Oxaliplatin induced peripheral neuropathy and associated factors among colorectal cancer patients in Tanzania.

Introduction
Globally, Colorectal cancer is among the most frequently occurring malignancy after breast, lung and prostate cancer [1]⁠. The incidence of colorectal cancer is high in both developed and developing worlds but mortality is more pronounced in the later. In East Africa, the data from Globocan 2020 showed 18,306 new cases and 13,236 deaths, whereas in Tanzania, colorectal cancer was ranked fourth with approximately 2354 new cases in 2020 and an estimated mortality of 1555 cases in the same year [1].
The management of colorectal cancer ranges from surgery, chemotherapy and radiotherapy which could be applied singly or in combination. Although colorectal cancer has a long preclinical progression which gives room to early diagnosis and so less aggressive treatment, most of the patients present to the hospital at advanced stage due to the nature of the disease in that it has less early symptoms making most patient advance in stage and definitely be candidates for chemotherapy [2, 3].
Discovery and licensing of oxaliplatin as an essential drug used in management of patients with colorectal cancer is one of major advances that changed the management and eventual outcome of patients with colorectal cancer. It started off being included in the treatment regimen for patients with metastatic disease after it showed promising results. It was later included as a pivotal mainstay drug used even in non-metastatic cancer patients. Due to the good treatment response and combination with other drugs, patients are now living longer than earlier anticipated even though there are toxicities associated with its use.
Peripheral neuropathy is one of the most debilitating and dose limiting toxicity of Oxaliplatin and this has been widely articulated in literature to occur in almost 85%-95% of the patients. Oxaliplatin induced peripheral neuropathy (OIPN) could be featured into two discrete forms, acute type which is short-lived and chiefly occurs in sensory form and might be triggered by cold exposure. The acute form has swift inception and persists briefly from hours to few days after treatment and is majorly reversible [4]. The other is the chronic type, which is characterized by steady and progressive onset with the chance of occurrence increasing with the cumulative dose of oxaliplatin [5]. Furthermore, the chronic type is also sensory in nature and is accompanied by sensory paresthesia’s and dysesthesias which could impair day-to-day activities and quality of life of patients [6–11].
No current management is available for OIPN and the only accessible medications are for symptoms mitigations to the elements of neuropathic pain [12, 13]. Among multiple methods that have been investigated to manage OIPN, duloxetine was found useful in management of oxaliplatin-associated neuropathic pain. Other medicines that have demonstrated to be effective include gabapentin, lamotrigine, and baclofen [14]. With OIPN reversibility in the majority of patients, several facilities have used a stop-and-go strategy to try and minimize the side effect. Two trials showed that the prevalence of neuropathy was less if the intermittent administration is given and hence minimize toxicity in patients who manifest with OIPN [15]. At ORCI, for symptom management, gabapentin and vitamin B supplements are prescribed to curb the effect of neuropathy. Management of colorectal cancer is in two folds. For colon cancer the treatment consists of adjuvant chemotherapy which include FOLFOX regimen and for metastatic stage, FOLFOX or CAPOX regimen is used. For rectal cancer, neo-adjuvant chemo-radiotherapy is used and consist of 5 fluorouracil plus leucovorin and radiotherapy followed by adjuvant chemotherapy. Modified FOLFOX 6 or 7 is used with or without bevacizumab in metastatic stage [16].
Despite the fact that oxaliplatin is widely used in our local setting to manage patients with colorectal cancer, there is scarcity of information about OIPN and its impact on patients at ORCI, Tanzania and sub-Saharan Africa at large. This shed the light on OIPN as adverse effect of oxaliplatin in treatment of CRC at ORCI and the information will be used for additional guideline development and in identifying at risk patients for available prophylactic and supportive measures in the management of patients with CRC at ocean road and Tanzania at large.

Methodology

Study design and subjects
This retrospective cohort study retrieved all consecutive patients confirmed with CRC who were on oxaliplatin-based regimen, given at least six cycles, aged 18 years or older and treated at ORCI from December 2021 to May 2022. Patients who have received other platinum/taxane-based chemotherapy previously and with pre-existing peripheral neuropathy prior to oxaliplatin therapy were excluded. The documents of all patients who met the inclusion criteria were retrieved from hospital records computer system and followed from time of initiation of oxaliplatin-based chemotherapy to maximum of six months post treatment. Phone calls were used to contact patients for enquiries about needed details that were missing from their files; questionnaires were administered over the phone by trained research assistants.
Social demographic and clinical information of the patients was extracted from the patients’ medical charts whereas other information regarding social demographic characteristics of the patients was supplemented by interviewing the patients. Questionnaires were used to interview and collect data. The questionnaire consisted of social demographic, clinical profile of the patients and adapted questionnaires. In this study, two sets of questionnaires were adapted and used to screen for OIPN and quantify the severity. Prevalence of acute OIPN was assessed using descriptive questionnaire that was previously used in other studies, Nadeen T. Ali et al. [17]
The questionnaire consists of 11 questions (appendix 1) that assess the temporary symptoms of peripheral neuropathy and the severity of OIPN was graded as the sum of number of symptoms. If the patients have 1–3 symptoms, the severity of acute OIPN was graded 1, if the sum of the symptoms is present between 4 and 6 then the severity was Grade 2 and if is 7–9 symptoms will be present then it was graded 3 and above 9 symptoms was graded as 4 [17].
EORT-CIPN20 questionnaire and Common terminology criteria for adverse events version 4.03 was used to screen and grade chronic OIPN respectively. This questionnaire has been used to assess chemotherapy induced peripheral neuropathy (CIPN) in many studies. It consists of sensory, motor and autonomic neurotoxicity subscale consisting of twenty questions items. Each item is measured on a Likert scale ranging from 1) Not at all to 4) Very much [17]. The QLQ-CIPN20 was assessed using a simple additive checklist that yielded a single, mean total score. The raw mean of the 18 items (excluding the two conditional items; difficulty using the pedals and difficulty getting or maintaining an erection) was computed and linearly converted to a 0-100 scale, as per EORTC scoring methods for symptom scales. Patients with higher score indicated a higher level of symptom burden [18]. Grading was done using Common terminology criteria for adverse events version 4.03 which has been broadly used to quantify the severity of chemotherapy induced neurotoxicity [15].
Muhimbili University of Health and Allied Sciences ethical review board (approval number: - MUHAS-REC-12-2021-913) provided ethical clearance and Permission to conduct the study at ORCI was sought from institutional ethical committee. Informed consent to participate in this study was obtained from all of the participants. This study was conducted in accordance with the Declaration of Helsinki.

Statistical analysis
Statistical Package for Social Science (SPSS v.23) was used in data analysis. Clinical and demographic data were analyzed using measures of descriptive statistics such frequency, percentage, mean and standard deviation. For Association between independent and dependent variable, univariate analysis (chi square) and multivariate analysis (logistic regression) was done.

Results

Characteristics of study subject and severity of peripheral neuropathy
A total of 73 subjects with a diagnosis of anal CRC were retrieved from December 2021 to May 2022; however, 62 subjects (84.9%) met the study inclusion criteria. The age ranged from 19 to 77 years with a mean of 52.47 ± 14.05. More than half (59.7%) of patients were male and 28 (40.3%) were female with male to female ratio of 1.5:1. Based on the comorbidity status it was found that 22.6% of the patients were hypertensive, 6.5% diabetic and 6.5% seroconverted (HIV). In this study 58.1% of the patient were prescribed CAPEOX regimen while 41.9% used FOLFOX, of which 56.4% got more than seven cycles of chemotherapy. The prevalence of acute symptoms of OIPN was 71% and chronic OIPN symptoms were 59.7% of colorectal cancer patients who were treated at ORCI during the study period (Table 1).

The proportion of patients who had acute OIPN symptoms was also examined among CRC patient during chemotherapy treatment. Among the symptoms, throat discomfort (77.42%) and tingling in mouth (82.26%) and jaw pain (87.10%) had highest prevalence. However, there were 32, 45 patients who respectively experienced acute symptoms such as burning in eyes and feeling shock down the back. Minority of CRC patients had problem with breathing (9.68%) among others (Table 2). On grading of acute OIPN, majority of the patients had Grade 2 in 43 (69.4%), Grade 1 were 11 (17.7%) patients, Grade 3 were 5 (8.1%) patients with grade 4 in 3 patients (4. 8%) (Fig. 1). For the assessment of severity for Chronic OIPN among colorectal cancer patient n = 37 (59.7%); maximum grade was grade 3 in 5 (13.5%) patients, grade 2 in 9 (24.3%) and grade 1 in 23 (62.2%) patients (Fig. 2).

Responses to the single item subscales of the QLQ-CIPN20 is shown in table. Among the symptoms assessed, the three most prevalent were, tingling sensation in fingers or hands (n = 5), tingling sensation of toes or feet (n = 2) and trouble opening jar (n = 2) which had high proportion of patients responding to very much on the Linkert scale. The study also revealed that 1.6% and below individuals could either experience one of the remaining symptoms or none at a relatively higher (very much) extent. Most other patient responded in great numbers to having less symptoms (a little) on the scale. 33.9% of the patient reported to have symptoms of feeling dizziness after standing up when compared to those who didn’t feel any symptoms of dizziness (Table 3).

Association of chronic OIPN and baseline variables
Chi-square test of independence was performed. Cumulative dose, BMI and number of cycles were the variables that were significantly associated with OIPN toxic effect in patients with colorectal cancer (p<0.05). Patients who receive relatively high cumulative dose (> 780mg/m2) seem to be more likely to develop OIPN toxic symptoms (83.33%) compared to individuals who receive doses ≤ 780mg/m2 (44.74%) with p<0.05. The analysis further showed that the number of cycles the patients received attributes to OIPN chronic toxicity. Similar pattern was observed for the case of BMI categories, there were more obese patients who developed OIPN toxicity (85.71%), followed by over weight (72.73%) compared to their respective categories (14.29% and 27.27%) who did not experience OIPN toxicity (Table 4).

Results indicated that the model was statistically significant finding, χ2 (6, N = 62) = 26.53, p = 0.001. The correct prediction rate was about 82.3%. The Wald tests showed that all the variables significantly predicted the dependent variable (OIPN toxicity). This inferred that number of cycles, cumulative dose and BMI were risk factors of OIPN chronic toxicity. However, hypertension was not associated to OIPN chronic toxicity among patients with colorectal malignancies.
The adjusted odds ratio of developing OIPN among patients who received ≤ 7 cycles was 0.227 less compared to those who received more than 7 cycles, meaning that patients who receive less cycles are at lower risk of developing OIPN than at those who receive 7 and more cycles (OR = 0.227, 95% CI (0.053–0.971), p = 0.046). In assessing relationship between BMI of the patients with colorectal cancer and OIPN toxicity after chemotherapy treatment, logistic regression test results revealed that BMI was a significant predictor of chronic OIPN toxicity (p<0.05). Furthermore, the model also revealed that colorectal cancer patients who are obese were 21.66 times higher at risk of suffering from chronic OIPN compared to individuals who have normal body weight (OR = 21.66, 95% CI (2.574, 182.244), p < 0.05). On the other end from this study, it was found that Cumulative Dose showed predictive effect on OIPN toxicity (OR = 0.104, 95% CI (0.022–0.502), p = 0.005). For instance, during chemotherapy treatment patients who receive lower doses (≤ 780mg/m2), their odds of developing chronic OIPN toxicity were lower than patients who received higher dose (> 780mg/m2) (Table 5).

Discussion
Colorectal cancer cases have been on the rise in sub-Saharan Africa in the past few years [1], making the use of oxaliplatin-based chemotherapy regimen as the mainstay in management of these cancer to increase. Oxaliplatin induced acute and chronic neuropathy is one of the doses limiting toxicities in management of CRC patients and to our best knowledge this is the first study in sub–Saharan Africa to assess associated factors and severity of this side effect.
The finding from our study demonstrated that the prevalence of acute and chronic OIPN was 71.0% and 59.7% respectively among our patients. The mean age of the patients was 52.47 with standard deviation ± 14.05 years and most of the patients had stage 3 and 4 diseases, which can be attributed to late presentation due to asymptomatic early stages of colorectal cancer. Majority of the patients developed grade 1 (17.7%) and 2 (69.4%) neuropathies in both acute and chronic cases. Most patients reported symptoms of tingling in mouth (82.26%), throat discomfort (77.42%) and jaw pain (87.10%) for acute symptoms. The most prevalent symptoms reported by the patients in chronic OIPN was tingling of fingers (8.1%) and toes (3.2%) in its most severe form. This finding demonstrate a favorable comparison with many other studies done internationally [19–21]. However, various factors such as study design, difference in study populations, and assessment tools used restrict comparisons between OIPN studies.
We also assessed the association of OIPN with age, Gender, number of cycles that the patient received, co-morbidities, Drug regime, Cancer stage and cumulative dose among others. The multivariate analysis of the results generally inferred that number of cycles, cumulative dose and BMI were independent risk factors of OIPN. This is in comparison with a systematic review study done by Jeremy. N .pulvers et al. [19]. This study did not observe a statistically significant difference in OIPN between patients treated with CAPEOX and FOLFOX. This is in contrast to a study done by Baek KK et al. where 266 patients were treated with FOLFOX every two weeks, and 79 received CAPEOX every three weeks. Their study found out that the chance of developing severe OIPN was significantly higher in those receiving FOLFOX compared to CAPEOX (p<0.01) [22]. This difference with our study can be explained in that 58.1% of our patients received CAPEOX at a dose of 130 mg/m2 and the cumulative dosages at 6 cycles and above were higher compared to 41.9% who received FOLFOX. Limited sample size could be another factor for this difference between our study and most published studies.
Body mass index was strongly associated with OIPN where patients who were overweight and obese had higher prevalence and severity in our study. Those who were overweight and obese were 6.59 times and 21.66 times, respectively, more at risk of getting chronic OIPN symptoms compared to individuals who have normal body weight, this was statistically significant and gives an inference that the higher the BMI the more the risk of an individual in developing chronic OIPN symptoms during chemotherapy treatment. This finding was in agreement with a systemic review study done by Marry Jesse et al. where twelve articles describing relationships between overweight/obesity and presence of chemotherapy induced peripheral neuropathy were analyzed. Associations between body mass index (BMI), body surface area (BSA), sarcopenic obesity and chemotherapy induced peripheral neuropathy incidence and severity were described in patients who had received platinum and taxanes compounds. Eleven out of 12 studies indicated increased incidence of peripheral neuropathy with higher BMI or BSA or sarcopenic obesity, and the two studies that analyzed severity of symptoms found an increase in those who were overweight/obese [23]. The mechanisms influencing this relationship is related to increased dosing of chemotherapy related to a higher body surface area as well as the link of obesity with metabolic syndrome. The systemic inflammation often present with metabolic syndrome can lead to peripheral and central sensitization in the pain transmission system as well as axonal damage and demyelination of the nerves occurring as a result of proinflammatory cytokines [23].
Our study also took into consideration the severity of OIPN and found out that in for acute symptoms, 17.7% of the patients had grade 1, 69.4% had grade 2, 8.1% grade 3 and 4.8% grade 4. In Chronic OIPN 62.2% of the patients had grade 1 neuropathy, 24.3% grade 2 and 13.5% with grade 3. There is significant difference when compared with a study done by Sreenivasulu Palugulla et al. [24]. Where 219 patients were studied and the severity of both acute and chronic OIPN were assessed. The grade of acute OIPN was Grade1 in 70 (64.8%) patients, Grade 2 in 33 (30.5%) patients, and Grade 3 in 5 (4.6%) patients. For chronic OIPN, most of the patients had Grade 1 in 71 (55.9%) patients, Grade 2 in 54 (42.5%) patients, Grade 3 in 2 patients (1.5%) [24]. There is obvious difference in severity for both the acute and chronic OIPN with grade 2 and 3 in our present study. The reason behind this may be due to the number of patients studied and possibly different genetic composition of patients in this study. This study is limited by the fact that data were retrieved retrospectively, and affected by poor documentation of some clinical information. There was also a limited sample size in spite of including all patients who fit the inclusion criteria during the study period. Single centered nature of study and absence of objective neuropathy confirmation like nerve conduction studies could be other limitations. Nevertheless, the study has highlighted the prevalence, severity, and factors associated with OIPN among CRC patients treated in our setting. We need well-designed prospective long-term studies in the future to evaluate patients treated over an extended period with a sufficient sample size.
In conclusion, Number of cycles, cumulative dose and BMI were the major factors associated with the neuropathy. Most of the patients experienced grade 1 and 2 neuropathy in both acute and chronic symptoms. Proper symptomatic management of OIPN during and after completion of chemotherapy will help patients in our setting to tolerate chemotherapy toxicity and improve their quality of life.

Supplementary Information