In this issue of JAIDS: J Acquir Immune Defic Syndr, Bateganya et al1 provide an important review of the recent experience with non-Hodgkin lymphoma (NHL) at the Uganda Cancer Institute (UCI) in Kampala. Of 160 individuals over the age of 13 with newly diagnosed NHL seen at the center between January 2004 and August 2008, nearly one-third were known to be HIV seropositive. This proportion far exceeds the estimated 5%-7% prevalence of HIV infection in the Ugandan adult (15-year to 49-year old) population2 and attests to the impact of HIV on NHL development in this clinical setting. The good news from this analysis is that HIV-infected individuals who received antiretroviral therapy during NHL treatment showed a survival distribution that was indistinguishable from that of individuals with NHL who were known to be HIV seronegative. However, the study also identifies a tremendous gulf between the median survival of patients with NHL (irrespective of HIV status) treated at UCI and that expected in high-income countries and highlights the need for improvements in cancer therapy and supportive care, patient follow-up, routine clinical documentation, and diagnostic capabilities in low-resource settings.
Despite administration of a standard chemotherapy regimen, CHOP (intravenously administered cyclophosphamide, doxorubicin and vincristine, and oral prednisone), to almost two-thirds of the study subjects, most of whom were young (ie, in an age range that would be expected to have the best prognosis in high-income countries), the estimated median survival, even among HIV-seronegative individuals, was less than 1 year, and the authors found no evidence that receipt of chemotherapy improved survival. The survival of even the HIV-uninfected individuals in the current series was similar to that reported when CHOP was used to treat NHL in HIV-infected individuals in the United States in the absence of antiretroviral therapy.3 It also compares unfavorably with a 3-year survival rate in excess of 50% reported from the United States when the same chemotherapy regimen was used to treat HIV-negative patients with intermediate-grade and high-grade NHL whose median age was more than 554 and with a 2-year survival rate of 57% among patients 60-80 years old with diffuse large B-cell lymphoma in Europe.5 In both of these studies, event-free survival after 2-3 years, which can be considered a reasonable surrogate for cure, was approximately 40%. By contrast, only 20 of 154 subjects (13%) with adequate follow-up data in the current study were known to be alive 1 year or more after receiving a diagnosis of NHL.1 These differences are not likely to be solely a consequence of better prognostic features in the United States and European studies, as the majority of patients in the cited trials showed poor prognostic features.
One could easily get bogged down in a discussion of the finer points of this analysis and specific issues related to the management of NHL. It is probably more useful, however, to consider some of the broader issues it raises about the provision of cancer care in resource-constrained environments and measures that would enhance the quality of cancer care, improve treatment outcomes, and support the collection of patient data that would inform the rational use of limited resources. Recent estimates that by 2030, 70% of newly reported cancers are expected to be in developing countries6 emphasize the urgency of addressing this problem.
First, there is the issue of the large proportion of patients who were lost to follow-up, a finding that is likely not unique to NHL or UCI. Without the resources to ascertain the fate of the many patients who failed to return-which could be a consequence of lack of treatment efficacy, disability, or death from potentially treatable therapy-related complications, socioeconomic factors that prevented travel to the treatment center, or, in the case of HIV-infected individuals, HIV-related morbidity and mortality-it is impossible to know what sort of intervention(s) might deal most effectively with patient attrition. Addressing each of these possibilities requires different strategies, for example, alternative treatment regimens, better supportive care, and financial and logistical support for return visits and access to drugs. Providing support for personnel who could actively seek out the causes for failure to return is a relatively low-cost measure that could inform future analyses of this sort and help direct the application of resources to areas of greatest need.
Another issue is that of incomplete data, which restricts the ability to derive meaningful conclusions from this sort of retrospective analysis and direct the way forward. For example, although UCI had standard procedures in place to stage NHL, the tumors of only about 60% of the patients in this study had been staged. A similar proportion of baseline lactate dehydrogenase values, another component of the International Prognostic Index, were missing for the patients in this study. Although information on other factors (ie, hemoglobin, constitutional “B” symptoms, gender) were more reliably available in the UCI medical records and shown to correlate with survival, the large amount of missing data on factors that have been consistently correlated with survival in other settings makes it hard to draw meaningful comparisons with studies conducted elsewhere. Similarly, few data were available on the reasons why many patients did not receive chemotherapy or discontinued it prematurely. One way to improve data quality, which would create a more useful record for future analyses of factors associated with treatment response and survival, would be to provide standardized templates for the specific clinical, laboratory, and demographic data that should be collected on all patients and to require documentation of reasons for departures from the standard.
This study also highlights the reliance on pathologic diagnoses and lymphoma classification rendered at referring centers, without central review and verification, and the limited ability to fully characterize the phenotypic and molecular profile of the tumors. As described recently by Mwanda et al,7 detailed immunophenotypic studies of biopsy specimens from HIV-infected patients in East Africa revealed a wider diversity of NHL subtypes than expected in developed countries. This diversity was not suspected based on the standard hematoxylin and eosin staining available locally, and could have important implications for understanding NHL pathogenesis in sub-Saharan Africa, and its therapy and prognosis. These findings highlight the urgent need for additional training and support for precise pathologic diagnosis of cancer as part of capacity-building efforts in low-resource settings.
Although the authors of the current study can not tell us why their patients showed such a short median survival even in the absence of HIV infection, it is likely that that this is at least partially attributable to treatment-related toxicity. Enhancing the ability to recognize and treat therapy-related complications is one way to approach this issue and should be an eventual aim, but could be costly to implement in many places, at least in the short term. An example of a more pragmatic approach, which has already been shown to be feasible in a small study conducted, in part, at UCI, is the use of a dose-modified oral regimen that maintains antitumor efficacy while at the same time utilizing the existing supportive care infrastructure.7 In this clinical trial, conducted between May 2001 and August 2005 at UCI and at the Kenyatta National Hospital in Nairobi, Kenya, 49 HIV-infected patients with NHL received a rationally selected regimen designed to avoid cardiotoxic agents (ie, anthracyclines) and immunosuppressive corticosteroids, to include agents that cross the blood-brain barrier, and to minimize myelosuppression. Although only 35% of potentially eligible patients were enrolled, raising questions about selection bias, the study was notable for inducing a high overall response rate (78%) and a median survival of 12.3 months, despite advanced-stage disease in the majority of patients and access to antiretroviral therapy in a minority. One third of the patients survived 5 years, treatment-related mortality was low, and the International Prognostic Index proved predictive of survival. Although these results require confirmation in larger trials that include universal access to antiretroviral therapy, this study provides an excellent example of the sort of creative strategies that can be used to help those living in areas with limited resources achieve the substantial benefits of chemotherapy that have been so well established elsewhere.
Some may argue that testing “weaker” cancer chemotherapy in low-resource settings is unethical and that instead we should be finding ways to provide Western-style treatment-in the case of stage III or IV NHL, this would typically involve a CHOP-like chemotherapy regimen administered with rituximab (and colony-stimulating factors, broad-spectrum antibiotics, and blood component support to deal with therapy complications). Although comparing the outcomes of low-tech and high-tech (and high-cost) treatment for NHLs and other cancers is a legitimate research question, it is not a foregone conclusion that the outcomes (ie, response rates, survival, toxicity) of what is considered “standard” therapy in high-income countries will be the same everywhere, even with appropriate support for such treatment. Many factors may influence the effects of cancer treatment and will need to be studied in different populations, including racial and ethnic variations in genetic factors that influence drug disposition, efficacy, and side effects.8
It is concerning that Bateganya et al1 did not mention the oral chemotherapy trial7 cited above. This study was in progress at UCI during a period that overlapped with the early part of the current study, in which CHOP is described as the only chemotherapy regimen received, so it is unclear whether some NHL patients treated at UCI were excluded from the analysis. Nonetheless, despite the limitations inherent in a retrospective analysis, and which the authors were otherwise careful to point out, this is a noteworthy article that adds to our relatively limited knowledge about cancer in Africa and suggests strategies to begin closing the substantial gaps between developed and developing countries in cancer treatment and survival.
1. Bateganya MH, Stanaway J, Brentlinger PE, et al. Predictors of survival after a diagnosis of non-Hodgkin's lymphoma in a resource-limited setting: a retrospective study on the impact of HIV infection and its treatment. J Acquired Immune Defic Syndr. 2011;56:312-319.
2. Hladik W, Musinguzi J, Kirungi W, et al. The estimated burden of HIV/AIDS in Uganda 2005-2010. AIDS. 2008;19:503-510.
3. Kaplan LD, Kahn JO, Crowe S, et al. Clinical and virologic effects of recombinant human granulocyte-macrophage colony-stimulating factor in patients receiving chemotherapy for human immunodeficiency virus-associated non-Hodgkin's lymphoma: results of a randomized trial. J Clin Oncol. 1991;9:929-940.
4. Fisher RI, Gaynor ER, Dahlberg S, et al. Comparison of a standard regimen (CHOP) with three intensive chemotherapy regimens for advanced non-Hodgkin's lymphoma. N Engl J Med. 1993;328:1002-1006.
5. Coiffier B, Lepage E, Briere J, et al. CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma. N Engl J Med. 2002;346:235-242.
6. Farmer D, Frenk J, Knaul FM, et al. Expansion of cancer care and control in countries of low and middle income: a call to action. Lancet. 2010;376:1186-1193.
7. Mwanda WO, Orem J, Fu P, et al. Dose-modified oral chemotherapy in the treatment of AIDS-related non-Hodgkin's lymphoma in East Africa. J Clin Oncol. 2009;27:3480-3488.
8. Evans WE, McLeod HL. Pharmacogenomics-drug disposition, drug targets, and side effects. N Engl J Med. 2003;348:538-549.
© 2011 Lippincott Williams & Wilkins, Inc.