Anal cancer survival: a socioeconomic analysis

Abstract

Introduction

Squamous cell carcinoma is the most common variant of anal malignancy. Certain disease-related factors have been established in determining survival. These include tumour size, differentiation and nodal involvement. Other factors such as HIV status, human papillomavirus infection, smoking and socioeconomic disparity may have important roles, however few data are available on the UK population. We aim to correlate social deprivation and survival of anal cancer patients at a tertiary centre.

Materials and methods

All consecutive cases diagnosed with anal squamous cell carcinoma and treated as per local protocol between July 2010 and April 2017 were included. The pathological and demographical details were collected from a prospectively maintained database. Socioeconomic deprivation was defined for each postcode using the Index of Multiple Deprivation decile compiled by local governments in England. Survival was estimated using Kaplan–Meier analysis and Cox regression was used to investigate the effect of different factors on overall survival.

Results

A total of 129 patients with anal squamous cell carcinoma over a median follow-up of 43 months were included. Overall survival for the entire patient cohort was 87.7% (95% confidence interval, CI, 82.0–93.7%), 75.5% (95% CI 67.5–84.5%) and 68.9% (95% CI 59.7–79.6%) at one year, three years and five years, respectively. On multivariate analysis, Index of Multiple Deprivation and income do not significantly influence overall survival (p = 0.79, hazard ratio, HR, 1.07; 95% CI 0.61–1.63), (p = 0.99, HR=1.00; 95% CI 0.61–1.63), respectively. Increased risk of death was observed for male sex (p = 0.02, HR=2.80; 95% CI 1.02–5.50) and larger tumour size (p = 0.01, HR=1.64; 95% CI 1.12–2.41).

Conclusion

In contrast to US studies, there is little difference in survival between the least deprived and most deprived groups. We attribute this to equal access to intensity-modulated radiation therapy-based chemoradiotherapy. Thus, a highly effective treatment made available to all mitigates any survival difference between socioeconomic groups.

Introduction

Squamous carcinoma of the anal canal is a rare disease which affects women more than men and is predisposed by the human papillomavirus and HIV infection.¹ The incidence of anal cancer has been on the rise in the UK and other Western countries.^2,3 Primary treatment is chemoradiotherapy, while abdominoperineal resection is reserved for salvage treatment in the case of recurrence or resistant disease.^4,5

Multiple prognostic factors have been identified in determining overall survival. Tumour size, lymph node involvement, skin ulceration and male sex are the worst prognostic factors shown in the European Organization for Research and Treatment of Cancer (EORTC)and RTOG 98–11 studies.^6–8

Tumour-related factors have been the main focus of research to date. However, the role of other factors such as socioeconomic status has not been explored widely. One epidemiological study involving 19,191 patients with anal cancer from the US National Cancer Registry between 1985 and 2000 has established that lower median household income is associated with poor outcome.⁶ This has been supported by another study involving 9,550 cases.⁹

We investigated the impact of socioeconomic status on the overall survival of patients with squamous cell carcinoma of the anus in a tertiary referral centre in the North East of England which spans a geographically large catchment area with a population of 1.2 million.¹⁰ Our centre is one of the first in the UK to adopt the use of intensity-modulated radiation therapy (IMRT) in the treatment of anal cancer in July 2010.

Materials and methods

Study population

Data on staging, demographics, follow-up and survival of all cancer patients between July 2010 and April 2017 were obtained from a prospectively maintained hospital database. After IMRT became available in James Cook University Hospital in July 2010, all consecutive patients diagnosed with anal squamous cell carcinoma between July 2010 and April 2017 were included. All patients included in the survival analysis have been uniformly treated with intensity-modulated radiotherapy-based chemoradiation with a curative intent and followed up at three-month intervals for the first year and six-monthly thereafter until five years.

Socioeconomic data

Socioeconomic deprivation was defined for each patient’s postcode using an Index of Multiple Deprivation (IMD) decile compiled by the Ministry of Housing, Communities and Local Government in England.¹¹ The IMD is the official measure of relative deprivation for areas in England known as lower-layer super output areas (LSOA) where the average population in an area is 1,500. Over 32,000 LSOAs are ranked and grouped into 10 equally sized deciles. The IMD is calculated using seven metrics: income, employment, health deprivation and disability, education skills and training, barriers to housing and services, crime and living environment.¹¹ As there are no definitive criteria to further categorise patients on a deprivation scale, patients were grouped into most deprived, A (IMD deciles 1–3), moderately deprived, B (IMD deciles 4–6), and least deprived, C (IMD deciles 7–10) to roughly achieve similar numbers in each group.

Analysis

Tumour size was classified using the American Joint Committee on Cancer staging system. The relation between relative deprivation and tumour stage at the time of diagnosis was studied using the chi-square (χ²) test. Survival was estimated using Kaplan–Meier analysis and log-rank tests were used to compare survival curves between groups. A multivariate Cox proportional hazards regression model was used to investigate the effect of different factors on the overall survival. Multivariate analysis included sex, age and either IMD group or income group as a metric derived from the IMD.

Results

A total of 147 cases with anal malignancy were included in this study. Of these, 15 patients were excluded from analysis due to histological diagnosis other than squamous cell carcinoma. These included adenocarcinoma (n = 10), malignant melanoma (n = 2), neuroendocrine carcinoma (n = 2) and adenosquamous carcinoma (n = 1). Among the 132 patients diagnosed with squamous cell carcinoma, one patient was excluded from survival analysis due to a missing address. Two further patients had cancer confined to the mucosal layer (Tis) and were excluded from survival analysis as they did not receive chemoradiotherapy. Of the remaining patients, women comprised 70.5% (n = 91, female to male ratio 2.39:1; Table 1) and the median age at diagnosis was 66.8 years (range 33–98 years; Table 1). Median follow-up was 27 months. Table 1 shows the demographic distribution of the three socioeconomic groups.

Table 1 .

Demographics, TNM staging and tumour grades by Index of Multiple Deprivation

		Sex (%)		T stage (%)					N stage (%)					M stage (%)		Grade (%)
IMD	Mean age	F	M	T1	T2	T3	T4	Tis	N0	N1	N2	N3	Nx	M0	M1	G1	G2	G3	Gx
A	65.1	73.7	26.3	14.0	31.6	21.1	29.8	3.50	60.0	18.2	16.4	3.6	1.8	98.2	1.8	19.3	43.9	29.8	7.0
B	67.3	69.2	30.8	12.8	38.5	20.5	28.2	0.00	43.6	20.5	18.0	5.1	12.8	97.4	2.6	28.2	30.8	30.8	10.2
C	69.0	68.6	31.4	31.4	17.1	25.7	25.7	0.00	48.6	22.9	22.9	2.8	2.8	97.1	2.9	17.1	40.0	40.0	2.9

A, most deprived; B, moderately deprived; C, least deprived; F, female; Gx, grade unidentified; M, male; M stage, metastasis; N stage, nodal stage at diagnosis; T stage, tumour size stage at diagnosis; Tis: carcinoma in situ confined to the mucosal layer

Within the most deprived group, 29.8% (n = 17) of the patients presented with a T4 stage compared with 28.2% (n = 11) in the middle group and 25.7% (n = 9) in the least deprived group. No significant difference was observed in tumour size at the time of presentation between the deprivation groups (χ² 10.43, df 8, p = 0.23).

Survival analysis

Overall survival for the entire patient cohort was 87.7% (82.0%–93.7%), 75.5% (67.5%–84.5%) and 68.9% (59.7%–79.6%) at 1 year, 3 years and 5 years, respectively (Figure 1). We found no significant difference in survival between the various IMD groups (p = 0.54; log-rank; Figure 2A) or between income groups (p = 0.50, log-rank; Figure 2B).

Figure 1 .

Overall survival of patients diagnosed with squamous cell carcinoma of the anus and treated with chemoradiotherapy (n = 129); 52% five-year survival, 2007–2013, Office for National Statistics³

Figure 2 .

Overall survival of: (a) Least deprived, moderately deprived and most deprived groups. Patients are grouped by IMD. Log-rank was used to test the null-hypothesis that there is no difference between the survival curves. (b) Low-, middle- and high-income groups. Patients are grouped by income. Log-rank was used to test the null-hypothesis that there is no difference between the survival curves.

Separate multivariate analysis models were created to analyse the effect of IMD group and income group on overall survival. IMD (p = 0.79, HR 1.07; 95% CI 0.61–1.63) and income group (p = 0.99, HR 1.00; 95% CI 0.61–1.63) do not significantly influence overall survival. Both models showed that male sex is a significant poor prognostic factor (p = 0.02, HR 2.80; 95% CI 1.02–5.50; values from IMD model). Having a T stage 1 tumour was associated with an improved prognosis, although significance was borderline (p = 0.05, HR 0.26; values from IMD model). Hazard ratios with 95% confidence intervals for the IMD and income group models are summarised in Table 2. The inclusion of N stage as a categorical variable produced a coefficient which converged on infinity, probably because it is strongly correlated with T stage, so it was not included in the final Cox models.

Table 2 .

Hazard ratios and p-values for the multivariate Cox proportional hazards models of Index of Multiple Deprivation and income group overall survival

	IMD model				Income model
	HR	Lower 95% CI	Upper 95% CI	p-value	HR	Lower 95% CI	Upper 95% CI	p-value
Age	1.03	0.99	1.07	0.12	1.03	0.99	1.07	0.12
Male sexa	2.80	1.21	6.43	0.02	2.80	1.21	6.47	0.02
T-stage 1a	0.26	0.07	1.01	0.05	0.26	0.07	1.02	0.05
T-stage 2a	0.46	0.15	1.42	0.18	0.46	0.15	1.39	0.17
T-stage 3a	1.74	0.63	4.78	0.28	1.76	0.64	4.83	0.27
Grade 2	0.71	0.25	1.97	0.51	0.70	0.25	1.93	0.49
Grade 3	0.79	0.28	2.26	0.66	0.79	0.28	2.26	0.66
IMD/Income group	1.07	0.67	1.69	0.79	1.00	0.65	1.54	0.99

CI, confidence interval; HR, hazard ratio; IMD, Index of Multiple Deprivation

^a Tumour grade at diagnosis, analysed as a categorical variable

Discussion

Anal cancer is a relatively curable disease with low rates of metastasis and locoregional recurrence.^12,13 An epidemiological study from Cancer Research UK compared the incidence of anal cancer between the most and least deprived areas.¹⁴ It showed incidence rates were 53% and 82% higher in female and male populations living in the most deprived areas, respectively. Previous studies have looked at the association between survival and socioeconomic status in a variety of cancers,^14–16 and some of the disparity in survival has been attributed to the stage at the time of presentation in colorectal cancers,^17,18 but other studies do not support this conclusion.^19,20

The tumour size and lymph node status are known prognostic factors; however, we do not see any difference in these two parameters at the time of presentation between IMD groups. Homogeneity of the stage at presentation across socioeconomic groups in our study may explain similar survival outcomes between socioeconomic groups.

Importantly, malignancies which carry good prognoses, such as acute lymphoblast leukaemia (ALL) and acute myeloid leukaemia (AML), show relatively little variation in survival between socioeconomic groups in the UK.^16,21,22 The National Cancer Intelligence Network (NCIN) report on anal cancer outcomes by deprivation showed that the difference in mortality between the least and most deprived groups in the UK was less than 1/100,000 of the population, while for colorectal, stomach and lung cancers the differences were 5, 5 and 45/100,000, respectively.¹⁶ For ALL, AML, chronic myeloid leukemia. Hodgkin’s lymphoma and cancers of the penis and testes, the difference between the least and most deprived groups was also less than 1/100,000.¹⁶ Thus it appears that the survival difference between socioeconomic groups is weaker for the most prognostically favourable malignancies. Although the same NCIN report has shown a difference in anal cancer mortality between socioeconomic groups, this amounts to only 65 excess deaths from a total of 1,634 cases (4.28%).¹⁶ It is not clear that all patients in this report received IMRT, whether deaths were directly due to anal cancer and, furthermore, multivariate survival analysis was not used.

Lung, stomach and colorectal cancers have worse prognosis among the deprived.^14,16 Certain patient related factors such as smoking, obesity, lack of exercise, motivation to adhere to prolonged treatment and pre-existing comorbidities could be contributing factors,^23–25 but further studies are warranted. We therefore hypothesise that the socioeconomic variation in anal cancer survival is small as this is a relatively curable disease. Our healthcare is free at the point of service and anal cancer patients almost universally receive IMRT-based chemoradiotherapy, which may explain the similar outcome between socioeconomic groups. This contrasts with the US studies which have shown a substantial survival difference across income groups bearing in mind that healthcare delivery in the US is variable.

There is conflicting evidence of the impact of HIV positivity on the outcome of patients with anal cancer. HIV-positive patients may have a poorer outcome in terms of treatment related toxicity and risk of local relapse.²⁶ The prevalence of HIV infection among patients with anal cancer has not been correlated to socioeconomic data, nonetheless there is good evidence that poverty drives the spread of HIV infection.²⁷ The uniform access to the healthcare system might have some role in attenuating the outcome difference among patients who are HIV-positive with anal cancer, but that is beyond the scope of this study.

Our study demonstrated that overall survival outcomes were comparable or superior to some pre-existing studies.^3,6 The five-year survival rate for patients treated with chemoradiotherapy in a recent population-based study in the United States was 62.1%⁶ and 52.0% in the UK (2007–2013).³ The slightly higher overall survival in our study could be explained on the basis of the following: the strict selection criteria for squamous cell carcinoma; higher level of expertise with anal carcinoma and designated teams for close follow-up by oncologists and surgeons at three-monthly intervals in the first year and six-monthly thereafter until five years at our specialised study centre.

One weakness of our study is the sample size given the rare nature of the disease, which may limit our ability to detect small survival differences. However, the uniformity of treatment, regular follow-up, accurate data collection facilitated by direct access to patients’ records has enabled us to perform multivariate statistical analysis. In addition, data on smoking habits were not available to compare smoking rates between socioeconomic groups in our patient cohort.

Conclusions

In our cohort, patients from the most deprived areas do not present with more locally advanced disease. In contrast to US studies, there is little difference in survival between least and most deprived groups. We attribute this to equal access to IMRT-based chemoradiotherapy. Thus, a highly effective treatment made available to all mitigates any survival difference between socioeconomic groups.