A retrospective review of cancer treatments and outcomes among Inuit referred from Nunavut, Canada

Background Cancer is a health concern in Inuit populations. Unique cultural, dietary, and genetic factors and geographic isolation influence cancer epidemiology in this group. Inuit-specific data about oncology treatments and survival outcomes in Canadian Inuit referred to urban treatment centres are lacking. Methods A retrospective chart review of Inuit patients referred to The Ottawa Hospital Cancer Centre (tohcc) from the Baffin region of Nunavut between 2000 and 2010 was conducted. Nunavut cancer registry data were used to establish the percentage of cancer cases referred and their survival outcomes. Results Of 307 cancer patients registered among Baffin-region Inuit, 216 [70% (63 men, 153 women)] were referred to tohcc for chemotherapy (ct) and radiation therapy (rt). Mean age in the referred group was 59.3 years (range: 25–89 years), and current smokers constituted half the group (52%). The cancers most commonly leading to referral in men were lung (55%), colorectal (19%), and nasopharyngeal (11%) cancers; in women, they were lung (46%), colorectal (24%), breast (10%), nasopharyngeal (6%), and cervical (5%) cancers. Of the 216 referred patients, 82 (38%) had already undergone surgery, and 18 (8%) received chemoradiation or rt only, all given with curative intent. Among the surgical patients referred, 33 (40%) and 23 (28%) went on to receive adjuvant ct and adjuvant rt respectively. Among 116 patients referred for palliative care, 64 (55%) received ct, 76 (66%) received rt, 43 (37%) received both ct and rt, and 19 (16%) received neither treatment. Median all-stage overall survival was 10 months for patients with lung cancer [95% confidence interval: 6.1 to 13.9 months] and 37 months for patients with colorectal cancer [95% confidence interval: 14.8 to 59.2 months]. Conclusions High uptake of palliative and adjuvant ct and rt was observed in the Inuit patients referred to tohcc. Lung cancer was the most common cancer in referred Inuit men and women. The survival rates for Inuit lung cancer patients referred to tohcc were comparable to those in the rest of Canada. Further research is required to understand reasons for non-referral of Canadian Inuit to tohcc.


INTRODUCTION
Historically, the seasonal lifestyle and healthy "country" diet high in vitamin A, protein, and omega-3-fatty acids has protected the circumpolar Inuit from many chronic Western diseases 1 .However, a dramatic increase in cancer among Inuit populations over the last several decades has been observed 2,3 .Changes in lifestyle, diet, living conditions, and environmental exposures have all been implicated in the increased cancer burden [4][5][6] .Despite the relative youthfulness of the Canadian Inuit population (median age: 22 years) 7 , cancer is now a leading cause of death, with half this population's 10-year reduction in life expectancy being attributable to premature death from cancer 8 .This recent increase in cancer has not been uniformly distributed across all cancer sites.Although nasopharyngeal and oral cancers have remained consistently high, and prostate cancers uniformly low, marked increases in cancers of the lung and colon have been observed 2,3 .Lung and colon cancers now make up more than half of all cancers among Canadian Inuit 2 .The high smoking rate in this population-which is higher by a factor of 4 than elsewhere in Canada 9 -cannot be ignored, given its influence on the increasing risk of lung, colon, and other cancers, and it is the focus of many ongoing public health promotion and cancer prevention strategies.
Inuit and other northern residents in Canada experience a number of barriers to accessing cancer diagnostic services and treatment.Given the absence of cancer treatment facilities in the Canadian Arctic, Northern patients are typically referred out-of-territory to Yellowknife, Ottawa, Winnipeg, and Edmonton for cancer treatment 10 .Since 1998, The Ottawa Hospital Cancer Centre (tohcc) has been the referral centre for chemotherapy (ct) or radiation therapy (rt) for cancer patients from the Baffin (Qikiqtaaluk) region of Nunavut.That region represents a landmass of 1 million square kilometers in Canada's Eastern Arctic, where half of Nunavut's population-85% of whom are Inuit-resides 7 .Patients referred from this region must travel a minimum of 2000 km by air to receive specialty oncology care.Indications for ct regimens (4-6 cycles over 12-18 weeks) or daily rt treatments (fractionation over 4-6 weeks) require extended stays away from family and community.In some circumstances, patients undergoing ct might be permitted to return to Nunavut between infusion treatments every 3 weeks, but the travel burden and the challenges of treatment away from community social support networks are significant.
There is an identified dearth of Inuit-specific Canadian health data to inform health care policies and programming 11 .Cancer research in Canadian Inuit populations is limited by the inconsistent inclusion of Inuit indicators in the cancer registries covering the four geographic regions making up the Inuit land claims settlements, collectively known as Inuit Nunangat.As a result, Inuit cancer estimates are commonly generated by data linkage of cancer registry data with place of residence using postal codes, not Inuit status 3 .Even in jurisdictions in which Inuitspecific indicators are available, the cancer treatment and outcomes of Inuit referred to urban treatment centres remain poorly understood.Given the increasingly important impact that cancer is having on the health of Inuit populations, Inuit-specific data are necessary to determine the resources and services required to reduce treatment barriers and increase access to cancer care.
To start addressing the information gap, we reviewed cancer cases among Inuit from Nunavut referred to tohcc for ct and rt during 2000-2010.The specific aim of the study was to characterize Inuit-specific cancer referral patterns by establishing the proportion of all cancer cases from the Baffin region being referred to tohcc, the types of cancers being treated, and the ct and rt treatments being given.As a secondary objective, survival outcomes of Inuit referred to our centre were examined using date-of-death data obtained from the Nunavut cancer registry.

METHODS
A retrospective chart review of all cancer cases among adults (≥18 years of age) referred to the tohcc for ct and rt between 1 January 2000 and 31 December 2010 from the Baffin region of Nunavut was conducted.To ensure that the data were Inuit-specific, only cancer cases among confirmed Canadian Inuit land claims beneficiaries based on Nunavut health insurance card coding information were included in the review.Socio-demographic (age, sex, smoking history), diagnostic (type and stage of cancer), and treatment-related information (prior surgery, ct, rt) were collected for all identified patients from the medical charts at tohcc and the Qikiqtani General Hospital in Iqaluit, Nunavut.Given the high rates of tuberculosis (tb) in Nunavut 12 and the potential for its reactivation during chemotherapy 13 , a history of active or latent tb was also recorded.
After completion of data collection, the demographic information, Inuit status, and date of death for patients referred to tohcc were cross-referenced with information obtained through the Nunavut cancer registry.The total number of cancer cases among Inuit from the Baffin region during 2000-2010 was also obtained from the Nunavut cancer registry and compared with the number of cases referred for ct or rt, or both, at tohcc.To allow for comparison with cancer count data from the Nunavut cancer registry, second primary cancers in patients successfully treated for a first primary were analyzed as unique cases.
Parametric and nonparametric statistics are used for the primarily descriptive variables in this retrospective review.Based on the time (in months) from the date of diagnosis to date of death, estimates of overall survival (all cancers, lung cancer, and colorectal cancer) were calculated using the Kaplan-Meier method.All patients alive at the time of analysis were censored at the date of last follow-up.The study was approved by The Ottawa Hospital Research Ethics Board and registered with the Nunavut Research Institute.

RESULTS
Based on Nunavut cancer registry data, 374 cancers were diagnosed in people from the Baffin region during 2000-2010, 307 of which were diagnosed in Inuit individuals.Of the 286 cases referred from Nunavut to tohcc during that time, 216 cancers in confirmed Canadian Inuk adults were identified and included in the analysis (Table i).The dataset included 3 patients, each with 2 sequential primaries, who were both times referred to tohcc.Mean age at diagnosis was 59.3 years (range: 25-89 years), with one quarter of the patients being diagnosed before the age of 52.Median age at diagnosis was 65 years for lung cancer and 59 years for colorectal cancer.More than half these patients were smokers at the time of diagnosis (52%), and more than one third (38%) had a history of either latent or active tb.

DISCUSSION
This paper details the results of the first Canadian Inuitspecific study of referral patterns, treatments, and outcomes for Inuit cancer patients referred to an urban centre for ct or rt assessment.The large number of patients referred to our centre with lung, colorectal, and nasopharyngeal cancer is consistent with the cancer incidence patterns reported by others 2,3 .To our knowledge, the present study is the first to objectively quantify the sex-based referral differences for Canadian Inuit patients that has been suggested in earlier qualitative reports 14 .
Although the referrals to tohcc in many cases mirrored earlier population-based incidence estimates, some notable exceptions are evident.The proportion of colorectal cancer patients referred accorded with the proportion of cases reported by the Government of Nunavut 15 over a similar time period (22% vs. 19%), but lung cancer patients constituted a greater proportion of referrals than would be  expected based on the Government of Nunavut incidence report (49% vs. 33%) 16 .Although the reason for the latter discordance is unknown, it could reflect the symptoms associated with lung cancer, which often require palliative treatment, and it clearly supports continued efforts to reduce the smoking rate in the Inuit population.Consistent with earlier reports 2,3 , prostate cancer was uncommon among Inuit men, despite representing a quarter of all cancers in men elsewhere in Canada 17 .
The 61% 1-year overall survival for lung cancer reported here is comparable to the 62% most recently reported by the Government of Nunavut based on all lung cancer cases in Nunavut between 1999 and 2010 16 .Further, the median survival of Inuit lung cancer patients treated at tohcc (10 months) accords with an earlier report from Nunavut suggesting a median survival of 10.2 months 18 .A high proportion of Inuit cancer patients referred to tohcc were found to have advanced disease at presentation, commensurate with findings in the rest of Canada 19,20 and such that the overall survival rates were comparable 20,21 .Specifically, the 61% 1-year lung cancer survival is better than the 37.1% 1-year relative survival reported for people diagnosed with lung cancer in Canada during 2001-2005 20 .Likewise, for colorectal cancer, the 80% 1-year overall survival is comparable to the 75% 1-year survival reported by the Government of Nunavut for 1999-2010 15 and the 84.3% 1-year relative survival reported by Ontario for 2006-2010 21 .However, the 59% 2-year survival of colorectal patients reported here is slightly less than the 75%-77% reported elsewhere in Canada during the same period 17,21 .
Our results also showed important sex differences in Inuit referrals to our cancer treatment centre.Notably, the numbers of lung and colorectal cancer patients referred for ct or rt, or both, were higher among Inuit women by a factor of between 2 and 4, despite a comparable incidence of those cancers among men and women in Nunavut over the period in question 15,16 .Our data support earlier qualitative reports based on focus group discussions involving Canadian Inuit, which suggest that Inuit men are less likely to seek cancer treatment 14 .Cultural and health service delivery characteristics in the North have both been suggested to play a role in sex differences in health-care-seeking behavior and the willingness of men to undergo treatment 4,14 .That information should be considered in any initiatives to increase cancer awareness among Canadian Inuit.
A number of factors could explain why slightly less than one third (30%) of cancer patients among the Inuit of the Baffin region diagnosed during 2000-2010 were not referred to tohcc for ct and rt assessment.
First, some patients might have been managed with surgery alone and might not have been referred for adjuvant therapy.Given the results of recent Canadian populationbased studies of adjuvant ct referral patterns, lack of referral for ct assessment after surgery is not uncommon 22,23 .In Ontario, a study of referral patterns for adjuvant ct in patients with nsclc resected during 2004-2006 (1718 men, 1636 women) found that only 55% were seen postoperatively by a medical oncologist, and only 31% received ct 22 .Similarly, a population-based study in Nova Scotia found that, among patients with stage ii or iii colon or rectal cancer undergoing curative-intent surgery during 2001-2005 (n = 1151), only 59% ever received adjuvant ct 23 .Among the 82 patients referred to tohcc that had initially been managed surgically, half (50%) received adjuvant therapy (ct or rt, or both).
Second, patients diagnosed with metastatic disease might choose to forgo treatment and decline referral for oncology services.Indeed, among all patients diagnosed with metastatic nsclc in the province of Ontario during 2005-2009 (n = 8113; mean age: 68 years), 76% did not receive ct 24 .Inuit patients referred to urban treatment centres for palliative treatment run the additional risk of not being able to return home if their condition deteriorates, possibly further influencing their willingness to be referred.Notably, of the 116 Inuit cancer patients referred to tohcc who were not eligible for curative therapy, a high percentage (84%) received ct or rt, or both.
Third, there is an identified dearth of culturally appropriate cancer resource information for Inuit people, which limits health literacy with respect to the prognosis of various cancers and can lead to fatalistic attitudes about cancer treatment outcomes 14 .The extent to which such attitudes might have influenced the observed treatment referral patterns is, however, uncertain.
Finally, despite the many clinical factors that influence the initiation of cancer treatment in both the adjuvant and palliative settings 25 , patient preference often plays a critical role in determining the treatment plan.Earlier studies have suggested that geographic remoteness can influence cancer treatment choices 26,27 , but the extent to which the travel burden experienced by Inuit patients affects their treatment-seeking behaviour and their treatment choices is poorly understood and requires further study.
This retrospective study is the first to detail cancer referral and treatment patterns among Canadian Inuit at one of the largest Canadian cancer centres, but it nevertheless has limitations.Our analysis included cancer patients referred to tohcc only for systemic ct and rt assessment and treatment.We cannot comment, therefore, on Inuit patients from the Baffin region referred to cancer treatment centres elsewhere in Canada.But given that tohcc is the primary referral centre for the Eastern Canadian Arctic, the number of patients referred to other cancer centres, although unknown, would likely be small.It is also likely that our study did not capture all hematologic malignancies referred from the Baffin region, because those cancers are commonly managed by hematologists and are not referred to cancer treatment centres.But because hematologic malignancies are uncommon among Inuit populations 1,3 , the effect of their incomplete capture in our study findings is likely to be small.In addition, although we know from registry data the total number of cancer patients diagnosed during the study period, we do not know the proportion of patients that were managed with surgery alone or that were treated elsewhere among the 30% of patients not referred to our treatment centre, nor do we know the number of patients that declined treatment or their reasons for doing so.Characterization of the reasons for non-referral to tohcc of Inuit cancer patients living in the Baffin region would require patient chart reviews in each of the region's 13 Inuit communities, which is beyond the scope of the current review.

CONCLUSIONS
Despite the added travel burden, uptake of ct and rt is good among Canadian Inuit patients referred to tohcc.In this patient cohort, stage of disease at presentation and overall survival for the most common cancers appear proportionally the same as in the rest of Canada.Despite comparable incidence rates of common cancers, Inuit women are more likely than Inuit men to be referred for ct and rt assessment.That sex difference should be considered in any initiative to improve uptake of cancer therapies in Inuit populations.Given that lung cancer represented half of all cases referred to tohcc, any strategies to improve cancer treatment and outcomes among Inuit should, as a priority, incorporate anti-smoking initiatives.Further study is required to better understand reasons for non-referral.Moving forward, increased collaboration between rural clinics and urban cancer centres might help to improve access and uptake of effective cancer therapies and improve quality of life for Canadian Inuit populations.

FIGURE 2
FIGURE 2 Lung cancer overall survival in referred Inuit patients, all stages.

TABLE I
Patient characteristics a Latent or active.