Dr. Ramsay - Independent Medical Opinion

Condition	Hemangiopericytoma
Date of Production	May 21, 2014
Doctor's Name	David A. Ramsay, MB ChB, DPhil (Oxon), FRCPC, FRCPath, MRCP (UK)

Thank you for your letter dated 13 February 2014 in which you have requested my opinion about this case. You have posed specific questions, which I have answered below.

This report has the following sections: Summary of my qualifications, list of materials received, clinical summary, responses to the questions outlined in your letter, and conclusions.

SUMMARY OF MY QUALIFICATIONS

I am a licensed Neuropathologist in the Province of Ontario. I have practised in London since 1990. I am a staff Neuropathologist at the London Health Sciences Centre and a Professor of Pathology at the University of Western Ontario. I am a Fellow of the Royal College of Physicians of Canada, a Fellow of the Royal College of Pathologists and a Member of the Royal College of Physicians of the United Kingdom. I was the Secretary-Treasurer of the Canadian Association of Neuropathologists from 1995 to 2000. I am a member of the Neurosurgery Examination Committee of the Royal College of Physicians and Surgeons of Canada (1994 - 2000 and 2004 - present).

My practice is in 'general' Neuropathology, which includes expertise in the diagnosis of brain tumours. I also have previous training and experience, but not special expertise, in Internal Medicine, Clinical Neurology and basic neuroscience. I teach Neuropathology to Medical Students and Residents in Neuropathology, Neurology and Neurosurgery and I have done so since I arrived in London.

The principal role of a clinical Neuropathologist is to diagnose and/or interpret diseases of the nervous system and their complications. The process of diagnosis is by examining the macroscopic ('naked eye') and microscopic appearances of tissues obtained during a surgical operation ('biopsies') or an autopsy. The interpretation of various neuropathological diagnoses entails the process of 'clinicopathological correlation', which involves the detailed analysis of clinical histories, other laboratory data, the findings from neuroimaging studies, and information available in the medical literature.

A copy of my curriculum vitae accompanies this letter; pages 10 and 11 provide a brief summary of my career.

Top of Page

LIST OF MATERIALS RECEIVED

You have provided me with the following materials:

• Medical file
• Left Occipital Meningioma SOC (Decision Number 100000397692)
• Metastatic Hemangiopericytoma SOC (Decision # 100001855714)
• Additional Documentation (re Decision # 100001855714)
• Exhibits G1-G7 re VRAB Hearing 5 June 2013
• Service Health Records

Please note that my opinion regarding the 'standards of care' provided by a pathologist should ideally include my examination of the various microscope slides concerning the case. I understand you do not have access to this material. My opinion is therefore based, among other things, on reading reports about the pathological findings rather than examining the original material. It is possible that my conclusions may change if I eventually do have the opportunity of examining the biopsy material.

Top of Page

CLINICAL SUMMARY

The Appellant, then a member of the Canadian Armed Forces, presented on 7 January 1999 at the age of 32 years having had a 'funny turn' while he was looking upward, repairing a fire alarm. The episode was characterized by loss of vision and hearing, and the symptoms resolved within a minute or two; there was no loss of consciousness or seizure activity.

His other medical history includes two documented episodes of headache, one in January 1991 that lasted approximately 1 week and the other, classed as migraine, on 31 July 1993.

In addition, in a statement dated 15 January 2002 the Appellant reported that while serving on a ship between April 1998 and August 1998 he had attended the sick-bay on several occasions, complaining of "constant sinus pain." The pain was relieved somewhat by nasal decongestants. He subsequently became "accustomed to living with the pain until the evening of 6 January 1999", when he experienced the 'funny turn' described above.

Throughout his naval career he had also been exposed to various potentially carcinogenic agents, including ionizing radiation (Drs Blood and Reaume have already addressed the possibility of a causal association between this occupational exposure and the tumor).

His examination and investigation in January 1999 revealed a focal visual defect and bilateral papilledema ¹. The report of the CT head scan on 13 January 1999 describes a "large enhancing necrotic mass in the left parietal-occipital area...most likely a glioblastoma." ubsequently, the report of an MRI scan carried out 2 days later, on 15 Jan 1999, notes "an intensely but inhomogeneous/y enhancing ² lobulated mass [in the] left occipita/lobe, extending upwards into the parietal lobe ... findings most consistent with aggressive glioma/glioblastoma multiforme... ³

A gross total surgical removal of the tumor was carried out on 19 January 1999. However, the neurosurgeon, Dr. Cameron, wrote in the operative report that:

It is not clear whether he meant local recurrence or metastasis, or both.

The surgical pathology report about the tumour (Department of Laboratory Medicine [DLM], Victoria General Hospital [VGH], Accession Number 99 01328), written by Dr. Shojania and signed on 22 January 1999, describes the microscopic features of the neoplasm as follows

These histological features appear to have been difficult to interpret because the Dr. Shojania consulted two other pathologists in the DLM at VGH, Drs Kelly and Cavers. The consensus diagnosis was an "atypical highly mitotic meningioma". This pathology laboratory does not have a neuropathologist but one or more of the pathologists may have developed a special expertise in this area.

The Appellant was treated with cranial radiation.

Some years later, in August 2010, he developed abdominal pain. Subsequent radiological investigation revealed multiple intrathoracic and intra-abdominal metastases. A core biopsy of one of the renal metastasis had the histological features of a hemangiopericytoma/solitary fibrous tumor ⁴. The slides from the Victoria General Hospital were re-examined and the brain tumor diagnosis revised to that of a primary meningeal hemangiopericytoma (noted in consultation reports from clinic visits on 19 November 2010 and 7 January 2011).

Despite treatment with chemotherapy, the Appellant died on 12 January 2013, 14 years after his brain tumor was diagnosed.

Top of Page

RESPONSES TO YOUR QUESTIONS

1. General Information on the Diagnosis of Brain Tumors
   • Could you describe the procedures, investigations and analyses that would be performed by pathologist when exercising a reasonable and prudent degree of skill, care and competence in the diagnosis of a brain tumor at the applicable period of time (in 1999)?

     Depending on the experience of pathologist and the commonness and complexity of the tumor in question, the reliable diagnosis of a brain tumor requires knowledge of the neuroimaging (e.g. , MRI, CT) appearance of the tumor, familiarity with the varied histological appearances of different types of brain tumors, the selective use of ancillary techniques to define the nature of the tumor (usually special staining and immunohistochemical methods, the latter described below), and, in complex or rare tumors in which there may be doubt about the diagnosis, a preparedness to refer the case to other pathologists for an expert opinion (usually neuropathologists in the case of brain tumors).

   • In particular, what investigations, procedures and analysis would be performed by reasonably prudent and diligent pathologist?

     In all cases sections from the tumor are examined after having been stained with a 'routine' stain (such as hematoxylin and eosin). In many cases, the tumour samples are also treated with different types of stain (which highlight different tissues) and one or more immunohistochemical methods.

     Immunohistochemistry is a technique that allows specific molecules (usually proteins) called 'antigens' in the tumor to be 'labelled' or 'stained'. Different types of tumors have different antigens; the presence or otherwise of various antigen types usually allow the tumor class to be determined.

     The immunohistochemical technique relies on: (a) the availability of specific antibodies that have been 'raised' in animals (the antigen in question is injected into the animal, the animal produces specific antibodies against the antigen, and then the antibodies are removed the animal's circulation for use in immunohistochemistry); (b) the application of a solution of these antibodies to a microscopic section; (c) the binding of the antibodies to specific proteins in the section; and (d) the attachment of a colour tag to the antibody so the 'labelled' binding sites are visualised when the glass slides are examined under a microscope. The terms 'immunopositive' and 'immunonegative' are used refer to the presence or absence respectively of a given antigen or protein.

   • Types of tumors and terminology
     • How does a sarcoma different from a meningioma? Are all sarcomas malignant?

       In very general terms, tumors are classified on the basis of their presumptive cell or tissue of origin; this classification is largely based on the location of the primary tumour and similarities in appearance between the tumour cells and the cells of the normal tissues at the site of the neoplasm. Thus, for example, melanomas arise from the pigmented cells of the skin (i.e., melanocytes), gliomas arise from the glial cells of the central nervous system (glial cells are the 'supporting' cells of the brain that maintain the function of the nerve cells) and meningiomas develop in the meninges, which are the membranes that line the surface of the brain.

       A 'sarcoma' is a malignant tumor that arises from the cells of one of the various 'supporting' structures of the body, including bone, muscle, fatty tissue, blood vessels and miscellaneous soft or 'fibrous' tissue; the corresponding tumours are osteosarcomas, myosarcomas, liposarcomas, angiosarcomas and fibrosarcomas (benign tumors arising from these tissues are osteoma, myomas, lipoma, angiomas and fibromas respectively). When the presumptive tissue of origin of the sarcoma cannot be determined the tumor is referred to generically as a 'sarcoma'.

       Hemangiopericytomas are a type of sarcoma whose cellular appearances suggest it is derived from pericytes, which are cells that form part of the wall of blood vessels. They can occur throughout the body. (The hemangiopericytoma is behaviourally and histologically distinct from angiosarcomas.)

       Sarcomas, which are malignant neoplasms, and meningiomas, which are usually benign tumors, are compared in terms of their behavior and their histological appearances. Regarding their behavior, sarcomas invade adjacent tissues, grow rapidly, recur at the primary site (even after resection), and may spread through blood vessels and the lymphatic system to other distant sites (i.e., they metastasise). In contrast, meningiomas compress adjacent tissues, grow slowly, recur much less frequently at the primary site than do sarcomas, and rarely metastasise. The invasive properties of sarcomas also make them difficult to remove completely whereas the 'contained' nature of a meningioma allows its complete removal.

       In terms of microscopic appearances, sarcomas are neoplasms composed of closely packed, usually elongated (i.e., 'spindled'), monomorphic (i.e., of similar in appearance), mitotically active cells. Meningiomas are less cellular, the morphology of their cells and their pattern of growth (i.e., architecture) are more varied and mitoses are either absent or sparse. The architecture of meningiomas is distinctive and includes various combinations of generally loosely packed spindle cells ('fibrous' pattern) and round cells that in sections form pavement-like patterns ('meningothelial' pattern); meningioma cells characteristically wrap themselves around one another to form distinctive 'whorls.'

       Referring to their immunohistochemical properties, both sarcomas and meningiomas are immunopositive for vimentin whereas immunopositivity for epithelial membrane antigen (EMA) is common in meningiomas and usually not found in sarcomas.

       The distinction using histological criteria between a sarcoma and a meningioma is therefore usually straight-forward; however, meningiomas are notorious for having a highly varied histological appearance and they may be 'aggressive' ('atypical' meningiomas) or malignant ('malignant meningiomas'); however atypical and malignant meningiomas ordinarily retain recognisable histological features of regular meningiomas.

     • What criteria or factors would a pathologist be expected to consider when diagnosing or classifying a brain tumor?

       These criteria and factors are described in the first section, "General Information on the Diagnosis of Brain Tumors."

     • What characteristics would indicate that the tumor is a malignant tumor or a benign tumor?

       Malignant tumors are distinguished from benign tumors by their capacity to metastasise and their potential for rapid and uncontrolled growth at their site of origin, both at presentation and as recurrences. Malignant tumors outside the brain commonly metastasize but metastases arising from brain tumors within the cranium are rare (however sinus invasion by a brain tumour may increase the risk of metastasis). The histological features of malignancy include one or more of the following: closely packed cells (i.e., 'cellularity'), marked variation in the shape and size of the cell nuclei, evidence of vigorous cell division (i.e., numerous mitoses) and foci of dead (i.e., 'necrotic') tissue (namely areas where the energy demands have exceeded the supply of essential nutrients).

       There are degrees of malignancy. With reference to brain tumors, the World Health Organization (WHO) grading scheme assigns grade I through IV to all 'brain' tumors (including tumours arising in the meninges), the higher the grade the more aggressive the tumor. Brain tumours that are WHO Grade I and II are histologically benign. (However these 'benign' tumours may eventually be fatal if they are located where they cannot safely be excised; for this reason, adjective 'well-differentiated' is often applied to WHO Grade I and II brain tumors.) Brain tumours that are WHO Grade Ill and IV brain tumors are histologically malignant and behave accordingly.

2. Diagnosis of and Distinction between a Meningioma and a Hemangiopericytoma?
   • What is the significance of the term "atypical" when used in reference to a meningioma?

     The term "atypical" indicates that something about the histological or cytological appearance of a tumor does not fit with the presumptive diagnosis or raises the possibility of malignant behaviour in an otherwise benign-appearing neoplasm. Nuclear atypia, for example, refers to nuclei that show more than the normal variation in their size and shape, which is a feature of malignancy.

     An "atypical" meningioma is a meningioma that has several features that collectively may indicate a more aggressive course than for a 'usual' meningioma. These features include a mitotic rate of greater than 4 per 10 high-power fields and at least 3 of the following features: 'Sheeting' (which is the loss of the characteristic and varied microscopic architecture of the meningioma), increased cellularity compared to a normal meningioma, large nucleoli (these structures are in the cell nuclei and play a role in the molecular and genetic control of the cell), and groups of undifferentiated small cells (i.e., small cells without the 'fibrous' or 'meningothelial' appearance of normal meningioma cells). An 'atypical' meningioma is WHO Grade 2 and has a greater risk of local recurrence than a WHO Grade I meningioma. Although the complete resection of the meningioma is often curative, some histologically benign meningiomas may behave in a malignant fashion by recurring frequently and many atypical meningiomas can have a benign course.

   • Could you explain the distinction between an atypical meningioma and a hemangiopericytoma?

     Atypical meningiomas are WHO Grade II. Hemangiopericytomas are graded either WHO Grade II or Grade Ill. The presence of 5 or more mitoses per 10 high power fields and high cellularity (amongst other features), which the Appellant's tumour had, indicate a WHO Grade Ill hemangiopericytoma.

     Atypical meningiomas and hemangiopericytomas have very distinctive histological appearances. They are described as follows in a textbook of surgical pathology, published in 1999 and widely read: ⁵

     "Atypical meningiomas show more cellularity and mitotic activity than WHO Grade I meningiomas. They may have large nucleoli, patternless sheet-like growth, and foci of necrosis. ... [Hemangiopericytomas are] highly cellular and mitotically active... rich in pericellular reticulin [and have] the same features as hemangiopericytoma elsewhere in the body. The… hemangiopericytomas are distinguished from benign meningiomas by their hypercellularity, higher mitotic index... while there are exceptions, they tend to lack markers other than mesenchymal markers."

     By markers, the authors are referring to immunohistochemical markers. Meningiomas are immunopositive for vimentin and epithelial membrane antigen (EMA) whereas hemangiopericytomas are immunopositive for vimentin and immunonegative for EMA. (This information and these markers were readily available in 1999. ⁶) Furthermore, unlike atypical meningiomas, the individual cells of hemangiopericytomas are typically surrounded by reticulin fibres, an easily stainable fibrillary constituent of the intercellular matrix (i.e., the space between individual cells) whereas meningioma cells are not.

     Relatively recent studies suggest that the 5 year and 10 year survival rates for atypical meningiomas are somewhat worse than for meningeal hemangiopericytomas (78% and 53% for atypical meningiomas and 92% and 68% respectively for hemangiopericytomas), that the recurrence rates for both tumours are similar (approximately 50%), approximately 20% of cases of hemangiopericytomas eventually generate metastasis, and metastases are rare for atypical meningiomas. ⁷

   • What does the term "highly mitotic" mean, and what role would the number of mitotic figures play in the classification of the tumor? Would the number [of] mitotic cells be considered indicative of a hemangiopericytoma? What other characteristics would be considered indicative of a hemangiopericytoma?

     A highly mitotically active tumor is more likely to be malignant. Mitoses tend to be more numerous in hemangiopericytomas than in atypical meningiomas. The other histological characteristics of a hemangiopericytoma are described in the previous section (i.e., dense cellularity, numerous mitoses, reticulin fibres around individual cells, immunopositivity for vimentin, immunonegativity for EMA).

3. Standard of care
   • In your opinion, would the failure to accurately diagnose a brain tumor as a hemangiopericytoma fall below the prevailing standard of care and professional practice for the diagnosis of a brain tumor in 1999?

     The MRI and CT features of the Appellant's tumor indicated a primary tumor of the brain substance but the findings during the surgical operation showed that the tumor originated from the meninges. Further, the histological appearance of the tumour, being predominantly a densely cellular, highly mitotically active neoplasm, did not have a readily recognisable pattern for the pathologists involved. Hence the neuroimaging, intraoperative and histological observations clearly indicated an unusual or unconventional intracranial tumor. Under these circumstances the pathologist should form a differential diagnosis and carry out additional testing to attempt to resolve the diagnosis. If the diagnosis cannot be resolved, or if doubt remains about it, the case should be referred to an appropriate specialist pathologist, in this case a neuropathologist.

     There is no evidence in the surgical pathology reports that the possibility of a hemangiopericytoma was considered by the pathologists, even though, as previously mentioned, meningeal examples of this tumor are described in the standard surgical pathology textbooks of the time. For this reason I suspect appropriate additional stains and immunohistochemical testing were not carried out, even though such testing was available then and the unusual features of the tumour warranted additional investigation. If the presence of pericellular reticulin and the lack of EMA immunolabeling had been found, a knowledgeable pathologist would have been alerted to an alternative diagnoses to a meningioma.

     It is clear from the surgical pathology report that the histological appearances presented difficulties of interpretation since the Dr. Shojania consulted with his colleagues. Given the peculiarities of the case already outlined it would have been appropriate to refer it to a neuropathologist, whether in Vancouver or elsewhere.

   • In your opinion based on the available evidence in the VRAB file, did the pathologist and other members of the medical team who were responsible for the diagnosis of the tumor in this case, exercise a reasonable level of skill and prudent medical judgment in arriving at the diagnosis of meningioma in 1999?

     From the foregoing, and admittedly with the benefit of hindsight, the Dr. Shojania should have included the possibility of a hemangiopericytoma in the differential diagnosis, a reticulin stain should have been performed and immunohistochemistry for vimentin and EMA should have been carried out. If these measure did not resolve the diagnosis, and the case should then have been referred to a neuropathologist for a second opinion.

   Comment regarding the Question of "Standard of Care":

   It is difficult for me to fairly determine whether the deficiencies outlined in this section "fall below the prevailing standard of care and professional practice" or represent the lack of "a reasonable level of skill and prudent medical judgment" on the part of the various parties involved simply because I do not have sufficient information about the prevailing patterns and customs of practice in the Victoria General Hospital at the time, nor do I have enough details about the specific circumstances surrounding the analysis of the biopsy.

   Nevertheless, given the information that I do have, whatever may be the explanation for the apparent shortcomings identified above, my opinion is that the pathological management of the case did fall short of what would reasonably have been expected in 1999.

Top of Page

CONCLUSIONS

The Appellant presented in January 1999 at 32 years of age with a large, malignant, meningeal-based brain tumor, which underwent 'gross total resection'. The attending pathologists had difficulty interpreting the histological appearance of the tumor but they finally concluded that it was an atypical meningioma. Atypical meningiomas have a significant risk of local recurrence (i.e. recurrence at the site of resection), but, in general, rarely spread outside the brain. In August 2010 the Appellant presented with symptoms of widespread intra-thoracic and intra-abdominal metastatic hemangiopericytoma. The histopathology of the brain tumor was reviewed at this time and the diagnosis revised to a meningeal hemangiopericytoma. Meningeal hemangiopericytomas carry a significant risk of metastasis. The Appellant was treated with chemotherapy but he died in January 2013, fourteen years after he first presented with his tumor.

The possibility of a hemangiopericytoma was not considered when the histology of the brain tumor was first examined and, accordingly, the appropriate stains and immunohistochemical methods that would have allowed the distinction to be made between an atypical meningioma and a meningeal hemangiopericytoma were not carried out. Despite the unusual appearance of the tumor a neuropathologist was not consulted. Subject to caveats mentioned above, my opinion is that the pathological management of the case fell short of what would reasonably have been expected in 1999.

I do not have the expertise to comment on the significance of these conclusions in regard to the Appellant's management. I believe that the key questions for consideration by a neuro­ oncologist are the following:

• What is the possibility that the headaches that the Appellant described for several months prior to his presentation were caused by the brain tumor? Were these headaches appropriately investigated? If the headaches were likely related to the brain tumor what is the possibility that the delay in diagnosis increased the likelihood of metastasis?
• If the diagnosis of a malignant (i.e., WHO Grade Ill) hemangiopericytoma had been made in January 1999, and given the invasion by the tumour of the sagittal sinus, would the patient have been treated differently? If so, what is the possibility that the clinical course would have been better?

I hope that the foregoing observations are useful for the deliberations of the Veterans Review and Appeal Board. I would be happy to assist with any additional questions that you or the Board might have about this case.

Top of Page

 

---

 

Footnotes

• ^{1. The visual defect was a "right homonymous inferior quadrantanopsia", which indicates a disorder in the left occipital lobe. Papilledema is swelling at the site where the optic nerve leaves the back of the eye, en route for the brain- it indicates significantly raised pressure in the cranium, often the result of an intracranial tumour.}
• ^{2. Contrast material is injected into a vein and concentrates in tumours with a substantial blood supply. The accumulation of the contrast material is visualised in CT and MRI scans as 'enhancement', which often indicates a malignant tumour. However, ordinary meningiomas also 'enhance', usually homogeneously so.}
• ^{3. Aggressive gliomas and glioblastomas (multiforme) are highly malignant tumours that grow in the brain (as opposed to the meninges).}
• ^{4. The phrase 'hemangiopericytoma/solitary fibrous tumor' reflects a recently evolving concept,since 1999, that solitary fibrous tumours and hemangiopericytomas form a spectrum of a single entity, the former being benign and the latter malignant.}
• ^{5. McKeever,Paul. The brain, spinal cord,and meninges. In: Sternberg, StephenS, Antonioli,Donald A, Carter, Daryl, Mills, Stacey E., Oberman, Harold A., editors. Diagnostic Surgical Pathology. 3rd edition. Georgetown,Tex. :New York, N.Y: Lippincott Williams & Wilkins; 1999. p. 389-493.}
• ^{6. (a) Winek RR,Scheithauer BW, Wick MR. Meningioma,meningeal hemangiopericytoma (angioblastic meningioma), peripheral hemangiopericytoma, and acoustic schwannoma. A comparative immunohistochemical study. Am J Surg Pathol. 1989 Apr;13(4):251-61. (b) Perry A,Scheithauer BW,Nascimento AG. The immunophenotypic spectrum of meningeal hemangiopericytoma: a comparison with fibrous meningioma and solitary fibrous tumor of meninges. Am J Surg Pathol. 1997 Nov;21(11):1354-60.}
• ^{7. (a) Rutkowski MJ, Jian BJ, Bloch 0, Chen C, Sughrue ME, Tihan T, et al. Intracranial hemangiopericytoma: clinical experience and treatment considerations in a modern series of 40 adult patients. Cancer. 2012 Mar 15; 118(6}:1628-36. (b) Durand A, Labrousse F, Jouvet A, Bauchet L, Kalama rides M, Menei P,et al. WHO Grade II and Ill meningiomas: a study of prognostic factors. J Neurooncol. 2009 Dec;95(3}:367-75.}