CORR Insights®: Should the Use of Biologic Agents in Patients With Renal and Lung Cancer Affect Our Surgical Management of Femoral Metastases?

Where Are We Now?

Decisions about the role of surgery and the most-appropriate bone fixation or reconstruction methods for a patient with osseous metastases often depend on the patient’s response to systemic oncologic treatment and prognosis for survival. Lung and renal cell carcinoma are among the most-common malignancies to metastasize to long bones, and osseous metastases are associated with a poorer prognosis for survival. But over the past decade, the increasing availability of systemic biologic agents, particularly receptor tyrosine kinase inhibitors and immunotherapy agents, has improved the life expectancy of many of these patients [11, 12].

As patients with osseous metastases from lung or renal cell carcinoma continue to live longer, it is important for clinicians to understand the implications of this longer life span, not only for the risk of skeletal-related events (pathologic fracture, need for radiation therapy, need for surgical stabilization, spinal cord compression, or hypercalcemia of malignancy) but also for the durability of surgical fixation constructs. Until recently, we have largely extrapolated from what we know about other primary carcinomas regarding the implications for clinical response. Among patients with metastatic breast carcinoma to bone, those with less burden of visceral disease, and therefore, longer estimated survival, have a higher rate of skeletal-related events than those with a greater burden of visceral disease [10].

Although numerous methods exist for estimating a patient’s anticipated duration of survival, each has inherent limitations. Advances in prognostication, with novel techniques such as Bayesian belief network analysis, let us incorporate the patient’s primary diagnosis into our predictions for their one-year survival [4]. Similarly, in patients with vertebral metastases, the modified Tokuhashi score accounts for the degree of osseous and visceral involvement, as well as the biology of the underlying malignancy [1]. Although the PathFX prognostication tool [9] incorporates the primary diagnosis, it does not currently account for the patient’s response to systemic agents, nor does it differentiate between completed versus impending pathologic fractures.

The timely and insightful report by Gutowski and colleagues [5] shows that the longer survival of patients with metastatic renal or lung carcinoma who respond to biologic agents also applies to the subset of these patients treated operatively for femoral metastases. Although some clinicians have suggested on the basis of anecdotal information that bone and visceral metastases may respond differently to systemic treatment with biologic agents, the low numbers of events in the current study prevented the authors from determining whether this is true. In patients whose visceral metastases responded to biologic agents, 7% had local progression in bone, compared with 3.5% of those who did not respond to biologic agents [5]. These findings may reflect differences in the efficacy of local treatments such as radiation, or they may be attributable to differences in length of survival between cohorts.

Where Do We Need To Go?

The implications of the findings in the current study [5] depend somewhat on the nature of this potentially different response to biologic agents between bone and visceral metastases. If further investigation establishes that treatment with some biologic agents yields responses that are similar between bone and visceral sites, then studies will be needed to determine how visceral response should be included in prognostication algorithms for survival in patients with osseous metastases.

In determining the response to treatment, one must consider the means of assessing visceral and osseous burden of disease. Many treatment centers and research protocols rely on the Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 to evaluate the oncologic response to systemic therapy for malignancy [3]. When using RECIST 1.1, bone scans, positron emission tomography, and conventional radiographs are considered inadequate modalities for assessing measurable disease in bone. They can be used only to assess for the development of new lesions or the resolution of previously noted sites of involvement. Further, with RECIST 1.1 criteria, the clinical response on CT or MRI can be used only for lytic or mixed lytic/blastic metastases with a measurable soft-tissue component. Despite the abundance of data (from studies using the RECIST 1.1 criteria [2, 8]) that is gathered to support regulatory approval of these biologic agents, investigators may have missed the opportunity to evaluate whether this difference in clinical response between bone and soft tissue is evident.

It is possible that the suggested difference in response to treatment between bone and soft tissue is less a result of the biologic agent used than of the efficacy of local treatment. The current analysis includes patients with megaprosthetic reconstructions, as well as those treated with closed intramedullary nailing. The degree of tumor excision, which would likely be greater in megaprosthetic reconstruction, may have major implications in the potential for local disease progression. Lin and colleagues [7] suggested that patients with oligometastatic renal cell carcinoma treated with tumor excision have a favorable prognosis. Similarly, the degree of radiosensitivity of the tumor may greatly influence local progression, regardless of the duration of patient survival. Most patients are indicated for therapeutic radiation to the site of surgically treated long-bone metastasis. This highlights the importance of a better understanding of local treatment options (excision or resection of the tumor; optimal delivery and dosage of radiation therapy; use of antiresorptive therapy; or other local adjuvant treatments) and how to improve their efficacy in patients with osseous metastases.

If bone and visceral metastases do respond differently to treatment with biologic agents, then understanding how this difference applies to each of the various biologic agents will be the next challenge. The current study [5] was performed at a high-volume cancer center, where surgeons likely consider anticipated response to systemic and local therapies before recommending a treatment plan. Although the low numbers of events may have resulted in some clinically important differences going undetected, any difference in treatment response between bone and visceral disease may also be underestimated because of skilled, anticipatory surgical decision-making.

How Do We Get There?

Despite having a relatively large number of patients for a single-institution study, the current analysis highlights the need for collaboration among treatment centers to evaluate clinical responses to biologic agents with larger samples on which to base robust conclusions. When using the RECIST 1.1 criteria, the evaluability of osseous response to systemic therapy is limited. Therefore, any analysis comparing visceral response versus bone response would likely require use of an additional system, such as the criteria of the International Union Against Cancer, the World Health Organization, or those of the MD Anderson Cancer Center [6]. A detailed comparison of responses in bone and viscera will likely require international collection and sharing of data on patient and implant survival to maximize the value in improving patient care. The PathFX tool has advanced our ability to predict survival in these patients, though developing a Bayesian prediction algorithm for prediction of implant survival can help us anticipate which treatment will provide durable pain control and maintenance of function for the patient’s lifetime.