of liver associated enzymes (AST: dF = 3, p = 0.018; ALT: dF = 3, p = 0.0362). Kidney (urea: dF = 3, p = 0.0155 and creatinine: dF = 3, p = 0.0073) and heart (creatine kinase: dF = 3, p = 0.0077) associated parameters increased after cisplatin injection. (Table 2).

OPEN IN VIEWER

Table 2.

Blood count and serum parameters of control, doxorubicin-, busulfan- and cisplatin-treated mice. Cisplatin and busulfan decreased the number of white blood cells (WBC). The number of red blood cells (RBC) was significantly lower after doxorubicin and busulfan treatment. While cisplatin and doxorubicin increased the number of platelets, busulfan treatment resulted in a reduced platelet count. Cisplatin was the only chemotherapeutic agent that increased serum aspartate transaminase (AST), whereas alanine aminotransferase (ALT) was elevated after both cisplatin and doxorubicin treatment. Furthermore, cisplatin increased urea and creatinine values, whereas doxorubicin treatment only resulted in higher creatinine values. All chemotherapeutics increased serum creatine kinase. Data are presented as means ( ± standard deviations). *p < 0.05.

	Control	Cisplatin			Doxorubicin		Busulfan
		Day 1	Day 2	Day 3	Day 8	Day 15	Day 3	Day 5	Day 10
	n = 9	n = 3	n = 3	n = 5	n = 3	n = 3	n = 3	n = 3	n = 5
WBC	6.7	5.6	5.5	3.7*	9.2	6.7	4.6	4.6	2.2*
103/µl	(±1.4)	(±0.6)	(±1.4)	(±1)	(±2)	(±2.8)	(±1.5)	(±2.8)	(±0.9)
RBC	9.3	10.2	9.4	9.8	9.6	8.1*	10	6.8	7.7*
106/µl	(±0.3)	(±0.3)	(±0.4)	(±0.3)	(±0.1)	(±0.7)	(±0.1)	(±3.9)	(±0.5)
platelets	879	797	894	1147*	1011	1124*	876	1217	480*
103/µl	(±178)	(±20)	(±187)	(±95)	(±23)	(±45)	(±106)	(±283)	(±149)
AST	109	122	201*	184	95	97	61	62	95
U/l	(±31)	(±18)	(±85)	(±60)	(±31)	(±10)	(±1)	(±11.2)	(±34)
ALT	62	58	89*	64	90*	55	56	65	78
U/l	(±8)	(±2)	(±24)	(±21)	(±20)	(±5)	(±11)	(±23)	(±27)
Urea	8.9	6.3	13.9*	11.1	7.7	7	8.5	7.5	9.4
mmol/l	(±1.8)	(±0.2)	(±6.4)	(±1.4)	(±0.2)	(±0.6)	(±0.6)	(±2.2)	(±2.3)
Creatinine	13.5	16.7	31.5*	22.3	15.3	16*	10.3	15.3	16
µmol/l	(±1.8)	(±1.5)	(±18.1)	(±5.4)	(±1.2)	(±1)	(±0.6)	(±3.1)	(±2.9)
Creatine kinase	104	181	340*	257	110	210*	113	128	376*
U/l	(±43)	(±44)	(±156)	(±197)	(±29)	(±21)	(±22)	(±51)	(±204)

Doxorubicin treatment

Body weight of doxorubicin-treated mice was almost stable during the first 13 days of treatment and started to decrease from day 14 onwards, resulting in a decrease of approximately 11% and a scoring of 6.7 ± 5.8 points. According to the score sheet, doxorubicin only had a mild strain (Figure 2(a)).

OPEN IN VIEWER

Figure 2.

Severity assessment after doxorubicin treatment. (a) Doxorubicin treatment lead to a slight decrease in body weight resulting in a mild severity for the animals. Although the score sheet gives no indication, all mice died suddenly on day 16 and 17. • timepoint of chemotherapy injection. (b) Body weight changes after repetitive injection of doxorubicin. Data presented as mean ±standard deviation, *p < 0.05, n = number of animals.

Blood analyses showed no change in white blood cell count, whereas red blood cell count was significantly reduced at day 15 (dF = 2, p = 0.0005). In contrast, the number of platelets was significantly higher (dF = 2, p = 0.029) (Table 2). Serum analyses showed an elevation of ALT in mice sacrificed on day 8 (dF = 2, p = 0.0016), whereas no change could be observed on day 15. Creatinine (dF = 2, p = 0.0015) and creatine kinase (dF = 2, p = 0.0037) were significantly higher on day 15. All other parameters were unaltered (Table 2).

Although score sheets only showed slight deviations from physiological values, all mice died before the end of the experiment (two mice were found dead in their cage at day 16 and three on day 17). A dissection of animals did not show any abnormalities. However, an increase in serum creatinine and creatine kinase in animals sacrificed on day 15 indicated an impaired function of heart and kidneys.

Busulfan treatment

Busulfan slightly decreased body weight of mice with a weight loss of approximately 8% at day 10 resulting in a score of 4.7 ± 4.1 points. The fur appearance led to a short-term increase in the score of the general state between day 3 (5.0 ± 0.0 points) and day 5 (3.1 ± 2.6 points) and the total score indicated a mild strain on the mice during the experiment (Figure 3(a)).

OPEN IN VIEWER

Figure 3.

Severity assessment after busulfan treatment. (a) Busulfan treated mice showed the lowest degree of severity in this experiment with a low degree of weight loss and some abnormalities in the fur appearance • timepoint of chemotherapy injection, (b) Body weight changes after repetitive injection of busulfan. Data presented as mean ±standard deviation, *p < 0.05, n = number of animals.

Busulfan significantly reduced white (dF = 3; p = 0.0012) and red blood cell (dF = 3; p = 0.0287), and platelet counts (dF = 3; p = 0.0004) at the last time point (Table 2). Serum analysis did not show alterations except for an increased creatine kinase value in mice sacrificed on day 10 (Table 2).

Influence of imaging on animal welfare assessment

No differences in the actual severity using these measures could be detected between animals subjected to chemotherapy with PET-CT imaging and animals only receiving chemotherapeutics (Figure 4).

OPEN IN VIEWER

Figure 4.

Influence of imaging on severity assessment. A comparison of mice that received only chemotherapy and mice that received chemotherapy and additional imaging did not show any difference in welfare assessment. Data presented as mean ±standard deviation, *p < 0.05; ↑ timepoint of imaging.

Discussion

The EU directive 2010/63/EU prescribes a prospective and an actual severity classification of procedures applied on animals. ² In the present study, the prospective evaluation classified the procedures as moderate severity resulting from a longer-lasting impairment of welfare after injection of chemotherapy and repeated anaesthesia during imaging. Although imaging procedures are classified as mild severity, ¹ it is assumed that repeated anaesthesia, a necessity for immobilization of animals during imaging, could lead to a higher severity. ⁹ Nevertheless, a comparison of animals subjected to chemotherapeutics with those subjected to chemotherapeutics and longitudinal imaging revealed no differences in the actual severity during the experiment. This is in line with a recent publication, where the impact of repeated isoflurane anaesthesia on C57BL/6JRj mice induced a short-term mild distress in the immediate post-anaesthetic period. ⁹ Although all animals experienced only mild strain during the experiment the duration of up to 15 days assigns the procedure to a moderate severity as defined by Annex VIII of the EU directive. ¹ Therefore, the prospective severity classification of the present study was in line with the governmental requirements.

In the current study, body weight was the most sensitive parameter for detecting alterations in animals' well-being. Nevertheless, the current scoring systems assigns 1 point for any alteration in body weight below 5%, not considering normal fluctuations of weight due to defaecation or urination. Weight measurements of 25 healthy female BALB/cAnNRj mice over 6 days revealed a variation of 2.4 ± 1.8%. Therefore, the threshold for scoring changes in body weight should be adapted.

The body weight of busulfan-treated animals was slightly reduced, and changes in serum parameters were less pronounced in these animals, whereas a strong reduction in blood cell numbers could be observed that was in line with changes found in the literature. ¹⁰

In contrast, cisplatin-treated animals showed the strongest decrease in body weight and multiple changes in blood and serum parameters. A single high dose of cisplatin is reported in the literature to increase serum creatinine and blood urea nitrogen levels 3 days after injection. ¹¹

Although doxorubicin-treated mice showed the lowest alteration of body weight and blood and serum parameters, all mice unexpectedly died before the end of the experiment. The applied dose of 4 mg/kg BW doxorubicin weekly for 3 weeks was chosen as it is known to induce cardiotoxicity in mice, ¹² whereas no spontaneous deaths were reported in the literature with comparable dosages and treatment duration.^12–14 No alterations of the animals' well-being could be observed one day before the animals died that would motivate a more frequent monitoring. An elevation of creatine kinase in animals euthanized on that day for histological analyses indicated a possible impairment of the heart function. Furthermore, increased serum creatinine levels pointed to the development of a nephrotic syndrome, which was reported after doxorubicin injection. ¹⁵ However, it could not be figured out whether the parameters we used to assess well-being did not accurately reflected the degree of the impairment of the animals' health states, whether their health condition deteriorated too rapidly to be captured after the last monitoring, or whether the animals had an unknown previous illness that resulted in the unexpected death.

From the score sheet only body weight scoring could be identified as an objective parameter of the animal's health state. All other parameters rather failed to detect changes, especially in the mice that died spontaneously. This may be due to the fact that the selected experiment-specific parameters mostly represent humane endpoints. Therefore, there is a need for a better adaptation of the score sheet for rodent-specific strain.

In this study, the body-weight reduction could be a result of reduced food intake possibly caused by fatigue, loss of appetite or nausea, common side effects of chemotherapy. Measuring food and water intake could be a simple method to assess alterations in eating behaviour. ¹⁶ Another sign for nausea could be ingestion of non-nutritive substances like bedding or cage enrichment material and should be observed. ¹⁷ Furthermore, alterations in locomotion can be a result of nausea or fatigue. One method described to assess these changes is the voluntary wheel running test, where mice are housed in cages equipped with running wheels to assess activity parameters.^18,19 A reduction of time spent on the wheel ¹⁸ or distance travelled ¹⁹ could be observed after administration of chemotherapeutics. Although the before mentioned parameters can be assessed easily without disturbance of the animals, a drawback is the fact that animals need to be single housed to detect changes in an individual animal. The impact of single versus group housing of mice on their well-being is still an ongoing discussion.^20,21 Furthermore, voluntary wheel running, a model of exercise in mice, could also alter the amount of chemotherapy-related measurable fatigue, as exercise is described to be beneficial for humans suffering from cancer related fatigue.^19,22

Another option to assess locomotion is the open field test, where mice are placed in a box of defined size and a camera located above the arena records the animals’ behaviour for a defined time. Outcome measures can give locomotion information of the animal, like travelled distance in a defined time and information about anxiety related behaviour detected by time spend in a certain area. ²³ It has been reported that mice showed reduced activity and a reduced time spent in the outer sections after chemotherapy. ²³ Although the open field test is an additional procedure, the severity of this procedure is defined by Annex VIII of EU directive 2010/63/EU as ‘below the threshold for regulation’. ¹

In conclusion, body weight was the main parameter detecting physiological changes in mice during chemotherapy. Regarding side effects of chemotherapy, more focus should be set on signs of nausea and fatigue as body-weight reduction could be a result of these signs. Useful measures to assess nausea and fatigue could be measuring food and water intake. Furthermore, changes in locomotion can be detected by voluntary wheel running or open field test, with the latter additionally enabling the assessment of anxiety relating behaviour.