To evaluate the incidence and risk factors for secondary solid tumors in Japan after allogeneic hematopoietic SCT (allo-HSCT), 2062 patients who had received allo-HSCT between 1984 and 2005 were retrospectively analyzed. Twenty-eight patients who developed 30 solid tumors were identified a median of 5.6 years after transplantation. The risk for developing tumors was 2.16-fold higher than that of the age- and sex-adjusted general population. The cumulative incidence of solid tumors at 10 years after allo-HSCT was 2.4%. The risk was significantly higher for tumors of the skin, oral cavity and esophagus (standard incidental ratio 40.23, 35.25 and 10.73, respectively). No increase in gastric, colon or lung cancer, despite being the most prevalent neoplasm in the Japanese, was observed. In multivariate analysis, occurrence of chronic GVHD and malignant lymphoma as a primary disease was associated with a higher risk for developing solid tumors. Eighteen patients are still alive, and their 5-year probability of survival since diagnosis of solid tumors is 59.7%. Our data suggest that the incidence and risk factors of secondary solid tumors in Japanese allo-HSCT recipients are comparable to those reported in Western countries and emphasize that the early detection of solid tumors has a crucial role in improving OS.
Allogeneic hematopoietic SCT (allo-HSCT) offers curative therapy for malignant and non-malignant hematological diseases. Improvements in outcomes after transplantation have led to a continuous increase of long-term survivors among allo-HSCT recipients. The development of secondary neoplasms is one of the serious late complications after allo-HSCT, and is associated with considerable morbidity and mortality.1, 2 Several studies reported that the incidence of secondary solid tumors in recipients aged 10 years or older ranged from 2.2 to 6.1%, and that TBI, T-cell depletion, GVHD and immunosuppressive therapy may be important risk factors.3, 4 Although these studies demonstrated that the risk of developing secondary solid tumors is significantly higher in allo-HSCT recipients in Western countries, only a few data documenting the incidence for secondary solid tumors in the Japanese population are available.5 Thus, we retrospectively analyzed 2062 Japanese allo-HSCT recipients registered to the Kanto Study Group for Cell Therapy database, in order to determine the incidence and risk factors of secondary solid tumors following allo-HSCT.
Patients and methods
Between July 1984 and June 2005, 2062 patients underwent allo-HSCT in 14 facilities of the Kanto Study Group for Cell Therapy, and were retrospectively evaluated. The occurrence of each non-hematological secondary solid tumor was identified by using the Kanto Study Group for Cell Therapy database, and detailed clinical information, including histological type of malignancy, site of occurrence, date of diagnosis, treatment and its outcome, were obtained for each secondary tumor by chart review. The tissue diagnoses of secondary solid tumors were given in each institution.
Patient characteristics are summarized in Table 1. The median duration of follow-up for surviving patients after allo-HSCT was 5.7 years, and 1290 (62.6%) patients were followed through to the end of study date. Majority of the patients were adults, with only 97 patients (4.7%) aged <18 years at the time of allo-HSCT. The median age at transplant was 36 years (range, 7–68 years). Most of the patients underwent allo-HSCT for a primary diagnosis of hematological malignancies including ALL (20.9%), AML (32.3%), CML (21.2%), myelodysplastic syndrome (10.5%) or malignant lymphoma (5.4%); the median time from diagnosis to allo-HSCT was 10.6 months (range, 0.3–503.5 months). Donor and hematopoietic stem cell sources included BM from related donors (n=852, 41.3%), peripheral blood from related donors (R-PBSCT; n=329, 16.0%), BM from unrelated donors (n=706, 34.2%) and cord blood from unrelated donors (n=160, 7.8%). TBI-based regimen was used for 68% and non-TBI regimen for 32% of the patients as the conditioning regimen. Prophylaxis for GVHD was attempted mainly with calcineurin inhibitor (CYA or tacrolimus) plus MTX in 2041 patients (99.0%). Acute GVHD (aGVHD, grade I–IV) developed in 1390 (67.4%) and chronic GVHD (cGVHD) in 1140 patients (55.3%). The median duration of follow-up was 3.3 years (range, 0.2–21.9 years) for all patients and 5.6 years (range, 0.3–21.9 years) for survivors.
The number of person-years at risk was calculated for each patient from the date of transplantation till the date of last contact, death, diagnosis of a new neoplasm or completion of the study (18 January 2006) in the order of their first occurrence. Age (by 5-year bands)-, sex- and year-specific incidence rates for all cancers at specific anatomical sites were applied to the appropriate person-years at risk to compute the expected numbers of cancers. Incidence of all cancers in the general population was obtained from the database of the population-based cancer registration in Japan.6 Standardized incidence ratios (SIRs) were then calculated by obtaining the ratio of the observed to the expected number of cases, accompanied by the 95% confidence intervals (CIs).7, 8 The cumulative incidence of developing a secondary malignancy was estimated after adjusting for the competing risk of patient death using Gray’s method.9, 10 We estimated the survival rate among patients who developed secondary tumors by the Kaplan–Meier method. Univariate Cox proportional hazards regression analysis was used to determine statistically significant predictor variables for the development of a secondary malignancy. A multivariate analysis was performed using variables significant at univariate levels, which were potentially associated with the development of a secondary malignancy. All tests were two-sided, and a P value of 0.05 was considered statistically significant.
Diagnosis and outcome of secondary solid tumors
In total, 28 patients developed 30 solid malignancies at a median of 5.6 years (range, 0.3–17.6 years) after allo-HSCT. These secondary tumors involved the skin (6), oral cavity or pharynx (11), esophagus (3), stomach (3), colon (3), lung (2), thyroid gland (1) and prostate (1). Characteristics of patients who developed a secondary solid malignancy are shown in Tables 1 and 2. The median age of these patients at the time of allo-HSCT was 35 years (range, 10–57 years), and the median age at the time of diagnosis with a secondary malignancy was 42 years (range, 17–61 years).
Among the 30 patients with secondary tumors, 29 received various therapies, with the exception of 1 patient with an early gastric cancer, for whom the diagnosis was made by autopsy after death from primary disease. Surgery was performed in 16 patients, and endoscopic mucosal resection was performed in 1 patient with gastric cancer, of which 16 operations were curative. All patients who received a curative operation, with the exception of one patient who died from cGVHD, survived for 1 to 94 months without the recurrence of solid tumors after the diagnosis of secondary solid tumors. Seven patients were treated with surgery and chemotherapy with or without radiotherapy, and five of seven survived for 6 to 48 months. Only one of the five patients who were treated with chemotherapy and/or radiotherapy survived 6 months from diagnosis. Overall, of the 28 patients, 1 died from primary disease, 2 from cGVHD and 7 from their secondary neoplasm at a median of 10.4 months (range, 0–23.2 months) after the diagnosis. The probability of a 5-year survival since diagnosis of secondary tumors was 59.7% (Figure 1).
Probability and risk factors of developing secondary solid tumors
The cumulative incidence of developing any secondary solid tumors at 5, 10 and 15 years after transplantation was 0.9, 2.4 and 3.7%, respectively (Figure 2). The overall risk of developing solid tumors had significantly increased in the study population; 30 cancers were observed compared with 13.90 cases expected in an age- and sex-matched general population (SIR 2.16; 95% CI 1.46–3.08). The overall risk was especially elevated for neoplasms of the skin (SIR 40.23; 95% CI 14.77–87.57), oral cavity or pharynx (SIR 35.25; 95% CI 17.59–63.06) and esophagus (SIR 10.73; 95% CI 2.21–31.36). In contrast, the risk of gastric cancer, a common cancer in Japan, had not increased compared with that in the general population (SIR 1.33; 95% CI 0.27–3.89) (Table 3).
In univariate analysis, R-PBSCT as the stem cell source (P=0.041), malignant lymphoma as the primary disease (P=0.014) and presence of cGVHD (P=0.05) were significant risk factors for development of a secondary solid tumor. Presence of aGVHD was marginally significant (P=0.062). These four factors were added to the Cox regression model in multivariate analysis. Two independent risk factors, malignant lymphoma as the primary disease and occurrence of cGVHD, were identified (P=0.005, relative risk (RR) 4.7 and P=0.043, RR 2.4, respectively). In the oral cavity and esophageal squamous cell carcinoma, these two risk factors were also significant in multivariate analysis (P=0.010, RR 8.1 and P=0.019, RR 4.9, respectively). Gray’s method was used to analyze these two risk factors to confirm this result. Malignant lymphoma as primary disease and occurrence of cGVHD remained significant (P=0.006 and 0.009, respectively). In contrast, no significant risk factor was identified for secondary solid tumors in other sites of the body.
Among the 2062 patients who received allo-HSCT, 30 secondary solid tumors occurred in 28 patients, resulting in an estimated cumulative risk of 0.9, 2.4 and 3.7% at 5, 10 and 15 years, respectively (Figure 2). The latency of secondary tumors was discovered at a median of 5.6 years (range, 0.3–17.6 years). The overall incidence was 2.16-fold higher compared with age- and sex-adjusted cancer rates (Table 3) for the general population.
This study is one of the largest surveys in Japan investigating secondary malignancies after allo-HSCT. These results are similar to those reported previously. Curtis et al.4 reported the cumulative incidence of secondary solid cancer as 2.2 and 6.7% at 10 and 15 years, respectively (RR 2.7 compared with the general population) in 19 229 allografts; this was the largest study to evaluate the incidence of secondary malignancies after allograft to date. Similarly, Baker et al.11 studied 3372 allogeneic and autologous HSCT recipients, and reported a cumulative incidence of solid cancers of 3.8% at 20 years with a SIR of 2.8 in a retrospective analysis. Shimada et al.5 reported that cumulative incidence at 5 and 10 years was 1.9 and 4.2%, respectively, in 809 Japanese HSCT recipients. But, these two studies included both allogeneic and autologous transplant recipients for analysis. Consistent with the previous reports, the current study confirmed the increasing incidence of secondary solid tumors, even after a period of observation of more than 15 years, thus emphasizing again the importance of longer follow-up in this population.
Various risk factors that may contribute to the development of secondary solid cancers after allo-HSCT have been identified. These include TBI as a part of the conditioning regimen, previous radiotherapy, immunodeficiency from incomplete recovery after allo-HSCT, the use of immunosuppressive agents, occurrence of aGVHD or cGVHD, and advanced age.12 Curtis et al.13 revealed that cGVHD and its therapy were strongly correlated with the risk for developing squamous cell carcinomas of the oral cavity and skin in a case–control study. The possible mechanism behind this correlation is that cGVHD leads to persistent inflammation in the involved organs, and may stimulate regeneration of the epithelium and subsequent emergence of neoplastic cells. Furthermore, instances of solid organ transplantation indicate that prolonged immunosuppressive therapy leading to impairment in immune surveillance is associated with an increased risk of both solid malignancies and non-Hodgkin’s lymphoma.14, 15 This may be caused by co-carcinogenic effects of pretransplant chemo-radiation resulting in genetic damage. Most instances of microsatellite instability of the colon or buccal non-neoplastic epithelium after allo-HSCT were often caused by aGVHD or cGVHD. These genomic alternations may also be implicated in the evolution of post transplant complications such as secondary malignancies.16 In our study, the occurrence of cGVHD was also identified as the risk factor for secondary tumors, particularly in the oral cavity and esophageal squamous cell carcinoma, where occurrence of secondary tumors was found to be 35- and 11-fold higher, respectively, than that expected in the general population (Table 3).
Only one case of squamous cell carcinoma of the skin was diagnosed in this study as compared with five non-squamous cell cutaneous malignancies. This is similar to the findings of previous reports for the Japanese population.5 However, this contrasts with the finding for patients in Western countries.3, 11 This disparity may be explained by the difference in cGVHD manifestation or in genetic susceptibilities between ethnic groups. The most common tumors observed in our study were those of the skin and oral cavity. The predominance of cutaneous and oral cancers is similar to that reported previously.17, 18 Although gastric, colon and lung cancers are prevalent in the Japanese population,6 the incidence of these neoplasms is not statistically different from that in the general population in this study.
Despite the well-known association between radiation and secondary solid malignancies, only two of the largest studies of secondary solid malignancies occurring after allo-HSCT describe a significant relationship with TBI. Curtis et al.4 reported that a significant dose related to a 2.7- to 4.4-fold higher risk with TBI. Socie et al.19 found a 3.1-fold risk with high-dose TBI. In contrast, other studies failed to demonstrate a significant association between secondary malignancies and TBI. Witherspoon et al.20 reported a 3.9-fold increased risk of all secondary malignancies after TBI in 2246 auto- and allografts, but this was not significant when assessing solid malignancies alone. Bhatia et al.21 showed a sixfold-increased risk for secondary malignancies in patients treated with TBI, which approached statistical significance (P=0.08). However, a more recent report with 6 more years of follow-up gave less conclusive results (RR 1.5, P=0.27).11
A TBI-containing conditioning regimen was not a statistically significant risk factor in this study. There may be several explanations for this finding. First, the follow-up period is still relatively short in comparison with the generally long latent period of radiogenic cancers. Several reports suggest that the risk of secondary solid malignancy after radiation exposure remains elevated for multiple decades;22, 23, 24 several radiogenic malignancies take a long period to develop, and hence may require much longer follow-up. Second, the risk of such cancers is frequently high among patients undergoing irradiation at a young age.19, 25 Third, certain tumor types, such as those in the brain, thyroid, salivary gland and bone connective tissue, occur in association with radiation exposure.26, 27, 28, 29 Two large-cohort studies in children indicated that brain and thyroid cancers accounted for the increased risk of TBI.2, 4 In this study, which mainly included adult patients, only one case of thyroid cancer was observed. These results indicate that a distinctive mechanism may participate in the evolution of different post transplant solid tumors.
An unexpected finding was that of the malignant lymphoma as a primary disease conferring significant risks of developing into a secondary solid tumor. Among all secondary tumors, malignant lymphoma as the primary disease had an RR of 4.7, which was significantly higher compared with acute leukemia. When the cohort was limited to secondary tumors in the oral cavity and esophagus, much high risk (RR 8.1) was demonstrated. This observation has not been reported previously in the literature, and its explanation remains to be elucidated. Because allo-HSCT is not routinely recommended for treating malignant lymphoma during the first CR in most cases, lymphoma patients who undergo allo-HSCT may be heavily treated to start off with. Pre-transplant treatment before the conditioning therapies, such as salvage therapy, radiotherapy and autologous HSCT, might have already predisposed those patients to develop secondary solid malignancy after allo-HSCT. Indeed, it has been reported that the incidence of lung cancer increases with increases in radiation dose and cycles of alkylating agents in Hodgkin’s lymphoma therapy.30 Unfortunately, we did not have comprehensive records of pre-transplant chemotherapy and/or radiotherapy, and hence could not assess the influence of pre-transplant therapy on the risk of developing a secondary solid tumor after allo-HSCT. It is necessary to estimate this finding carefully, because patients of malignant lymphoma are only 5.4% of all patients.
Few data exist about treatment of secondary solid tumors; and the prognosis of secondary malignancies is generally considered to be poor, depending on an early diagnosis at a potentially curable stage. Our results are consistent with those reported by Favre-Schmuziger et al.31 They reported five patients developing secondary solid tumors after allo-HSCT, who were treated as de novo tumor, and four out of five are alive without tumor recurrence. All patients in our study who received curative care overcame their secondary tumors. Because the risk of developing secondary malignancies after allo-HSCT continues to increase with time, all transplant recipients should be followed for the long term to detect cancers at an early stage.